Search results for: temporal convolutional net

Authors: Jéssica F. Rodrigues, Amanda C. Andrade, Sabrina C. Bastos, Sandra B. Coelho, Ana Carla M. Pinheiro

Abstract:

Currently, there is a great demand for the use of natural sweeteners due to the harmful effects of the high sugar and artificial sweeteners consumption on the health. Miracle fruit, which is known for its unique ability to modify the sour taste in sweet taste, has been shown to be a good alternative sweetener. However, it has a high production cost, being important to optimize lower contents to be used. Thus, the aim of this study was to assess the effect of different miracle fruit contents on the temporal (Time-intensity - TI and Temporal Dominance of Sensations - TDS) sensory profile and overall linking of lemonade, to determine the better content to be used as a natural sweetener in sour beverages. TI and TDS results showed that the concentrations of 150 mg, 300 mg and 600 mg miracle fruit were effective in reducing the acidity and promoting the sweet perception in lemonade. Furthermore, the concentrations of 300 mg and 600 mg obtained similar profiles. Through the acceptance test, the concentration of 300 mg miracle fruit was shown to be an efficient substitute for sucrose and sucralose in lemonade, once they had similar hedonic values between ‘I liked it slightly’ and ‘I liked it moderately’. Therefore, 300mg miracle fruit consists in an adequate content to be used as a natural sweetener of lemonade. The results of this work will help the food industry on the efficient application of a new natural sweetener- the Miracle fruit extract in sour beverages, reducing costs and providing a product that meets the consumer desires.

Keywords: acceptance, natural sweetener, temporal dominance of sensations, time-intensity

Procedia PDF Downloads 221

Authors: Ananya Ananya, Karthik Rao

Abstract:

Accurate segmentation of knee cartilage in 3-D magnetic resonance (MR) images for quantitative assessment of volume is crucial for studying and diagnosing osteoarthritis (OA) of the knee, one of the major causes of disability in elderly people. Radiologists generally perform this task in slice-by-slice manner taking 15-20 minutes per 3D image, and lead to high inter and intra observer variability. Hence automatic methods for knee cartilage segmentation are desirable and are an active field of research. This paper presents design and experimental evaluation of 2D convolutional neural networks based fully automated methods for knee cartilage segmentation in 3D MRI. The architectures are validated based on 40 test images and 60 training images from SKI10 dataset. The proposed methods segment 2D slices one by one, which are then combined to give segmentation for whole 3D images. Proposed methods are modified versions of U-net and dilated convolutions, consisting of a single step that segments the given image to 5 labels: background, femoral cartilage, tibia cartilage, femoral bone and tibia bone; cartilages being the primary components of interest. U-net consists of a contracting path and an expanding path, to capture context and localization respectively. Dilated convolutions lead to an exponential expansion of receptive field with only a linear increase in a number of parameters. A combination of modified U-net and dilated convolutions has also been explored. These architectures segment one 3D image in 8 – 10 seconds giving average volumetric Dice Score Coefficients (DSC) of 0.950 - 0.962 for femoral cartilage and 0.951 - 0.966 for tibia cartilage, reference being the manual segmentation.

Keywords: convolutional neural networks, dilated convolutions, 3 dimensional, fully automated, knee cartilage, MRI, segmentation, U-net

Procedia PDF Downloads 232

Authors: Barbara Janari

Abstract:

This paper focuses on temporality in J.M. Coetzee’s first novel, Dusklands, to argue that the novel’s fractured, disjointed temporality is intricately linked to the representations of the war in Vietnam and the colonial project in South Africa. The disrupted temporalities in the novel eschew chronological plots and linear time in favour of narratives that subvert the notion of historical progress to suggest instead the coextensive, multivalent ways in which the past and present interpenetrate one another. The disruption of temporal flow in the novel is evident in its form – the novel comprises two novellas that are juxtaposed, with the first part (‘The Vietnam War’) set centuries before the second part (‘The Narrative of Jacobus Coetzee’). The juxtaposition of the two novellas suggests history’s sometimes overlapping and lateral, rather than linear, movement. The novel’s form is extended in its montage narrative structure, which works to extend its temporal range. The temporal disjunction is reinforced, firstly, by Coetzee’s textual strategies, which include the subversion and critique of realism, parody, repetition, and the narrative technique of montage, and secondly, by the novel’s thematic concerns, which focus on the ways in which American domination can be linked to the colonial quest from earlier times. The complex structure of various strands and levels of authorship slows down the narrative’s temporal flow, requiring the reader to spend a fair amount of time unraveling the various parts of the narrative and relating them to each other. The structure epitomizes reflexive referencing, in which the reader can only make sense of the narrative by going back and forth and connecting various parts of it. The narrative structure also emphasizes the underlying similarities in the brutality that marked these two distinct historical events, epitomized by the drive towards subjection and domination by the novel’s two protagonists, Eugene Dawn and Jacobus Coetzee. The links and overlapping strands between the two novellas emphasize the ways in which the historical truth of colonial discourse becomes as much a myth as the propaganda program in Vietnam.

Keywords: disjunction, juxtaposition, montage, temporality

Procedia PDF Downloads 35

Authors: Bikis Muhammed, Sehra Sedigh Sarvestani, Ali R. Hurson, Lasanthi Gamage

Abstract:

Vehicular traffic events have overly complex spatial correlations and temporal interdependencies and are also influenced by environmental events such as weather conditions. To capture these spatial and temporal interdependencies and make more realistic vehicular traffic predictions, graph neural networks (GNN) based traffic prediction models have been extensively utilized due to their capability of capturing non-Euclidean spatial correlation very effectively. However, most of the already existing GNN-based traffic prediction models have some limitations during learning complex and dynamic spatial and temporal patterns due to the following missing factors. First, most GNN-based traffic prediction models have used static distance or sometimes haversine distance mechanisms between spatially separated traffic observations to estimate spatial correlation. Secondly, most GNN-based traffic prediction models have not incorporated environmental events that have a major impact on the normal traffic states. Finally, most of the GNN-based models did not use an attention mechanism to focus on only important traffic observations. The objective of this paper is to study and make real-time vehicular traffic predictions while incorporating the effect of weather conditions. To fill the previously mentioned gaps, our prediction model uses a real-time driving distance between sensors to build a distance matrix or spatial adjacency matrix and capture spatial correlation. In addition, our prediction model considers the effect of six types of weather conditions and has an attention mechanism in both spatial and temporal data aggregation. Our prediction model efficiently captures the spatial and temporal correlation between traffic events, and it relies on the graph attention network (GAT) and Bidirectional bidirectional long short-term memory (Bi-LSTM) plus attention layers and is called GAT-BILSTMA.

Keywords: deep learning, real time prediction, GAT, Bi-LSTM, attention

Procedia PDF Downloads 47

Authors: Anas H. Aljemely, Jianping Xuan

Abstract:

Rolling bearing fault diagnosis plays a pivotal issue in the rotating machinery of modern manufacturing. In this research, a raw vibration signal and improved deep learning method for bearing fault diagnosis are proposed. The multi-dimensional scales of raw vibration signals are selected for evaluation condition monitoring system, and the deep learning process has shown its effectiveness in fault diagnosis. In the proposed method, employing an Exponential linear unit (ELU) layer in a convolutional neural network (CNN) that conducts the identical function on positive data, an exponential nonlinearity on negative inputs, and a particular convolutional operation to extract valuable features. The identification results show the improved method has achieved the highest accuracy with a 100-dimensional scale and increase the training and testing speed.

Keywords: bearing fault prognostics, developed CNN model, multiple-scale evaluation, deep learning features

Procedia PDF Downloads 173

Authors: Kamal KC, Zhendong Yin, Dasen Li, Zhilu Wu

Abstract:

Intra-plant disease classification based on leaf images is a challenging computer vision task due to similarities in texture, color, and shape of leaves with a slight variation of leaf spot; and external environmental changes such as lighting and background noises. Deep convolutional neural network (DCNN) has proven to be an effective tool for binary classification. In this paper, two methods for binary classification of diseased plant leaves using DCNN are presented; model created from scratch and transfer learning. Our main contribution is a thorough evaluation of 4 networks created from scratch and transfer learning of 5 pre-trained models. Training and testing of these models were performed on a plant leaf images dataset belonging to 16 distinct classes, containing a total of 22,265 images from 8 different plants, consisting of a pair of healthy and diseased leaves. We introduce a deep CNN model, Optimized MobileNet. This model with depthwise separable CNN as a building block attained an average test accuracy of 99.77%. We also present a fine-tuning method by introducing the concept of a convolutional block, which is a collection of different deep neural layers. Fine-tuned models proved to be efficient in terms of accuracy and computational cost. Fine-tuned MobileNet achieved an average test accuracy of 99.89% on 8 pairs of [healthy, diseased] leaf ImageSet.

Keywords: deep convolution neural network, depthwise separable convolution, fine-grained classification, MobileNet, plant disease, transfer learning

Procedia PDF Downloads 157

Authors: Parisa Golbayani, Dan Wang, Ionut¸ Florescu

Abstract:

In this work we implement machine learning techniques to financial statement reports in order to asses company’s credit rating. Specifically, the work analyzes the performance of four neural network architectures (MLP, CNN, CNN2D, LSTM) in predicting corporate credit rating as issued by Standard and Poor’s. The paper focuses on companies from the energy, financial, and healthcare sectors in the US. The goal of this analysis is to improve application of machine learning algorithms to credit assessment. To accomplish this, the study investigates three questions. First, we investigate if the algorithms perform better when using a selected subset of important features or whether better performance is obtained by allowing the algorithms to select features themselves. Second, we address the temporal aspect inherent in financial data and study whether it is important for the results obtained by a machine learning algorithm. Third, we aim to answer if one of the four particular neural network architectures considered consistently outperforms the others, and if so under which conditions. This work frames the problem as several case studies to answer these questions and analyze the results using ANOVA and multiple comparison testing procedures.

Keywords: convolutional neural network, long short term memory, multilayer perceptron, credit rating

Procedia PDF Downloads 208

Authors: Yue Hao

Abstract:

COVID-19 vaccines not only protect individuals but society as a whole. In this case, having an understanding of the change and trend of vaccination conditions may shed some light on revising and making up-to-date policies regarding large-scale public health promotions and calls in order to lead and encourage the adoption of COVID-19 vaccines. However, vaccination status change over time and vary from place to place hidden patterns that were not fully explored in previous research. In our research, we took advantage of the spatial-temporal analytical methods in the domain of geographic information science and captured the spatial-temporal changes regarding COVID-19 vaccination status in the United States during 2020 and 2021. After conducting the emerging hot spots analysis on both the state level data of the US and county level data of California we found that: (1) at the macroscopic level, there is a continuously increasing trend of the vaccination rate in the US, but there is a variance on the spatial clusters at county level; (2) spatial hotspots and clusters with high vaccination amount over time were clustered around the west and east coast in regions like California and New York City where are densely populated with considerable economy conditions; (3) in terms of the growing trend of the daily vaccination among, Los Angeles County alone has very high statistics and dramatic increases over time. We hope that our findings can be valuable guidance for supporting future decision-making regarding vaccination policies as well as directing new research on relevant topics.

Keywords: COVID-19 vaccine, GIS, space time cube, spatial-temporal analysis

Procedia PDF Downloads 57

Authors: JooYoung Lee, Rasheed Hussain

Abstract:

Understanding the relationships among communities of users is the key to blueprint the evolution of human society. Majority of people are equipped with GPS devices, such as smart phones and smart cars, which can trace their whereabouts. In this paper, we discover communities of device users based on real locations in a given time frame. We, then, study the associations of discovered communities, referred to as temporal communities, and generate temporal and probabilistic association rules. The rules describe how strong communities are associated. By studying the generated rules, we can automatically extract underlying hierarchies of communities and permanent communities such as work places.

Keywords: association rules, community detection, evolution of communities, spatiotemporal

Procedia PDF Downloads 331

Authors: Wei-Han Wang, Hsiang-Yu Yu, Mau-Sun Hua

Abstract:

Theory of Mind (ToM) refers to the ability to infer another’s mental state. With appropriate ToM, one can behave well in social interactions. A growing body of evidence has demonstrated that patients with temporal lobe epilepsy (TLE) may have damaged ToM due to impact on regions of the underlying neural network of ToM. However, the question of whether there is cerebral laterality for ToM functions remains open. This study aimed to examine whether there is cerebral lateralization for ToM abilities in TLE patients. Sixty-seven adult TLE patients and 30 matched healthy controls (HC) were recruited. Patients were classified into right (RTLE), left (LTLE), and bilateral (BTLE) TLE groups on the basis of a consensus panel review of their seizure semiology, EEG findings, and brain imaging results. All participants completed an intellectual test and four tasks measuring basic and advanced ToM. The results showed that, on all ToM tasks; (1)each patient group performed worse than HC; (2)there were no significant differences between LTLE and RTLE groups; (3)the BTLE group performed the worst. It appears that the neural network responsible for ToM is distributed evenly between the cerebral hemispheres.

Keywords: cerebral lateralization, social cognition, temporal lobe epilepsy, theory of mind

Procedia PDF Downloads 399

Authors: Janani Selvaraj, M. Prashanthi Devi, P. B. Harathi

Abstract:

Epidemiologic studies conducted over several decades have provided evidence to suggest that long-term exposure to elevated ambient levels of particulate air pollution is associated with increased mortality. Air quality risk management is significant in developing countries and it highlights the need to understand the role of ecologic covariates in the association between air pollution and mortality. Several new methods show promise in exploring the geographical distribution of disease and the identification of high risk areas using epidemiological maps. However, the addition of the temporal attribute would further give us an in depth idea of the disease burden with respect to forecasting measures. In recent years, new methods developed in the reanalysis were useful for exploring the spatial structure of the data and the impact of spatial autocorrelation on estimates of risk associated with exposure to air pollution. Based on this, our present study aims to explore the spatial and temporal distribution of the lung cancer cases in the Coimbatore district of Tamil Nadu in relation to air pollution risk areas. A spatio temporal moving average method was computed using the CrimeStat software and visualized in ArcGIS 10.1 to document the spatio temporal movement of the disease in the study region. The random walk analysis performed showed the progress of the peak cancer incidences in the intersection regions of the Coimbatore North and South taluks that include major commercial and residential regions like Gandhipuram, Peelamedu, Ganapathy, etc. Our study shows evidence that daily exposure to high air pollutant concentration zones may lead to the risk of lung cancer. The observations from the present study will be useful in delineating high risk zones of environmental exposure that contribute to the increase of cancer among daily commuters. Through our study we suggest that spatially resolved exposure models in relevant time frames will produce higher risks zones rather than solely on statistical theory about the impact of measurement error and the empirical findings.

Keywords: air pollution, cancer, spatio-temporal analysis, India

Procedia PDF Downloads 486

Authors: Isaac K. E. Ampomah, Seong-Bae Park, Sang-Jo Lee

Abstract:

Over the past decade, there have been promising developments in Natural Language Processing (NLP) with several investigations of approaches focusing on Recognizing Textual Entailment (RTE). These models include models based on lexical similarities, models based on formal reasoning, and most recently deep neural models. In this paper, we present a sentence encoding model that exploits the sentence-to-sentence relation information for RTE. In terms of sentence modeling, Convolutional neural network (CNN) and recurrent neural networks (RNNs) adopt different approaches. RNNs are known to be well suited for sequence modeling, whilst CNN is suited for the extraction of n-gram features through the filters and can learn ranges of relations via the pooling mechanism. We combine the strength of RNN and CNN as stated above to present a unified model for the RTE task. Our model basically combines relation vectors computed from the phrasal representation of each sentence and final encoded sentence representations. Firstly, we pass each sentence through a convolutional layer to extract a sequence of higher-level phrase representation for each sentence from which the first relation vector is computed. Secondly, the phrasal representation of each sentence from the convolutional layer is fed into a Bidirectional Long Short Term Memory (Bi-LSTM) to obtain the final sentence representations from which a second relation vector is computed. The relations vectors are combined and then used in then used in the same fashion as attention mechanism over the Bi-LSTM outputs to yield the final sentence representations for the classification. Experiment on the Stanford Natural Language Inference (SNLI) corpus suggests that this is a promising technique for RTE.

Keywords: deep neural models, natural language inference, recognizing textual entailment (RTE), sentence-to-sentence relation

Procedia PDF Downloads 324

Authors: Christoph Linse, Thomas Martinetz

Abstract:

Recent findings have shown that Neural Networks generalize also in over-parametrized regimes with zero training error. This is surprising, since it is completely against traditional machine learning wisdom. In our empirical study we fortify these findings in the domain of fine-grained image classification. We show that very large Convolutional Neural Networks with millions of weights do learn with only a handful of training samples and without image augmentation, explicit regularization or pretraining. We train the architectures ResNet018, ResNet101 and VGG19 on subsets of the difficult benchmark datasets Caltech101, CUB_200_2011, FGVCAircraft, Flowers102 and StanfordCars with 100 classes and more, perform a comprehensive comparative study and draw implications for the practical application of CNNs. Finally, we show that VGG19 with 140 million weights learns to distinguish airplanes and motorbikes with up to 95% accuracy using only 20 training samples per class.

Keywords: convolutional neural networks, fine-grained image classification, generalization, image recognition, over-parameterized, small data sets

Procedia PDF Downloads 58

Authors: Tri Sulistyaningtyas, Yani Suryani, Dana Waskita, Linda Handayani Sukaemi, Ferry Fauzi Hermawan

Abstract:

The enthusiasm of foreigners studying the Indonesian language by Foreign Speakers (BIPA) was documented in a sitcom "International Class". Tone and stress when they speak the Indonesian language is unique and different from Indonesian pronunciation. By using the Praat program, this research aims to describe prosodic Indonesian language which is spoken by ‘International Class” actors consisting of Abbas from Nigeria, Lee from Korea, and Kotaro from Japan. Data for the research are taken from the video sitcom "International Class" that aired on Indonesian television. The results of this study revealed that pitch movement that arises when pronouncing Indonesian sentences was up and down gradually, there is also a rise and fall sharply. In terms of stress, respondents tend to contain a lot of stress when pronouncing Indonesian sentences. Meanwhile, in terms of temporal structure, the duration pronouncing Indonesian sentences tends to be longer than that of Indonesian speakers.

Keywords: melodic structure, temporal structure, prosody, experimental phonetics, international class

Procedia PDF Downloads 278

Authors: Aadhya Kaul, Gautam Manocha, Preeti Nagrath

Abstract:

Our environment comprises of a wide variety of plants that are similar to each other and sometimes the similarity between the plants makes the identification process tedious thus increasing the workload of the botanist all over the world. Now all the botanists cannot be accessible all the time for such laborious plant identification; therefore, there is an urge for a quick classification model. Also, along with the identification of the plants, it is also necessary to classify the plant as healthy or not as for a good lifestyle, humans require good food and this food comes from healthy plants. A large number of techniques have been applied to classify the plants as healthy or diseased in order to provide the solution. This paper proposes one such method known as anomaly detection using autoencoders using a set of collections of leaves. In this method, an autoencoder model is built using Keras and then the reconstruction of the original images of the leaves is done and the threshold loss is found in order to classify the plant leaves as healthy or diseased. A dataset of plant leaves is considered to judge the reconstructed performance by convolutional autoencoders and the average accuracy obtained is 71.55% for the purpose.

Keywords: convolutional autoencoder, anomaly detection, web application, FLASK

Procedia PDF Downloads 136

Authors: Ali Ben Abbes, Imed Riadh Farah

Abstract:

Due to the considerable advances in Multi-Temporal Satellite Images (MTSI), remote sensing application became more accurate. Recently, many advances in modeling MTSI are developed using various models. The purpose of this article is to present an overview of studies using Hidden Markov Model (HMM). First of all, we provide a background of using HMM and their applications in this context. A comparison of the different works is discussed, and possible areas and challenges are highlighted. Secondly, we discussed the difference on vegetation monitoring as well as urban growth. Nevertheless, most research efforts have been used only stationary data. From another point of view, in this paper, we describe a new non-stationarity HMM, that is defined with a set of parts of the time series e.g. seasonal, trend and random. In addition, a new approach giving more accurate results and improve the applicability of the HMM in modeling a non-stationary data series. In order to assess the performance of the HMM, different experiments are carried out using Moderate Resolution Imaging Spectroradiometer (MODIS) NDVI time series of the northwestern region of Tunisia and Landsat time series of tres Cantos-Madrid in Spain.

Keywords: multi-temporal satellite image, HMM , nonstationarity, vegetation, urban

Procedia PDF Downloads 321

Authors: Amal Khalifa, Nicolas Vana Santos

Abstract:

Steganography has been known for centuries as an efficient approach for covert communication. Due to its popularity and ease of access, image steganography has attracted researchers to find secure techniques for hiding information within an innocent looking cover image. In this research, we propose a novel deep-learning approach to digital image steganography. The proposed method, DeepWaveletFusion, uses convolutional neural networks (CNN) to hide a secret image into a cover image of the same size. Two CNNs are trained back-to-back to merge the Discrete Wavelet Transform (DWT) of both colored images and eventually be able to blindly extract the hidden image. Based on two different image similarity metrics, a weighted gain function is used to guide the learning process and maximize the quality of the retrieved secret image and yet maintaining acceptable imperceptibility. Experimental results verified the high recoverability of DeepWaveletFusion which outperformed similar deep-learning-based methods.

Keywords: deep learning, steganography, image, discrete wavelet transform, fusion

Procedia PDF Downloads 44

Authors: María L. Jalón, Feargal Brennan

Abstract:

A non-stationary stochastic optimization methodology is applied to an OWC (oscillating water column) to find the design that maximizes the wave energy extraction. Different temporal cycles are considered to represent the long-term variability of the wave climate at the site in the optimization problem. The results of the non-stationary stochastic optimization problem are compared against those obtained by a stationary stochastic optimization problem. The comparative analysis reveals that the proposed non-stationary optimization provides designs with a better fit to reality. However, the stationarity assumption can be adequate when looking at averaged system response.

Keywords: non-stationary stochastic optimization, oscillating water, temporal variability, wave energy

Procedia PDF Downloads 337

Authors: Allyne Topaz

Abstract:

Intraoral dehiscence compromises free fibula flaps following mandibular reconstruction. Salivary contamination risks thrombosis of microvascular anastomosis and hardware infection. The superficial temporal artery islandized flap (STAIF) offers an efficient, non-microsurgical reconstructive option for regaining intraoral competency for a time sensitive complication. Methods: The STAIF flap is based on the superficial temporal artery coursing along the anterior hairline. The flap is mapped with assistance of the doppler probe. The width of the skin paddle is taken based on the ability to close the donor site. The flap is taken down to the level of the zygomatic arch and tunneled into the mouth. Results: We present a case of a patient who underwent mandibular reconstruction with a free fibula flap after a traumatic shotgun wound. The patient developed repeated intraoral dehiscence following failed local buccal and floor of mouth flaps leading to salivary contamination of the flap and hardware. The intraoral dehiscence was successfully salvaged on the third attempt with a STAIF flap. Conclusions: Intraoral dehiscence creates a complication requiring urgent attention to prevent loss of free fibula flap after mandibular reconstruction. The STAIF is a non-microsurgical option for restoring intraoral competency. This robust, axially vascularized skin paddle may be split for intra- and extra-oral coverage, as needed and can be an important tool in the reconstructive armamentarium.

Keywords: free fibula flap, intraoral dehiscence, mandibular reconstruction, superficial temporal artery islandized flap

Procedia PDF Downloads 108

Authors: Ammarah Irum, Muhammad Ali Tahir

Abstract:

Document-level sentiment analysis in Urdu is a challenging Natural Language Processing (NLP) task due to the difficulty of working with lengthy texts in a language with constrained resources. Deep learning models, which are complex neural network architectures, are well-suited to text-based applications in addition to data formats like audio, image, and video. To investigate the potential of deep learning for Urdu sentiment analysis, we implemented five different deep learning models, including Bidirectional Long Short Term Memory (BiLSTM), Convolutional Neural Network (CNN), Convolutional Neural Network with Bidirectional Long Short Term Memory (CNN-BiLSTM), and Bidirectional Encoder Representation from Transformer (BERT). In this study, we developed a hybrid deep learning model called BiLSTM-Single Layer Multi Filter Convolutional Neural Network (BiLSTM-SLMFCNN) by fusing BiLSTM and CNN architecture. The proposed and baseline techniques are applied on Urdu Customer Support data set and IMDB Urdu movie review data set by using pre-trained Urdu word embedding that are suitable for sentiment analysis at the document level. Results of these techniques are evaluated and our proposed model outperforms all other deep learning techniques for Urdu sentiment analysis. BiLSTM-SLMFCNN outperformed the baseline deep learning models and achieved 83%, 79%, 83% and 94% accuracy on small, medium and large sized IMDB Urdu movie review data set and Urdu Customer Support data set respectively.

Keywords: urdu sentiment analysis, deep learning, natural language processing, opinion mining, low-resource language

Procedia PDF Downloads 37

Authors: Ying Su, Morgan C. Wang

Abstract:

Long-term time series forecasting is an important research area for automated machine learning (AutoML). Currently, forecasting based on either machine learning or statistical learning is usually built by experts, and it requires significant manual effort, from model construction, feature engineering, and hyper-parameter tuning to the construction of the time series model. Automation is not possible since there are too many human interventions. To overcome these limitations, this article proposed to use recurrent neural networks (RNN) through the memory state of RNN to perform long-term time series prediction. We have shown that this proposed approach is better than the traditional Autoregressive Integrated Moving Average (ARIMA). In addition, we also found it is better than other network systems, including Fully Connected Neural Networks (FNN), Convolutional Neural Networks (CNN), and Nonpooling Convolutional Neural Networks (NPCNN).

Keywords: automated machines learning, autoregressive integrated moving average, neural networks, time series analysis

Procedia PDF Downloads 48

Authors: Yuta Chishima, Tatsuya Murakami, Michael McKay

Abstract:

Temporal focus is described as one component of an individual’s time perspective and defined as the attention individuals devote to thinking about the past, present, and future. It affects how people incorporate perceptions about past experiences, current situations, and future expectations into their attitudes, cognitions, and behavior. The 12-item Temporal Focus Scale (TFS) is comprised of three-factors (past, current and future focus). The purpose of this study was to examine the reliability and validity of TFS scores in Japanese adolescents and young adults. The TFS was translated into Japanese by a professional translator, and the original author confirmed the back translated items. Study 1 involved 979 Japanese university students aged 18-25 years old in a questionnaire-based study. The hypothesized three-factor structure (with reliability) was confirmed, although there were problems with item 10. Internal consistency estimates for scores without item 10 were over .70, and test-retest reliability was also adequate. To verify the concurrent and convergent validity, we tested the relationship between TFS scores and life satisfaction, time perspective, self-esteem, and career efficacy. Results of correlational analyses supported our hypotheses. Specifically, future focus was strongly correlated to career efficacy, while past and current focus was not. Study 2 involved 1030 Japanese junior and junior high school students aged 12-18 years old in a questionnaire-based study, and results of multigroup analyses supported the age invariance of the TFS.

Keywords: Japanese, reliability, scale, temporal focus, validity

Procedia PDF Downloads 323

Authors: Tesnim Charrad, Kaouther Nouira, Ahmed Ferchichi

Abstract:

In order to reduce the number of deaths due to heart problems, we propose the use of Hierarchical Temporal Memory Algorithm (HTM) which is a real time anomaly detection algorithm. HTM is a cortical learning algorithm based on neocortex used for anomaly detection. In other words, it is based on a conceptual theory of how the human brain can work. It is powerful in predicting unusual patterns, anomaly detection and classification. In this paper, HTM have been implemented and tested on ECG datasets in order to detect cardiac anomalies. Experiments showed good performance in terms of specificity, sensitivity and execution time.

Keywords: cardiac anomalies, ECG, HTM, real time anomaly detection

Procedia PDF Downloads 189

Authors: Rashini Maduka, C. R. Wijesinghe, A. R. Weerasinghe

Abstract:

Drug-drug interactions (DDIs) can happen when two or more drugs are taken together. Today DDIs have become a serious health issue due to adverse drug effects. In vivo and in vitro methods for identifying DDIs are time-consuming and costly. Therefore, in-silico-based approaches are preferred in DDI identification. Most machine learning models for DDI prediction are used chemical and biological drug properties as features. However, some drug features are not available and costly to extract. Therefore, it is better to make automatic feature engineering. Furthermore, people who have diabetes already suffer from other diseases and take more than one medicine together. Then adverse drug effects may happen to diabetic patients and cause unpleasant reactions in the body. In this study, we present a model with a graph convolutional autoencoder and a graph decoder using a dataset from DrugBank version 5.1.3. The main objective of the model is to identify unknown interactions between antidiabetic drugs and the drugs taken by diabetic patients for other diseases. We considered automatic feature engineering and used Known DDIs only as the input for the model. Our model has achieved 0.86 in AUC and 0.86 in AP.

Keywords: drug-drug interaction prediction, graph embedding, graph convolutional networks, adverse drug effects

Procedia PDF Downloads 69

Authors: P. Karthick, K. Mahesh

Abstract:

Video has become an increasingly significant component of our digital everyday contact. With the advancement of greater contents and shows of the resolution, its significant volume poses serious obstacles to the objective of receiving, distributing, compressing, and revealing video content of high quality. In this paper, we propose the primary beginning to complete a deep video compression model that jointly upgrades all video compression components. The video compression method involves splitting the video into frames, comparing the images using convolutional neural networks (CNN) to remove duplicates, repeating the single image instead of the duplicate images by recognizing and detecting minute changes using generative adversarial network (GAN) and recorded with long short-term memory (LSTM). Instead of the complete image, the small changes generated using GAN are substituted, which helps in frame level compression. Pixel wise comparison is performed using K-nearest neighbours (KNN) over the frame, clustered with K-means, and singular value decomposition (SVD) is applied for each and every frame in the video for all three color channels [Red, Green, Blue] to decrease the dimension of the utility matrix [R, G, B] by extracting its latent factors. Video frames are packed with parameters with the aid of a codec and converted to video format, and the results are compared with the original video. Repeated experiments on several videos with different sizes, duration, frames per second (FPS), and quality results demonstrate a significant resampling rate. On average, the result produced had approximately a 10% deviation in quality and more than 50% in size when compared with the original video.

Keywords: video compression, K-means clustering, convolutional neural network, generative adversarial network, singular value decomposition, pixel visualization, stochastic gradient descent, frame per second extraction, RGB channel extraction, self-detection and deciding system

Procedia PDF Downloads 161

Authors: Shibo Wei, Ting Jiang

Abstract:

Convolutional-recurrent neural networks (CRN) have achieved much success recently in the speech enhancement field. The common processing method is to use the convolution layer to compress the feature space by multiple upsampling and then model the compressed features with the LSTM layer. At last, the enhanced speech is obtained by deconvolution operation to integrate the global information of the speech sequence. However, the feature space compression process may cause the loss of information, so we propose to model the upsampling result of each step with the residual LSTM layer, then join it with the output of the deconvolution layer and input them to the next deconvolution layer, by this way, we want to integrate the global information of speech sequence better. The experimental results show the network model (RES-CRN) we introduce can achieve better performance than LSTM without residual and overlaying LSTM simply in the original CRN in terms of scale-invariant signal-to-distortion ratio (SI-SNR), speech quality (PESQ), and intelligibility (STOI).

Keywords: convolutional-recurrent neural networks, speech enhancement, residual LSTM, SI-SNR

Procedia PDF Downloads 171

Authors: M. Bilal Ishfaq, Adnan N. Qureshi

Abstract:

COVID-19 is the most infectious disease these days, it was first reported in Wuhan, the capital city of Hubei in China then it spread rapidly throughout the whole world. Later on 11 March 2020, the World Health Organisation (WHO) declared it a pandemic. Since COVID-19 is highly contagious, it has affected approximately 219M people worldwide and caused 4.55M deaths. It has brought the importance of accurate diagnosis of respiratory diseases such as pneumonia and COVID-19 to the forefront. In this paper, we propose a hybrid approach for the automated detection of COVID-19 using medical imaging. We have presented the hybridization of manually extracted and convolutional features. Our approach combines Haralick texture features and convolutional features extracted from chest X-rays and CT scans. We also employ a minimum redundancy maximum relevance (MRMR) feature selection algorithm to reduce computational complexity and enhance classification performance. The proposed model is evaluated on four publicly available datasets, including Chest X-ray Pneumonia, COVID-19 Pneumonia, COVID-19 CTMaster, and VinBig data. The results demonstrate high accuracy and effectiveness, with 0.9925 on the Chest X-ray pneumonia dataset, 0.9895 on the COVID-19, Pneumonia and Normal Chest X-ray dataset, 0.9806 on the Covid CTMaster dataset, and 0.9398 on the VinBig dataset. We further evaluate the effectiveness of the proposed model using ROC curves, where the AUC for the best-performing model reaches 0.96. Our proposed model provides a promising tool for the early detection and accurate diagnosis of COVID-19, which can assist healthcare professionals in making informed treatment decisions and improving patient outcomes. The results of the proposed model are quite plausible and the system can be deployed in a clinical or research setting to assist in the diagnosis of COVID-19.

Keywords: COVID-19, feature engineering, artificial neural networks, radiology images

Procedia PDF Downloads 45

Authors: Umair Rashid

Abstract:

Normally softmax loss is used as the supervision signal in face recognition (FR) system, and it boosts the separability of features. In the last two years, a number of techniques have been proposed by reformulating the original softmax loss to enhance the discriminating power of Deep Convolutional Neural Networks (DCNNs) for FR system. To learn angularly discriminative features Cosine-Margin based softmax has been adjusted as monotonically decreasing angular function, that is the main challenge for angular based softmax. On that issue, we propose monotonically decreasing element for Cosine-Margin based softmax and also, we discussed the effect of different monotonically decreasing parameters on angular Margin softmax for FR system. We train the model on publicly available dataset CASIA- WebFace via our proposed monotonically decreasing parameters for cosine function and the tests on YouTube Faces (YTF, Labeled Face in the Wild (LFW), VGGFace1 and VGGFace2 attain the state-of-the-art performance.

Keywords: deep convolutional neural networks, cosine margin face recognition, softmax loss, monotonically decreasing parameter

Procedia PDF Downloads 68

Authors: Necati Içer

Abstract:

Crop-hail insurance plays a vital role in managing risks and reducing the financial consequences of hail damage on crop production. Predicting insurance premium rates with short-term data is a major difficulty in numerous nations because of the unique characteristics of hailstorms. This study aims to suggest a feasible approach for establishing equitable premium rates in crop-hail insurance for nations with short-term insurance data. The primary goal of the rate-making process is to determine premium rates for high and zero loss costs of villages and enhance their credibility. To do this, a technique was created using the author's practical knowledge of crop-hail insurance. With this approach, the rate-making method was developed using a range of temporal and spatial factor combinations with both hypothetical and real data, including extreme cases. This article aims to show how to incorporate the temporal and spatial elements into determining fair premium rates using short-term insurance data. The article ends with a suggestion on the ultimate premium rates for insurance contracts.

Keywords: crop-hail insurance, premium rate, short-term insurance data, spatial and temporal parameters

Procedia PDF Downloads 18

Authors: Tiago do Carmo Nogueira, Cássio Dener Noronha Vinhal, Gélson da Cruz Júnior, Matheus Rudolfo Diedrich Ullmann

Abstract:

Generating automatic image descriptions through natural language is a challenging task. Image captioning is a task that consistently describes an image by combining computer vision and natural language processing techniques. To accomplish this task, cutting-edge models use encoder-decoder structures. Thus, Convolutional Neural Networks (CNN) are used to extract the characteristics of the images, and Recurrent Neural Networks (RNN) generate the descriptive sentences of the images. However, cutting-edge approaches still suffer from problems of generating incorrect captions and accumulating errors in the decoders. To solve this problem, we propose a model based on the encoder-decoder structure, introducing a module that generates the weights according to the importance of the word to form the sentence, using the part-of-speech (PoS). Thus, the results demonstrate that our model surpasses state-of-the-art models.

Keywords: gated recurrent units, caption generation, convolutional neural network, part-of-speech

Procedia PDF Downloads 70