Search results for: supervised classifiers

Authors: Despoina Mouratidis, Katia Lida Kermanidis

Abstract:

Machine translation (MT) is a very important task in Natural Language Processing (NLP). MT evaluation is crucial in MT development, as it constitutes the means to assess the success of an MT system, and also helps improve its performance. Several methods have been proposed for the evaluation of (MT) systems. Some of the most popular ones in automatic MT evaluation are score-based, such as the BLEU score, and others are based on lexical similarity or syntactic similarity between the MT outputs and the reference involving higher-level information like part of speech tagging (POS). This paper presents a language-independent machine learning framework for classifying pairwise translations. This framework uses vector representations of two machine-produced translations, one from a statistical machine translation model (SMT) and one from a neural machine translation model (NMT). The vector representations consist of automatically extracted word embeddings and string-like language-independent features. These vector representations used as an input to a multi-layer neural network (NN) that models the similarity between each MT output and the reference, as well as between the two MT outputs. To evaluate the proposed approach, a professional translation and a "ground-truth" annotation are used. The parallel corpora used are English-Greek (EN-GR) and English-Italian (EN-IT), in the educational domain and of informal genres (video lecture subtitles, course forum text, etc.) that are difficult to be reliably translated. They have tested three basic deep learning (DL) architectures to this schema: (i) fully-connected dense, (ii) Convolutional Neural Network (CNN), and (iii) Long Short-Term Memory (LSTM). Experiments show that all tested architectures achieved better results when compared against those of some of the well-known basic approaches, such as Random Forest (RF) and Support Vector Machine (SVM). Better accuracy results are obtained when LSTM layers are used in our schema. In terms of a balance between the results, better accuracy results are obtained when dense layers are used. The reason for this is that the model correctly classifies more sentences of the minority class (SMT). For a more integrated analysis of the accuracy results, a qualitative linguistic analysis is carried out. In this context, problems have been identified about some figures of speech, as the metaphors, or about certain linguistic phenomena, such as per etymology: paronyms. It is quite interesting to find out why all the classifiers led to worse accuracy results in Italian as compared to Greek, taking into account that the linguistic features employed are language independent.

Keywords: machine learning, machine translation evaluation, neural network architecture, pairwise classification

Procedia PDF Downloads 103

Authors: Nassima Noufail, Sara Bouhali

Abstract:

In this work, we develop a semi-supervised solution for the purpose of action detection in videos and propose an efficient algorithm for video segmentation. The approach is divided into video segmentation, feature extraction, and classification. In the first part, a video is segmented into clips, and we used the K-means algorithm for this segmentation; our goal is to find groups based on similarity in the video. The application of k-means clustering into all the frames is time-consuming; therefore, we started by the identification of transition frames where the scene in the video changes significantly, and then we applied K-means clustering into these transition frames. We used two image filters, the gaussian filter and the Laplacian of Gaussian. Each filter extracts a set of features from the frames. The Gaussian filter blurs the image and omits the higher frequencies, and the Laplacian of gaussian detects regions of rapid intensity changes; we then used this vector of filter responses as an input to our k-means algorithm. The output is a set of cluster centers. Each video frame pixel is then mapped to the nearest cluster center and painted with a corresponding color to form a visual map. The resulting visual map had similar pixels grouped. We then computed a cluster score indicating how clusters are near each other and plotted a signal representing frame number vs. clustering score. Our hypothesis was that the evolution of the signal would not change if semantically related events were happening in the scene. We marked the breakpoints at which the root mean square level of the signal changes significantly, and each breakpoint is an indication of the beginning of a new video segment. In the second part, for each segment from part 1, we randomly selected a 16-frame clip, then we extracted spatiotemporal features using convolutional 3D network C3D for every 16 frames using a pre-trained model. The C3D final output is a 512-feature vector dimension; hence we used principal component analysis (PCA) for dimensionality reduction. The final part is the classification. The C3D feature vectors are used as input to a multi-class linear support vector machine (SVM) for the training model, and we used a multi-classifier to detect the action. We evaluated our experiment on the UCF101 dataset, which consists of 101 human action categories, and we achieved an accuracy that outperforms the state of art by 1.2%.

Keywords: video segmentation, action detection, classification, Kmeans, C3D

Procedia PDF Downloads 49

Authors: Ju Peng Poh, Jun Yu Charles Lee, Jonathan Chew Hoe Khoo

Abstract:

The emergence of digital twin technology, a digital replica of physical world, has improved the real-time access to data from sensors about the performance of buildings. This digital transformation has opened up many opportunities to improve the management of the building by using the data collected to help monitor consumption patterns and energy leakages. One example is the integration of predictive models for anomaly detection. In this paper, we use the GAM (Generalised Additive Model) for the anomaly detection of Air Handling Units (AHU) power consumption pattern. There is ample research work on the use of GAM for the prediction of power consumption at the office building and nation-wide level. However, there is limited illustration of its anomaly detection capabilities, prescriptive analytics case study, and its integration with the latest development of digital twin technology. In this paper, we applied the general GAM modelling framework on the historical data of the AHU power consumption and cooling load of the building between Jan 2018 to Aug 2019 from an education campus in Singapore to train prediction models that, in turn, yield predicted values and ranges. The historical data are seamlessly extracted from the digital twin for modelling purposes. We enhanced the utility of the GAM model by using it to power a real-time anomaly detection system based on the forward predicted ranges. The magnitude of deviation from the upper and lower bounds of the uncertainty intervals is used to inform and identify anomalous data points, all based on historical data, without explicit intervention from domain experts. Notwithstanding, the domain expert fits in through an optional feedback loop through which iterative data cleansing is performed. After an anomalously high or low level of power consumption detected, a set of rule-based conditions are evaluated in real-time to help determine the next course of action for the facilities manager. The performance of GAM is then compared with other approaches to evaluate its effectiveness. Lastly, we discuss the successfully deployment of this approach for the detection of anomalous power consumption pattern and illustrated with real-world use cases.

Keywords: anomaly detection, digital twin, generalised additive model, GAM, power consumption, supervised learning

Procedia PDF Downloads 125

Authors: Camila Lemos Pinto, Juliana Alves Carneiro, Patrícia Borges Botelho, João Felipe Mota

Abstract:

Sarcopenia, a syndrome characterized by progressive and generalized loss of skeletal muscle mass and strength, currently affects over 50 million people and increases the risk of adverse outcomes such as physical disability, poor quality of life and death. The aim of this study was to examine the efficacy of creatine supplementation associated with resistance training on muscle mass in the elderly. A 12-week, double blind, randomized, parallel group, placebo controlled trial was conducted. Participants were randomly allocated into one of the following groups: placebo with resistance training (PL+RT, n=14) and creatine supplementation with resistance training (CR + RT, n=13). The subjects from CR+RT group received 5 g/day of creatine monohydrate and the subjects from the PL+RT group were given the same dose of maltodextrin. Participants were instructed to ingest the supplement on non-training days immediately after lunch and on training days immediately after resistance training sessions dissolved in a beverage comprising 100 g of maltodextrin lemon flavored. Participants of both groups undertook a supervised exercise training program for 12 weeks (3 times per week). The subjects were assessed at baseline and after 12 weeks. The primary outcome was muscle mass, assessed by dual energy X-ray absorptiometry (DXA). The secondary outcome included diagnose participants with one of the three stages of sarcopenia (presarcopenia, sarcopenia and severe sarcopenia) by skeletal muscle mass index (SMI), handgrip strength and gait speed. CR+RT group had a significant increase in SMI and muscle (p<0.0001), a significant decrease in android and gynoid fat (p = 0.028 and p=0.035, respectively) and a tendency of decreasing in body fat (p=0.053) after the intervention. PL+RT only had a significant increase in SMI (p=0.007). The main finding of this clinical trial indicated that creatine supplementation combined with resistance training was capable of increasing muscle mass in our elderly cohort (p=0.02). In addition, the number of subjects diagnosed with one of the three stages of sarcopenia at baseline decreased in the creatine supplemented group in comparison with the placebo group (CR+RT, n=-3; PL+RT, n=0). In summary, 12 weeks of creatine supplementation associated with resistance training resulted in increases in muscle mass. This is the first research with elderly of both sexes that show the same increase in muscle mass with a minor quantity of creatine supplementation in a short period. Future long-term research should investigate the effects of these interventions in sarcopenic elderly.

Keywords: creatine, dietetic supplement, elderly, resistance training

Procedia PDF Downloads 447

Authors: Aayushi Pandey, Manoj Kumar Pandey, Ashutosh Tiwari, Kireet Kumar

Abstract:

Glaciers play an important role in climate change and are sensitive phenomena of global climate change scenario. Glaciers in Himalayas are unique as they are predominantly valley type and are located in tropical, high altitude regions. These glaciers are often covered with debris which greatly affects ablation rate of glaciers and work as a sensitive indicator of glacier health. The aim of this study is to map high altitude Glacier surface with a focus on glacial lake and debris estimation using different techniques in Nagling glacier of dhauliganga basin in Himalayan region. Different Image Classification techniques i.e. thresholding on different band ratios and supervised classification using maximum likelihood classifier (MLC) have been used on high resolution sentinel 2A level 1c satellite imagery of 14 October 2017.Here Near Infrared (NIR)/Shortwave Infrared (SWIR) ratio image was used to extract the glaciated classes (Snow, Ice, Ice Mixed Debris) from other non-glaciated terrain classes. SWIR/BLUE Ratio Image was used to map valley rock and Debris while Green/NIR ratio image was found most suitable for mapping Glacial Lake. Accuracy assessment was performed using high resolution (3 meters) Planetscope Imagery using 60 stratified random points. The overall accuracy of MLC was 85 % while the accuracy of Band Ratios was 96.66 %. According to Band Ratio technique total areal extent of glaciated classes (Snow, Ice ,IMD) in Nagling glacier was 10.70 km2 nearly 38.07% of study area comprising of 30.87 % Snow covered area, 3.93% Ice and 3.27 % IMD covered area. Non-glaciated classes (vegetation, glacial lake, debris and valley rock) covered 61.93 % of the total area out of which valley rock is dominant with 33.83% coverage followed by debris covering 27.7 % of the area in nagling glacier. Glacial lake and Debris were accurately mapped using Band ratio technique Hence, Band Ratio approach appears to be useful for the mapping of debris covered glacier in Himalayan Region.

Keywords: band ratio, Dhauliganga basin, glacier mapping, Himalayan region, maximum likelihood classifier (MLC), Sentinel-2 satellite image

Procedia PDF Downloads 201

Authors: Ella Sèdé Maforikan

Abstract:

Accurate land cover mapping is essential for effective environmental monitoring and natural resources management. This study focuses on assessing the classification performance of two satellite datasets and evaluating the impact of different input feature combinations on classification accuracy in the Beterou catchment, situated in the northern part of Benin. Landsat-8 and Sentinel-2 images from June 1, 2020, to March 31, 2021, were utilized. Employing the Random Forest (RF) algorithm on Google Earth Engine (GEE), a supervised classification categorized the land into five classes: forest, savannas, cropland, settlement, and water bodies. GEE was chosen due to its high-performance computing capabilities, mitigating computational burdens associated with traditional land cover classification methods. By eliminating the need for individual satellite image downloads and providing access to an extensive archive of remote sensing data, GEE facilitated efficient model training on remote sensing data. The study achieved commendable overall accuracy (OA), ranging from 84% to 85%, even without incorporating spectral indices and terrain metrics into the model. Notably, the inclusion of additional input sources, specifically terrain features like slope and elevation, enhanced classification accuracy. The highest accuracy was achieved with Sentinel-2 (OA = 91%, Kappa = 0.88), slightly surpassing Landsat-8 (OA = 90%, Kappa = 0.87). This underscores the significance of combining diverse input sources for optimal accuracy in land cover mapping. The methodology presented herein not only enables the creation of precise, expeditious land cover maps but also demonstrates the prowess of cloud computing through GEE for large-scale land cover mapping with remarkable accuracy. The study emphasizes the synergy of different input sources to achieve superior accuracy. As a future recommendation, the application of Light Detection and Ranging (LiDAR) technology is proposed to enhance vegetation type differentiation in the Beterou catchment. Additionally, a cross-comparison between Sentinel-2 and Landsat-8 for assessing long-term land cover changes is suggested.

Keywords: land cover mapping, Google Earth Engine, random forest, Beterou catchment

Procedia PDF Downloads 30

Authors: Tausif A. Ishtiaque, Zarrin T. Tasin, Kazi S. Akter

Abstract:

Current trend of urban expansion, especially in the developing countries has caused significant changes in land cover, which is generating great concern due to its widespread environmental degradation. Energy consumption of the cities is also increasing with the aggravated heat island effect. Distribution of land surface temperature (LST) is one of the most significant climatic parameters affected by urban land cover change. Recent increasing trend of LST is causing elevated temperature profile of the built up area with less vegetative cover. Gradual change in land cover, especially decrease in vegetative cover is enhancing the Urban Heat Island (UHI) effect in the developing cities around the world. Increase in the amount of urban vegetation cover can be a useful solution for the reduction of UHI intensity. LST and Normalized Difference Vegetation Index (NDVI) have widely been accepted as reliable indicators of UHI and vegetation abundance respectively. Chittagong, the second largest city of Bangladesh, has been a growth center due to rapid urbanization over the last several decades. This study assesses the intensity of UHI in Chittagong city by analyzing the relationship between LST and NDVI based on the type of land use/land cover (LULC) in the study area applying an integrated approach of Geographic Information System (GIS), remote sensing (RS), and regression analysis. Land cover map is prepared through an interactive supervised classification using remotely sensed data from Landsat ETM+ image along with NDVI differencing using ArcGIS. LST and NDVI values are extracted from the same image. The regression analysis between LST and NDVI indicates that within the study area, UHI is directly correlated with LST while negatively correlated with NDVI. It interprets that surface temperature reduces with increase in vegetation cover along with reduction in UHI intensity. Moreover, there are noticeable differences in the relationship between LST and NDVI based on the type of LULC. In other words, depending on the type of land usage, increase in vegetation cover has a varying impact on the UHI intensity. This analysis will contribute to the formulation of sustainable urban land use planning decisions as well as suggesting suitable actions for mitigation of UHI intensity within the study area.

Keywords: land cover change, land surface temperature, normalized difference vegetation index, urban heat island

Procedia PDF Downloads 252

Authors: Newton Muhury, Armando A. Apan, Tek Maraseni

Abstract:

This study modelled the relationships between vegetation response and available water below the soil surface using the Terra’s Moderate Resolution Imaging Spectroradiometer (MODIS) generated Normalised Difference Vegetation Index (NDVI) and soil water content (SWC) data. The Soil & Water Assessment Tool (SWAT) interface known as ArcSWAT was used in ArcGIS for the groundwater analysis. The SWAT model was calibrated and validated in SWAT-CUP software using 10 years (2001-2010) of monthly streamflow data. The average Nash-Sutcliffe Efficiency during the calibration and validation was 0.54 and 0.51, respectively, indicating that the model performances were good. Twenty years (2001-2020) of monthly MODIS NDVI data for three different types of vegetation (forest, shrub, and grass) and soil water content for 43 sub-basins were analysed using the WEKA, machine learning tool with a selection of two supervised machine learning algorithms, i.e., support vector machine (SVM) and random forest (RF). The modelling results show that different types of vegetation response and soil water content vary in the dry and wet season. For example, the model generated high positive relationships (r=0.76, 0.73, and 0.81) between the measured and predicted NDVI values of all vegetation in the study area against the groundwater flow (GW), soil water content (SWC), and the combination of these two variables, respectively, during the dry season. However, these relationships were reduced by 36.8% (r=0.48) and 13.6% (r=0.63) against GW and SWC, respectively, in the wet season. On the other hand, the model predicted a moderate positive relationship (r=0.63) between shrub vegetation type and soil water content during the dry season, which was reduced by 31.7% (r=0.43) during the wet season. Our models also predicted that vegetation in the top location (upper part) of the sub-basin is highly responsive to GW and SWC (r=0.78, and 0.70) during the dry season. The results of this study indicate the study region is suitable for seasonal crop production in dry season. Moreover, the results predicted that the growth of vegetation in the top-point location is highly dependent on groundwater flow in both dry and wet seasons, and any instability or long-term drought can negatively affect these floodplain vegetation communities. This study has enriched our knowledge of vegetation responses to groundwater in each season, which will facilitate better floodplain vegetation management.

Keywords: ArcSWAT, machine learning, floodplain vegetation, MODIS NDVI, groundwater

Procedia PDF Downloads 89

Authors: Elia Favarelli, Enrico Testi, Andrea Giorgetti

Abstract:

Autonomous structural health monitoring (SHM) of many structures and bridges became a topic of paramount importance for maintenance purposes and safety reasons. This paper proposes a set of machine learning (ML) tools to perform automatic feature selection and detection of anomalies in a bridge from vibrational data and compare different feature extraction schemes to increase the accuracy and reduce the amount of data collected. As a case study, the Z-24 bridge is considered because of the extensive database of accelerometric data in both standard and damaged conditions. The proposed framework starts from the first four fundamental frequencies extracted through operational modal analysis (OMA) and clustering, followed by density-based time-domain filtering (tracking). The fundamental frequencies extracted are then fed to a dimensionality reduction block implemented through two different approaches: feature selection (intelligent multiplexer) that tries to estimate the most reliable frequencies based on the evaluation of some statistical features (i.e., mean value, variance, kurtosis), and feature extraction (auto-associative neural network (ANN)) that combine the fundamental frequencies to extract new damage sensitive features in a low dimensional feature space. Finally, one class classifier (OCC) algorithms perform anomaly detection, trained with standard condition points, and tested with normal and anomaly ones. In particular, a new anomaly detector strategy is proposed, namely one class classifier neural network two (OCCNN2), which exploit the classification capability of standard classifiers in an anomaly detection problem, finding the standard class (the boundary of the features space in normal operating conditions) through a two-step approach: coarse and fine boundary estimation. The coarse estimation uses classics OCC techniques, while the fine estimation is performed through a feedforward neural network (NN) trained that exploits the boundaries estimated in the coarse step. The detection algorithms vare then compared with known methods based on principal component analysis (PCA), kernel principal component analysis (KPCA), and auto-associative neural network (ANN). In many cases, the proposed solution increases the performance with respect to the standard OCC algorithms in terms of F1 score and accuracy. In particular, by evaluating the correct features, the anomaly can be detected with accuracy and an F1 score greater than 96% with the proposed method.

Keywords: anomaly detection, frequencies selection, modal analysis, neural network, sensor network, structural health monitoring, vibration measurement

Procedia PDF Downloads 91

Authors: Melis Inalpulat, Levent Genc

Abstract:

Land Use Land Cover (LULC) changes due to human activities and natural causes have become a major environmental concern. Assessment of temporal remote sensing data provides information about LULC impacts on environment. Land Surface Temperature (LST) is one of the important components for modeling environmental changes in climatological, hydrological, and agricultural studies. In this study, LULC changes (September 7, 1984 and July 8, 2014) especially in agricultural lands together with population changes (1985-2014) and LST status were investigated using remotely sensed and census data in South Marmara Watershed, Turkey. LULC changes were determined using Landsat TM and Landsat OLI data acquired in 1984 and 2014 summers. Six-band TM and OLI images were classified using supervised classification method to prepare LULC map including five classes including Forest (F), Grazing Land (G), Agricultural Land (A), Water Surface (W), and Residential Area-Bare Soil (R-B) classes. The LST image was also derived from thermal bands of the same dates. LULC classification results showed that forest areas, agricultural lands, water surfaces and residential area-bare soils were increased as 65751 ha, 20163 ha, 1924 ha and 20462 ha respectively. In comparison, a dramatic decrement occurred in grazing land (107985 ha) within three decades. The population increased % 29 between years 1984-2014 in whole study area. Along with the natural causes, migration also caused this increase since the study area has an important employment potential. LULC was transformed among the classes due to the expansion in residential, commercial and industrial areas as well as political decisions. In the study, results showed that agricultural lands around the settlement areas transformed to residential areas in 30 years. The LST images showed that mean temperatures were ranged between 26-32 °C in 1984 and 27-33 °C in 2014. Minimum temperature of agricultural lands was increased 3 °C and reached to 23 °C. In contrast, maximum temperature of A class decreased to 41 °C from 44 °C. Considering temperatures of the 2014 R-B class and 1984 status of same areas, it was seen that mean, min and max temperatures increased by 2 °C. As a result, the dynamism of population, LULC and LST resulted in increasing mean and maximum surface temperatures, living spaces/industrial areas and agricultural lands.

Keywords: census data, landsat, land surface temperature (LST), land use land cover (LULC)

Procedia PDF Downloads 367

Authors: Shaukat Ali Shahee, Usha Ananthakumar

Abstract:

A data set exhibits class imbalance problem when one class has very few examples compared to the other class, and this is also referred to as between class imbalance. The traditional classifiers fail to classify the minority class examples correctly due to its bias towards the majority class. Apart from between-class imbalance, imbalance within classes where classes are composed of a different number of sub-clusters with these sub-clusters containing different number of examples also deteriorates the performance of the classifier. Previously, many methods have been proposed for handling imbalanced dataset problem. These methods can be classified into four categories: data preprocessing, algorithmic based, cost-based methods and ensemble of classifier. Data preprocessing techniques have shown great potential as they attempt to improve data distribution rather than the classifier. Data preprocessing technique handles class imbalance either by increasing the minority class examples or by decreasing the majority class examples. Decreasing the majority class examples lead to loss of information and also when minority class has an absolute rarity, removing the majority class examples is generally not recommended. Existing methods available for handling class imbalance do not address both between-class imbalance and within-class imbalance simultaneously. In this paper, we propose a method that handles between class imbalance and within class imbalance simultaneously for binary classification problem. Removing between class imbalance and within class imbalance simultaneously eliminates the biases of the classifier towards bigger sub-clusters by minimizing the error domination of bigger sub-clusters in total error. The proposed method uses model-based clustering to find the presence of sub-clusters or sub-concepts in the dataset. The number of examples oversampled among the sub-clusters is determined based on the complexity of sub-clusters. The method also takes into consideration the scatter of the data in the feature space and also adaptively copes up with unseen test data using Lowner-John ellipsoid for increasing the accuracy of the classifier. In this study, neural network is being used as this is one such classifier where the total error is minimized and removing the between-class imbalance and within class imbalance simultaneously help the classifier in giving equal weight to all the sub-clusters irrespective of the classes. The proposed method is validated on 9 publicly available data sets and compared with three existing oversampling techniques that rely on the spatial location of minority class examples in the euclidean feature space. The experimental results show the proposed method to be statistically significantly superior to other methods in terms of various accuracy measures. Thus the proposed method can serve as a good alternative to handle various problem domains like credit scoring, customer churn prediction, financial distress, etc., that typically involve imbalanced data sets.

Keywords: classification, imbalanced dataset, Lowner-John ellipsoid, model based clustering, oversampling

Procedia PDF Downloads 389

Authors: Akinwumi Omotayo, Badu-B Apraku, Joseph Olobasola, Petra Abdul Saghir, Yinka Sobowale

Abstract:

In southwestern Nigeria, climate change has led to loss of at least two months of rainfall. Consequently, only one cycle of maize can now be grown because of the shorter duration of rainy season as against two cycles in the past. The Early and Extra-early maturing varieties of maize were originally developed for the semi-arid and arid zones of West and Central Africa where there are seasonal challenges of water threatening optimum performance of the traditional maize grown, which are commonly late in maturity (115 to 120 days). The early varieties of maize mature in 90 to 95 days; while the Extra-Early maize varieties reach physiological maturity in less than 90 days. It was broadly hypothesized that the extra early varieties of maize could mitigate the effects of climate change in southwestern Nigeria with higher levels of rainfall by reinstating the original two cycles of rain-fed maize crop. Trials were therefore carried out in southwestern Nigeria on the possibility of adapting the extra early maize to mitigate the effects of climate change. The trial was the Mother/Baby design. The mother trial involves the evaluation of extra-early varieties following ideal recommendations and closely supervised centrally at the University research farm and the Agricultural Development Programmes (ADPs). This requires farmers to observe and evaluate the technology and the management regime meant to precede the second stage of evaluation at several satellite farmers field managed by selected farmers. The Baby Trial is expected to provide a realistic assessment of the technology by farmers in their own environment. A stratified selection of thirty farmers for the Baby Trial ensured appropriate representation across the different categories of the farming population by age and gender. Data from the trials indicate that extra early maize can be grown in two cycles rain fed in south west Nigeria and a third and fourth cycle could be obtained with irrigation. However the long duration varieties outyielded the extra early maize in both the mother and baby trials. When harvested green, the extra early maize served as source of food between March and May when there was scarcity of food. This represents a major advantage. The study recommends that further work needs to be done to improve the yield of extra early maize to encourage farmers to adopt.

Keywords: adaptation, climate change, extra early, maize varieties, mitigation

Procedia PDF Downloads 168

Authors: Yarra Sulina, Nunna Tagore Sai Priya, Ankhi Banerjee

Abstract:

With the rapid growth in tourist arrivals, there has been a huge demand for the growth of infrastructure in the destinations. With the increasing preference of tourists to stay near attractions, there has been a considerable change in the land use around tourist sites. However, with the inclusion of certain regulations and guidelines provided by the authorities based on the nature of tourism activity and geographical constraints, the pattern of growth of built form is different for various tourist sites. Therefore, this study explores the patterns of growth of built-up for a decade from 2009 to 2019 through two-dimensional and three-dimensional analysis. Land use maps are created through supervised classification of satellite images obtained from LANDSAT 4-5 and LANDSAT 8 for 2009 and 2019, respectively. The overall expansion of the built-up area in the region is analyzed in relation to the distance from the city's geographical center and the tourism-related growth regions are identified which are influenced by the proximity of tourist attractions. The primary tourist sites of various destinations with different geographical characteristics and tourism activities, that have undergone a significant increase in built-up area and are occupied with tourism-related infrastructure are selected for further study. Proximity analysis of the tourism-related growth sites is carried out to delineate the influence zone of the tourist site in a destination. Further, a temporal analysis of volumetric growth of built form is carried out to understand the morphology of the tourist precincts over time. The Digital Surface Model (DSM) and Digital Terrain Model (DTM) are used to extract the building footprints along with building height. Factors such as building height, and building density are evaluated to understand the patterns of three-dimensional growth of the built area in the region. The study also explores the underlying reasons for such changes in built form around various tourist sites and predicts the impact of such growth patterns in the region. The building height and building density around tourist site creates a huge impact on the appeal of the destination. The surroundings that are incompatible with the theme of the tourist site have a negative impact on the attractiveness of the destination that leads to negative feedback by the tourists, which is not a sustainable form of development. Therefore, proper spatial measures are necessary in terms of area and volume of the built environment for a healthy and sustainable environment around the tourist sites in the destination.

Keywords: sustainable tourism, growth patterns, land-use changes, 3-dimensional analysis of built-up area

Procedia PDF Downloads 52

Authors: Erick C. Valverde, Wansu Lim

Abstract:

Facial emotion recognition (FER) system has been recently developed for more advanced computer vision applications. The ability to identify human emotions would enable smart healthcare facility to diagnose mental health illnesses (e.g., depression and stress) as well as better human social interactions with smart technologies. The FER system involves two steps: 1) face detection task and 2) facial emotion recognition task. It classifies the human expression in various categories such as angry, disgust, fear, happy, sad, surprise, and neutral. This system requires intensive research to address issues with human diversity, various unique human expressions, and variety of human facial features due to age differences. These issues generally affect the ability of the FER system to detect human emotions with high accuracy. Early stage of FER systems used simple supervised classification task algorithms like K-nearest neighbors (KNN) and artificial neural networks (ANN). These conventional FER systems have issues with low accuracy due to its inefficiency to extract significant features of several human emotions. To increase the accuracy of FER systems, deep learning (DL)-based methods, like convolutional neural networks (CNN), are proposed. These methods can find more complex features in the human face by means of the deeper connections within its architectures. However, the inference speed and computational costs of a DL-based FER system is often disregarded in exchange for higher accuracy results. To cope with this drawback, an optimized DL-based FER system is proposed in this study.An extreme version of Inception V3, known as Xception model, is leveraged by applying different network optimization methods. Specifically, network pruning and quantization are used to enable lower computational costs and reduce memory usage, respectively. To support low resource requirements, a 68-landmark face detector from Dlib is used in the early step of the FER system.Furthermore, a DL compiler is utilized to incorporate advanced optimization techniques to the Xception model to improve the inference speed of the FER system. In comparison to VGG-Net and ResNet50, the proposed optimized DL-based FER system experimentally demonstrates the objectives of the network optimization methods used. As a result, the proposed approach can be used to create an efficient and real-time FER system.

Keywords: deep learning, face detection, facial emotion recognition, network optimization methods

Procedia PDF Downloads 78

Authors: Fagner O. Serrano, Danielle R. Bouchard, Todd A. Duhame

Abstract:

Walking cadence (steps/min) is an effective way to prescribe exercise so an individual can reach a moderate intensity, which is recommended to optimize health benefits. To our knowledge, there is no study on the required walking cadence to reach a moderate intensity for people that present chronic conditions or risk factors for chronic conditions such as Cardiovascular Diseases (CVD). The objectives of this study were: 1- to identify the walking cadence needed for people at risk of CVD to a reach moderate intensity, and 2- to develop and test an equation using clinical variables to help professionals working with individuals at risk of CVD to estimate the walking cadence needed to reach moderate intensity. Ninety-one people presenting a minimum of two risk factors for CVD completed a medically supervised graded exercise test to assess maximum oxygen consumption at the first visit. The last visit consisted of recording walking cadence using a foot pod Garmin FR-60 and a Polar heart rate monitor, aiming to get participants to reach 40% of their maximal oxygen consumption using a portable metabolic cart on an indoor flat surface. The equation to predict the walking cadence needed to reach moderate intensity in this sample was developed as follows: The sample was randomly split in half and the equation was developed with one half of the participants, and validated using the other half. Body mass index, height, stride length, leg height, body weight, fitness level (VO2max), and self-selected cadence (over 200 meters) were measured using objective measured. Mean walking cadence to reach moderate intensity for people age 64.3 ± 10.3 years old at risk of CVD was 115.8  10.3 steps per minute. Body mass index, height, body weight, fitness level, and self-selected cadence were associated with walking cadence at moderate intensity when evaluated in bivariate analyses (r ranging from 0.22 to 0.52; all P values ≤0.05). Using linear regression analysis including all clinical variables associated in the bivariate analyses, body weight was the significant predictor of walking cadence for reaching a moderate intensity (ß=0.24; P=.018) explaining 13% of walking cadence to reach moderate intensity. The regression model created was Y = 134.4-0.24 X body weight (kg).Our findings suggest that people presenting two or more risk factors for CVD are reaching moderate intensity while walking at a cadence above the one officially recommended (116 steps per minute vs. 100 steps per minute) for healthy adults.

Keywords: cardiovascular disease, moderate intensity, older adults, walking cadence

Procedia PDF Downloads 417

Authors: Martyn Queen, Diane Crone, Andrew Parker, Saul Bloxham

Abstract:

Rationale: There is a growing body of evidence that supports the use of physical activity during and after cancer treatment. However, activity levels for patients remain low. As more cancer patients are treated successfully, and treatment costs continue to escalate, physical activity may be a promising adjunct to a person-centred healthcare approach to recovery. Aim: The aim was to further understand how physical activity may enhance the recovery process for a group of mixed-site cancer patients. Objectives: The research investigated longitudinal changes in physical activity and perceived the quality of life between two and six month’s post-exercise interventions. It also investigated support systems that enabled patients to sustain these perceived changes. Method: The respondent cohort comprised 14 mixed-site cancer patients aged 43-70 (11 women, 3 men), who participated in a two-phase physical activity intervention that took place at a university in the South West of England. Phase 1 consisted of an eight-week structured physical activity programme; Phase 2 consisted of four months of non-supervised physical activity. Semi-structured interviews took place three times over six months with each participant. Grounded theory informed the data collection and analysis which, in turn, facilitated theoretical development. Findings: Our findings propose three theories on the impact of physical activity for recovering cancer patients: 1) Knowledge gained through a structured exercise programme can enable recovering cancer patients to independently sustain physical activity to four-month follow-up. 2) Sustaining physical activity for six months promotes positive changes in the quality of life indicators of chronic fatigue, self-efficacy, the ability to self-manage and energy levels. 3) Peer support from patients facilitates adherence to a structured exercise programme and support from a spouse, or life partner facilitates independently sustained physical activity to four-month follow-up. Conclusions: This study demonstrates that qualitative research can provide an evidence base that could be used to support future care plans for cancer patients. Findings also demonstrate that a physical activity intervention can be effective at helping cancer patients recover from the side effects of their treatment, and recommends that physical activity should become an adjunct therapy alongside traditional cancer treatments.

Keywords: physical activity, health, cancer recovery, quality of life, support systems, qualitative, grounded theory, person-centred healthcare

Procedia PDF Downloads 256

Authors: Cemil Akcay, Goksun Yerlikaya

Abstract:

Buildings are the living spaces of people with an active role in every aspect of life in today's world. While some structures have survived from the early ages, most of the buildings that completed their lifetime have not transported to the present day. Nowadays, buildings that do not meet the social, economic, and safety requirements of the age return to life with a transformation process. This transformation is called urban transformation. Urban transformation is the renewal of the areas with a risk of disaster and the technological infrastructure required by the structure. The transformation aims to prevent damage to earthquakes and other disasters by rebuilding buildings that have completed their non-earthquake-resistant economic life. It is essential to decide on other issues related to conversion and transformation in places where most of the building stock should transform into the first-degree earthquake belt, such as Istanbul. In urban transformation, property owners, local authority, and contractor must deal at a common point. Considering that hundreds of thousands of property owners are sometimes in the areas of transformation, it is evident how difficult it is to make the deal and decide. For the optimization of these decisions, the use of game theory is foreseeing. The main problem in this study is that the urban transformation is carried out in place, or the building or buildings are transport to a different location. There are many stakeholders in the Istanbul University Cerrahpaşa Medical Faculty Campus, which is planned to be carried out in the process of urban transformation, was tried to solve the game theory applications. An analysis of the decisions given on a real urban transformation project and the logical suitability of decisions taken without the use of game theory were also supervised using game theory. In each step of this study, many decision-makers are classifying according to a specific logical sequence, and in the game trees that emerged as a result of this classification, Nash balances were tried to observe, and optimum decisions were determined. All decisions taken for this project have been subjected to two significant differentiated comparisons using game theory, and as decisions are taken without the use of game theory, and according to the results, solutions for the decision phase of the urban transformation process introduced. The game theory model developed from beginning to the end of the urban transformation process, particularly as a solution to the difficulty of making rational decisions in large-scale projects with many participants in the decision-making process. The use of a decision-making mechanism can provide an optimum answer to the demands of the stakeholders. In today's world for the construction sector, it is also seeing that the game theory is a non-surprising consequence of the fact that it is the most critical issues of planning and making the right decision in future years.

Keywords: urban transformation, the game theory, decision making, multi-actor project

Procedia PDF Downloads 102

Authors: Noureddine Henka, Sami Tazi, Mohamad Assaad

Abstract:

The electrical grid is a highly intricate system designed to transfer electricity from production areas to consumption areas. The Transmission System Operator (TSO) is responsible for ensuring the efficient distribution of electricity and maintaining the grid's safety and quality. However, due to the increasing integration of intermittent renewable energy sources, there is a growing level of uncertainty, which requires a faster responsive approach. A potential solution involves the use of electrical segmentation, which involves creating coherence zones where electrical disturbances mainly remain within the zone. Indeed, by means of coherent electrical zones, it becomes possible to focus solely on the sub-zone, reducing the range of possibilities and aiding in managing uncertainty. It allows faster execution of operational processes and easier learning for supervised machine learning algorithms. Electrical segmentation can be applied to various applications, such as electrical control, minimizing electrical loss, and ensuring voltage stability. Since the electrical grid can be modeled as a graph, where the vertices represent electrical buses and the edges represent electrical lines, identifying coherent electrical zones can be seen as a clustering task on graphs, generally called community detection. Nevertheless, a critical criterion for the zones is their ability to remain resilient to the electrical evolution of the grid over time. This evolution is due to the constant changes in electricity generation and consumption, which are reflected in graph structure variations as well as line flow changes. One approach to creating a resilient segmentation is to design robust zones under various circumstances. This issue can be represented through a multiplex graph, where each layer represents a specific situation that may arise on the grid. Consequently, resilient segmentation can be achieved by conducting community detection on this multiplex graph. The multiplex graph is composed of multiple graphs, and all the layers share the same set of vertices. Our proposal involves a model that utilizes a unified representation to compute a flattening of all layers. This unified situation can be penalized to obtain (K) connected components representing the robust electrical segmentation clusters. We compare our robust segmentation to the segmentation based on a single reference situation. The robust segmentation proves its relevance by producing clusters with high intra-electrical perturbation and low variance of electrical perturbation. We saw through the experiences when robust electrical segmentation has a benefit and in which context.

Keywords: community detection, electrical segmentation, multiplex graph, power grid

Procedia PDF Downloads 45

Authors: Uswah Ahmad Khan, Muhammad Moazam Fraz, Humayoon Shafique Satti, Qasim Aziz

Abstract:

Approximately 10-14% of the global population experiences a functional disorder known as irritable bowel syndrome (IBS). The disorder is defined by persistent abdominal pain and an irregular bowel pattern. IBS significantly impairs work productivity and disrupts patients' daily lives and activities. Although IBS is widespread, there is still an incomplete understanding of its underlying pathophysiology. This study aims to help characterize the phenotype of IBS patients by differentiating the comorbidities found in IBS patients from those in non-IBS patients using machine learning algorithms. In this study, we extracted samples coding for IBS from the UK BioBank cohort and randomly selected patients without a code for IBS to create a total sample size of 18,000. We selected the codes for comorbidities of these cases from 2 years before and after their IBS diagnosis and compared them to the comorbidities in the non-IBS cohort. Machine learning models, including Decision Trees, Gradient Boosting, Support Vector Machine (SVM), AdaBoost, Logistic Regression, and XGBoost, were employed to assess their accuracy in predicting IBS. The most accurate model was then chosen to identify the features associated with IBS. In our case, we used XGBoost feature importance as a feature selection method. We applied different models to the top 10% of features, which numbered 50. Gradient Boosting, Logistic Regression and XGBoost algorithms yielded a diagnosis of IBS with an optimal accuracy of 71.08%, 71.427%, and 71.53%, respectively. Among the comorbidities most closely associated with IBS included gut diseases (Haemorrhoids, diverticular diseases), atopic conditions(asthma), and psychiatric comorbidities (depressive episodes or disorder, anxiety). This finding emphasizes the need for a comprehensive approach when evaluating the phenotype of IBS, suggesting the possibility of identifying new subsets of IBS rather than relying solely on the conventional classification based on stool type. Additionally, our study demonstrates the potential of machine learning algorithms in predicting the development of IBS based on comorbidities, which may enhance diagnosis and facilitate better management of modifiable risk factors for IBS. Further research is necessary to confirm our findings and establish cause and effect. Alternative feature selection methods and even larger and more diverse datasets may lead to more accurate classification models. Despite these limitations, our findings highlight the effectiveness of Logistic Regression and XGBoost in predicting IBS diagnosis.

Keywords: comorbidities, disease association, irritable bowel syndrome (IBS), predictive analytics

Procedia PDF Downloads 86

Authors: Ankit Tyagi, Reet Kamal Tiwari, Naveen James

Abstract:

Landslide is the major geo-environmental problem of Himalaya because of high ridges, steep slopes, deep valleys, and complex system of streams. They are mainly triggered by rainfall and earthquake and causing severe damage to life and property. In Uttarakhand, the Tehri reservoir rim area, which is situated in the lesser Himalaya of Garhwal hills, was selected for landslide hazard zonation (LHZ). The study utilized different types of data, including geological maps, topographic maps from the survey of India, Landsat 8, and Cartosat DEM data. This paper presents the use of a weighted overlay method in LHZ using fourteen causative factors. The various data layers generated and co-registered were slope, aspect, relative relief, soil cover, intensity of rainfall, seismic ground shaking, seismic amplification at surface level, lithology, land use/land cover (LULC), normalized difference vegetation index (NDVI), topographic wetness index (TWI), stream power index (SPI), drainage buffer and reservoir buffer. Seismic analysis is performed using peak horizontal acceleration (PHA) intensity and amplification factors in the evaluation of the landslide hazard index (LHI). Several digital image processing techniques such as topographic correction, NDVI, and supervised classification were widely used in the process of terrain factor extraction. Lithological features, LULC, drainage pattern, lineaments, and structural features are extracted using digital image processing techniques. Colour, tones, topography, and stream drainage pattern from the imageries are used to analyse geological features. Slope map, aspect map, relative relief are created by using Cartosat DEM data. DEM data is also used for the detailed drainage analysis, which includes TWI, SPI, drainage buffer, and reservoir buffer. In the weighted overlay method, the comparative importance of several causative factors obtained from experience. In this method, after multiplying the influence factor with the corresponding rating of a particular class, it is reclassified, and the LHZ map is prepared. Further, based on the land-use map developed from remote sensing images, a landslide vulnerability study for the study area is carried out and presented in this paper.

Keywords: weighted overlay method, GIS, landslide hazard zonation, remote sensing

Procedia PDF Downloads 106

Authors: Sandra Figueiredo, Alexandra Pereira, Ana Oliveira, Idia Brito, Ivaniltan Jones, Joana Moreira, Madalena Silva, Maria Paraíba, Milene Silva, Tânia Pinho

Abstract:

In the present society, we are facing behaviours mainly in young individuals that might be considered trends of culture or psychopathology. Both contexts are challenges for Education, Psychology and Health. This paper examines nine case studies specifically in Educational Psychology with the main goal to identify and define phenomena contexts in school culture, the psychopathology involved and to present a psychological intervention for each case. The research was conducted by university students in the period of March 2017-June 2017, in Portugal, and the childhood was focused. The case studies explored the cyberbullying; the bullying - victims and bullies’ perspectives; the obsessive compulsive disorder; perception and inclusion of children from homoparental families; inclusion of foreign students in the higher education system; blindness and the inclusion in physical curricular activities; influence of doc-reality and media in attitudes and self-esteem; and the morningness and eveningness types learning in the same school timetables. The university students were supervised during their research analysis and two methods were available for the intervention research study: the meta-analysis and the empirical study. In the second phase, the pedagogical intervention was designed for the different educational contexts in analysis, especially concerning the school environments. The evidence of literature and the empirical studies showed new trends of school’ behaviours and educational disturbances that require further research and effective (and adequate to age, gender, nationality and culture) pedagogical instruments. Respecting the instruments, on the one hand, to identify behaviors, habits or pathologies and highlight the role and training of teachers, psychologists and health professionals, on the other hand, to promote the early intervention and to enhance healthy child development and orientation of the families. To respond to both milestones, this paper present nine pedagogical techniques and measures that will be discussed on their impact concerning advances for the psychological and educational intervention, centered in the individual and in the new generations of family’ cultures.

Keywords: behaviour, culture trends, educational intervention, psychopathology, obsessive compulsive disorder, cyberbullying, bullying, homoparental families, sleep influence, blindness and sports at school, inclusion of foreign students, media influence in behaviour

Procedia PDF Downloads 182

Authors: G. Singer, M. Golan

Abstract:

Providing students with learning disabilities (LD) with extra time to grant them equal access to the exam is a necessary but insufficient condition to compensate for their LD; there should also be a clear indication that the additional time was actually used. For example, if students with LD use more time than students without LD and yet receive lower grades, this may indicate that a different accommodation is required. If they achieve higher grades but use the same amount of time, then the effectiveness of the accommodation has not been demonstrated. The main goal of this study is to evaluate the effect of including parameters related to LD and extended exam time, along with other commonly-used characteristics (e.g., student background and ability measures such as high-school grades), on the ability of ordinal decision-tree algorithms to predict exam performance. We use naturally-occurring data collected from hundreds of undergraduate engineering students. The sub-goals are i) to examine the improvement in prediction accuracy when the indicator of exam performance includes 'actual time used' in addition to the conventional indicator (exam grade) employed in most research; ii) to explore the effectiveness of extended exam time on exam performance for different courses and for LD students with different profiles (i.e., sets of characteristics). This is achieved by using the patterns (i.e., subgroups) generated by the algorithms to identify pairs of subgroups that differ in just one characteristic (e.g., course or type of LD) but have different outcomes in terms of exam performance (grade and time used). Since grade and time used to exhibit an ordering form, we propose a method based on ordinal decision-trees, which applies a weighted information-gain ratio (WIGR) measure for selecting the classifying attributes. Unlike other known ordinal algorithms, our method does not assume monotonicity in the data. The proposed WIGR is an extension of an information-theoretic measure, in the sense that it adjusts to the case of an ordinal target and takes into account the error severity between two different target classes. Specifically, we use ordinal C4.5, random-forest, and AdaBoost algorithms, as well as an ensemble technique composed of ordinal and non-ordinal classifiers. Firstly, we find that the inclusion of LD and extended exam-time parameters improves prediction of exam performance (compared to specifications of the algorithms that do not include these variables). Secondly, when the indicator of exam performance includes 'actual time used' together with grade (as opposed to grade only), the prediction accuracy improves. Thirdly, our subgroup analyses show clear differences in the effect of extended exam time on exam performance among different courses and different student profiles. From a methodological perspective, we find that the ordinal decision-tree based algorithms outperform their conventional, non-ordinal counterparts. Further, we demonstrate that the ensemble-based approach leverages the strengths of each type of classifier (ordinal and non-ordinal) and yields better performance than each classifier individually.

Keywords: actual exam time usage, ensemble learning, learning disabilities, ordinal classification, time extension

Procedia PDF Downloads 68

Authors: Mehrnaz Mostafavi

Abstract:

The assessment and categorization of incidental lung nodules present a considerable challenge in healthcare, often necessitating resource-intensive multiple computed tomography (CT) scans for growth confirmation. This research addresses this issue by introducing a distinct computational approach leveraging radiomics and deep-learning methods. However, understanding local services is essential before implementing these advancements. With diverse tracking methods in place, there is a need for efficient and accurate identification approaches, especially in the context of managing lung nodules alongside pre-existing cancer scenarios. This study explores the integration of text-based algorithms in medical data curation, indicating their efficacy in conjunction with machine learning and deep-learning models for identifying lung nodules. Combining medical images with text data has demonstrated superior data retrieval compared to using each modality independently. While deep learning and text analysis show potential in detecting previously missed nodules, challenges persist, such as increased false positives. The presented research introduces a Structured-Query-Language (SQL) algorithm designed for identifying pulmonary nodules in a tertiary cancer center, externally validated at another hospital. Leveraging natural language processing (NLP) and machine learning, the algorithm categorizes lung nodule reports based on sentence features, aiming to facilitate research and assess clinical pathways. The hypothesis posits that the algorithm can accurately identify lung nodule CT scans and predict concerning nodule features using machine-learning classifiers. Through a retrospective observational study spanning a decade, CT scan reports were collected, and an algorithm was developed to extract and classify data. Results underscore the complexity of lung nodule cohorts in cancer centers, emphasizing the importance of careful evaluation before assuming a metastatic origin. The SQL and NLP algorithms demonstrated high accuracy in identifying lung nodule sentences, indicating potential for local service evaluation and research dataset creation. Machine-learning models exhibited strong accuracy in predicting concerning changes in lung nodule scan reports. While limitations include variability in disease group attribution, the potential for correlation rather than causality in clinical findings, and the need for further external validation, the algorithm's accuracy and potential to support clinical decision-making and healthcare automation represent a significant stride in lung nodule management and research.

Keywords: lung cancer diagnosis, structured-query-language (SQL), natural language processing (NLP), machine learning, CT scans

Procedia PDF Downloads 37

Authors: Bjornar Kjellstadli, Shaher A. I. Shalfawi

Abstract:

The aim of this study was to investigate the effect of strength training Cluster set-method compared to traditional set-method 30 m sprinting time and maximum strength in squats and bench-press. Thirteen Physical Education students, 7 males and 6 females between the age of 19-28 years old were recruited. The students were random divided in three groups. Traditional set group (TSG) consist of 2 males and 2 females aged (±SD) (22.3 ± 1.5 years), body mass (79.2 ± 15.4 kg) and height (177.5 ± 11.3 cm). Cluster set group (CSG) consist of 3 males and 2 females aged (22.4 ± 3.29 years), body mass (81.0 ± 24.0 kg) and height (179.2 ± 11.8 cm) and a control group (CG) consist of 2 males and 2 females aged (21.5 ± 2.4 years), body mass (82.1 ± 17.4 kg) and height (175.5 ± 6.7 cm). The intervention consisted of performing squat and bench press at 70% of 1RM (twice a week) for 8 weeks using 10 repetition and 4 sets. Two types of strength-training methods were used , cluster set (CS) where the participants (CSG) performed 2 reps 5 times with a 10 s recovery in between reps and 50 s recovery between sets, and traditional set (TS) where the participants (TSG) performed 10 reps each set with 90 s recovery in between sets. The pre-tests and post-tests conducted were 1 RM in both squats and bench press, and 10 and 30 m sprint time. The 1RM test were performed with Eleiko XF barbell (20 kg), Eleiko weight plates, rack and bench from Hammerstrength. The speed test was measured with the Brower speed trap II testing system (Brower Timing Systems, Utah, USA). The participants received an individualized training program based on the pre-test of the 1RM. In addition, a mid-term test of 1RM was carried out to adjust training intensity. Each training session were supervised by the researchers. Beast sensors (Milano, Italy) were also used to monitor and quantify the training load for the participants. All groups had a statistical significant improvement in bench press 1RM (TSG 1RM from 56.3 ± 28.9 to 66 ± 28.5 kg; CSG 1RM from 69.8 ± 33.5 to 77.2 ± 34.1 kg and CG 1RM from 67.8 ± 26.6 to 72.2 ± 29.1 kg), whereas only the TSG (1RM from 84.3 ± 26.8 to 114.3 ± 26.5 kg) and CSG (1RM from 100.4 ± 33.9 to 129 ± 35.1 kg) had a statistical significant improvement in Squats 1RM (P < 0.05). However, a between groups examination reveals that there were no marked differences in 1RM squat performance between TSG and CSG (P > 0.05) and both groups had a marked improvements compared to the CG (P < 0.05). On the other hand, no differences between groups were observed in Bench press 1RM. The within groups results indicate that none of the groups had any marked improvement in the distances from 0-10 m and 10-30 m except the CSG which had a notable improvement in the distance from 10-30 m (-0.07 s; P < 0.05). Furthermore, no differences in sprinting abilities were observed between groups. The results from this investigation indicate that traditional set strength training at 70% of 1RM gave close results compared to Cluster set strength training at the same intensity. However, the results indicate that the cluster set had an effect on flying time (10-30 m) indicating that the velocity at which those repetitions were performed could be the explanation factor of this this improvement.

Keywords: physical performance, 1RM, pushing velocity, velocity based training

Procedia PDF Downloads 139

Authors: Asma Omri, Noureddine Benothman, Sofiane Sayahi, Fethi Tlili, Hichem Besbes

Abstract:

Cycling, a popular mode of transportation, can also be perilous due to cyclists' vulnerability to collisions with vehicles and obstacles. This paper presents an innovative cyclist safety system based on radar technology designed to offer real-time collision risk warnings to cyclists. The system incorporates a low-power radar sensor affixed to the bicycle and connected to a microcontroller. It leverages radar point cloud detections, a clustering algorithm, and a supervised classifier. These algorithms are optimized for efficiency to run on the TI’s AWR 1843 BOOST radar, utilizing a coarse classification approach distinguishing between cars, trucks, two-wheeled vehicles, and other objects. To enhance the performance of clustering techniques, we propose a 2-Level clustering approach. This approach builds on the state-of-the-art Density-based spatial clustering of applications with noise (DBSCAN). The objective is to first cluster objects based on their velocity, then refine the analysis by clustering based on position. The initial level identifies groups of objects with similar velocities and movement patterns. The subsequent level refines the analysis by considering the spatial distribution of these objects. The clusters obtained from the first level serve as input for the second level of clustering. Our proposed technique surpasses the classical DBSCAN algorithm in terms of geometrical metrics, including homogeneity, completeness, and V-score. Relevant cluster features are extracted and utilized to classify objects using an SVM classifier. Potential obstacles are identified based on their velocity and proximity to the cyclist. To optimize the system, we used the View of Delft dataset for hyperparameter selection and SVM classifier training. The system's performance was assessed using our collected dataset of radar point clouds synchronized with a camera on an Nvidia Jetson Nano board. The radar-based cyclist safety system is a practical solution that can be easily installed on any bicycle and connected to smartphones or other devices, offering real-time feedback and navigation assistance to cyclists. We conducted experiments to validate the system's feasibility, achieving an impressive 85% accuracy in the classification task. This system has the potential to significantly reduce the number of accidents involving cyclists and enhance their safety on the road.

Keywords: 2-level clustering, coarse classification, cyclist safety, warning system based on radar technology

Procedia PDF Downloads 45

Authors: Julia Ive, Lucia Specia

Abstract:

Understanding and detecting toxicity is an important problem to support safer human interactions online. Our work focuses on the important problem of contextual toxicity detection, where automated classifiers are tasked with determining whether a short textual segment (usually a sentence) is toxic within its conversational context. We use “toxicity” as an umbrella term to denote a number of variants commonly named in the literature, including hate, abuse, offence, among others. Detecting toxicity in context is a non-trivial problem and has been addressed by very few previous studies. These previous studies have analysed the influence of conversational context in human perception of toxicity in controlled experiments and concluded that humans rarely change their judgements in the presence of context. They have also evaluated contextual detection models based on state-of-the-art Deep Learning and Natural Language Processing (NLP) techniques. Counterintuitively, they reached the general conclusion that computational models tend to suffer performance degradation in the presence of context. We challenge these empirical observations by devising better contextual predictive models that also rely on NLP data augmentation techniques to create larger and better data. In our study, we start by further analysing the human perception of toxicity in conversational data (i.e., tweets), in the absence versus presence of context, in this case, previous tweets in the same conversational thread. We observed that the conclusions of previous work on human perception are mainly due to data issues: The contextual data available does not provide sufficient evidence that context is indeed important (even for humans). The data problem is common in current toxicity datasets: cases labelled as toxic are either obviously toxic (i.e., overt toxicity with swear, racist, etc. words), and thus context does is not needed for a decision, or are ambiguous, vague or unclear even in the presence of context; in addition, the data contains labeling inconsistencies. To address this problem, we propose to automatically generate contextual samples where toxicity is not obvious (i.e., covert cases) without context or where different contexts can lead to different toxicity judgements for the same tweet. We generate toxic and non-toxic utterances conditioned on the context or on target tweets using a range of techniques for controlled text generation(e.g., Generative Adversarial Networks and steering techniques). On the contextual detection models, we posit that their poor performance is due to limitations on both of the data they are trained on (same problems stated above) and the architectures they use, which are not able to leverage context in effective ways. To improve on that, we propose text classification architectures that take the hierarchy of conversational utterances into account. In experiments benchmarking ours against previous models on existing and automatically generated data, we show that both data and architectural choices are very important. Our model achieves substantial performance improvements as compared to the baselines that are non-contextual or contextual but agnostic of the conversation structure.

Keywords: contextual toxicity detection, data augmentation, hierarchical text classification models, natural language processing

Procedia PDF Downloads 137

Authors: Leif Inge Tjelta, Gerd Lise Nordbotten, Cathrine Nyhus Hagum, Merete Hagen Helland

Abstract:

Overweight and obesity can lead to numerous health problems, and inactive people are more often overweight and obese compared to physically active people. Even a moderate weight loss can improve cardiovascular and endocrine disease risk factors. The aim of the study was to examine to what extent overweight and obese adults starting up with two weekly intensive running sessions had an increase in aerobic capacity, reduction in BMI and waist circumference and changes in body composition after 33 weeks of training. An additional aim was to see if there were differences between participants who, in addition to training, also received lifestyle modification education, including practical cooking (nutritional guidance and training group (NTG =32)) compared to those who were not given any nutritional guidance (training group (TG=40)). 72 participants (49 women), mean age of 46.1 ( ± 10.4) were included. Inclusion Criteria: Previous untrained and inactive adults in all age groups, BMI ≥ 25, desire to become fitter and reduce their BMI. The two weekly supervised training sessions consisted of 10 min warm up followed by 20 to 21 min effective interval running where the participants’ heart rate were between 82 and 92% of hearth rate maximum. The sessions were completed with ten minutes whole body strength training. Measures of BMI, waist circumference (WC) and 3000m running time were performed at the start of the project (T1), after 15 weeks (T2) and at the end of the project (T3). Measurements of fat percentage, muscle mass, and visceral fat were performed at T1 and T3. Twelve participants (9 women) from both groups, who all scored around average on the 3000 m pre-test, were chosen to do a VO₂max test at T1 and T3. The NTG were given ten theoretical sessions (80 minutes each) and eight practical cooking sessions (140 minutes each). There was a significant reduction in bout groups for WC and BMI from T1 to T2. There was not found any further reduction from T2 to T3. Although not significant, NTG reduced their WC more than TG. For both groups, the percentage reduction in WC was similar to the reduction in BMI. There was a decrease in fat percentage in both groups from pre-test to post-test, whereas, for muscle mass, a small, but insignificant increase was observed for both groups. There was a decrease in 3000m running time for both groups from T1 to T2 as well as from T2 to T3. The difference between T2 and T3 was not statistically significant. The 12 participants who tested VO₂max had an increase of 2.86 ( ± 3.84) mlkg⁻¹ min⁻¹ in VO₂max and 3:02 min (± 2:01 min) reduction in running time over 3000 m from T1 until T3. There was a strong, negative correlation between the two variables. The study shows that two intensive running session in 33 weeks can increase aerobic fitness and reduce BMI, WC and fat percent in inactive adults. Cost guidance in addition to training will give additional effect.

Keywords: interval training, nutritional guidance, fitness, BMI

Procedia PDF Downloads 121

Authors: Mohammad Karimi Moridani, Mohammad Abdi Zadeh, Zahra Shahiazar Mazraeh

Abstract:

In recent years, the use of intelligent systems in biomedical engineering has increased dramatically, especially in the diagnosis of various diseases. Also, due to the relatively simple recording of the electrocardiogram signal (ECG), this signal is a good tool to show the function of the heart and diseases associated with it. The aim of this paper is to design an intelligent system for automatically detecting a normal electrocardiogram signal from abnormal one. Using this diagnostic system, it is possible to identify a person's heart condition in a very short time and with high accuracy. The data used in this article are from the Physionet database, available in 2016 for use by researchers to provide the best method for detecting normal signals from abnormalities. Data is of both genders and the data recording time varies between several seconds to several minutes. All data is also labeled normal or abnormal. Due to the low positional accuracy and ECG signal time limit and the similarity of the signal in some diseases with the normal signal, the heart rate variability (HRV) signal was used. Measuring and analyzing the heart rate variability with time to evaluate the activity of the heart and differentiating different types of heart failure from one another is of interest to the experts. In the preprocessing stage, after noise cancelation by the adaptive Kalman filter and extracting the R wave by the Pan and Tampkinz algorithm, R-R intervals were extracted and the HRV signal was generated. In the process of processing this paper, a new idea was presented that, in addition to using the statistical characteristics of the signal to create a return map and extraction of nonlinear characteristics of the HRV signal due to the nonlinear nature of the signal. Finally, the artificial neural networks widely used in the field of ECG signal processing as well as distinctive features were used to classify the normal signals from abnormal ones. To evaluate the efficiency of proposed classifiers in this paper, the area under curve ROC was used. The results of the simulation in the MATLAB environment showed that the AUC of the MLP and SVM neural network was 0.893 and 0.947, respectively. As well as, the results of the proposed algorithm in this paper indicated that the more use of nonlinear characteristics in normal signal classification of the patient showed better performance. Today, research is aimed at quantitatively analyzing the linear and non-linear or descriptive and random nature of the heart rate variability signal, because it has been shown that the amount of these properties can be used to indicate the health status of the individual's heart. The study of nonlinear behavior and dynamics of the heart's neural control system in the short and long-term provides new information on how the cardiovascular system functions, and has led to the development of research in this field. Given that the ECG signal contains important information and is one of the common tools used by physicians to diagnose heart disease, but due to the limited accuracy of time and the fact that some information about this signal is hidden from the viewpoint of physicians, the design of the intelligent system proposed in this paper can help physicians with greater speed and accuracy in the diagnosis of normal and patient individuals and can be used as a complementary system in the treatment centers.

Keywords: neart rate variability, signal processing, linear and non-linear features, classification methods, ROC Curve

Procedia PDF Downloads 224

Authors: Anup Anand Deshmukh, Catherine Soladie, Renaud Seguier

Abstract:

Emotion plays a key role in many applications like healthcare, to gather patients’ emotional behavior. Unlike typical ASR (Automated Speech Recognition) problems which focus on 'what was said', it is equally important to understand 'how it was said.' There are certain emotions which are given more importance due to their effectiveness in understanding human feelings. In this paper, we propose an approach that models human stress from audio signals. The research challenge in speech emotion detection is finding the appropriate set of acoustic features corresponding to an emotion. Another difficulty lies in defining the very meaning of emotion and being able to categorize it in a precise manner. Supervised Machine Learning models, including state of the art Deep Learning classification methods, rely on the availability of clean and labelled data. One of the problems in affective computation is the limited amount of annotated data. The existing labelled emotions datasets are highly subjective to the perception of the annotator. We address the first issue of feature selection by exploiting the use of traditional MFCC (Mel-Frequency Cepstral Coefficients) features in Convolutional Neural Network. Our proposed Emo-CNN (Emotion-CNN) architecture treats speech representations in a manner similar to how CNN’s treat images in a vision problem. Our experiments show that Emo-CNN consistently and significantly outperforms the popular existing methods over multiple datasets. It achieves 90.2% categorical accuracy on the Emo-DB dataset. We claim that Emo-CNN is robust to speaker variations and environmental distortions. The proposed approach achieves 85.5% speaker-dependant categorical accuracy for SAVEE (Surrey Audio-Visual Expressed Emotion) dataset, beating the existing CNN based approach by 10.2%. To tackle the second problem of subjectivity in stress labels, we use Lovheim’s cube, which is a 3-dimensional projection of emotions. Monoamine neurotransmitters are a type of chemical messengers in the brain that transmits signals on perceiving emotions. The cube aims at explaining the relationship between these neurotransmitters and the positions of emotions in 3D space. The learnt emotion representations from the Emo-CNN are mapped to the cube using three component PCA (Principal Component Analysis) which is then used to model human stress. This proposed approach not only circumvents the need for labelled stress data but also complies with the psychological theory of emotions given by Lovheim’s cube. We believe that this work is the first step towards creating a connection between Artificial Intelligence and the chemistry of human emotions.

Keywords: deep learning, brain chemistry, emotion perception, Lovheim's cube

Procedia PDF Downloads 124

Authors: Olena Kavats, Dmitry Khramov, Kateryna Sergieieva, Vladimir Vasyliev, Iurii Kavats

Abstract:

Geoinformation technologies of space agromonitoring are a means of operative decision making support in the tasks of managing the agricultural sector of the economy. Existing technologies use satellite images in the optical range of electromagnetic spectrum. Time series of optical images often contain gaps due to the presence of clouds and haze. A geoinformation technology is created. It allows to fill gaps in time series of optical images (Sentinel-2, Landsat-8, PROBA-V, MODIS) with radar survey data (Sentinel-1) and use information about agrometeorological conditions of the growing season for individual monitoring years. The technology allows to perform crop classification and mapping for spring-summer (winter and spring crops) and autumn-winter (winter crops) periods of vegetation, monitoring the dynamics of crop state seasonal changes, crop yield forecasting. Crop classification is based on supervised classification algorithms, takes into account the peculiarities of crop growth at different vegetation stages (dates of sowing, emergence, active vegetation, and harvesting) and agriculture land state characteristics (row spacing, seedling density, etc.). A catalog of samples of the main agricultural crops (Ukraine) is created and crop spectral signatures are calculated with the preliminary removal of row spacing, cloud cover, and cloud shadows in order to construct time series of crop growth characteristics. The obtained data is used in grain crop growth tracking and in timely detection of growth trends deviations from reference samples of a given crop for a selected date. Statistical models of crop yield forecast are created in the forms of linear and nonlinear interconnections between crop yield indicators and crop state characteristics (temperature, precipitation, vegetation indices, etc.). Predicted values of grain crop yield are evaluated with an accuracy up to 95%. The developed technology was used for agricultural areas monitoring in a number of Great Britain and Ukraine regions using EOS Crop Monitoring Platform (https://crop-monitoring.eos.com). The obtained results allow to conclude that joint use of Sentinel-1 and Sentinel-2 images improve separation of winter crops (rapeseed, wheat, barley) in the early stages of vegetation (October-December). It allows to separate successfully the soybean, corn, and sunflower sowing areas that are quite similar in their spectral characteristics.

Keywords: geoinformation technology, crop classification, crop yield prediction, agricultural monitoring, EOS Crop Monitoring Platform

Procedia PDF Downloads 402