Search results for: engineering dataset

Authors: Salma El Hajjami, Jamal Malki, Alain Bouju, Mohammed Berrada

Abstract:

With the expansion of machine learning and data mining in the context of Big Data analytics, the common problem that affects data is class imbalance. It refers to an imbalanced distribution of instances belonging to each class. This problem is present in many real world applications such as fraud detection, network intrusion detection, medical diagnostics, etc. In these cases, data instances labeled negatively are significantly more numerous than the instances labeled positively. When this difference is too large, the learning system may face difficulty when tackling this problem, since it is initially designed to work in relatively balanced class distribution scenarios. Another important problem, which usually accompanies these imbalanced data, is the overlapping instances between the two classes. It is commonly referred to as noise or overlapping data. In this article, we propose an approach called: One Side Behavioral Noise Reduction (OSBNR). This approach presents a way to deal with the problem of class imbalance in the presence of a high noise level. OSBNR is based on two steps. Firstly, a cluster analysis is applied to groups similar instances from the minority class into several behavior clusters. Secondly, we select and eliminate the instances of the majority class, considered as behavioral noise, which overlap with behavior clusters of the minority class. The results of experiments carried out on a representative public dataset confirm that the proposed approach is efficient for the treatment of class imbalances in the presence of noise.

Keywords: machine learning, imbalanced data, data mining, big data

Procedia PDF Downloads 134

Authors: Daniel Pinheiro Da Silva Junior, Aline Paes, Daniel De Oliveira, Christiano Lacerda Ghuerren, Marcio Duran

Abstract:

The Courts of Accounts are institutions responsible for overseeing and point out irregularities of Public Administration expenses. They have a high demand for processes to be analyzed, whose decisions must be grounded on severity laws. Despite the existing large amount of processes, there are several cases reporting similar subjects. Thus, previous decisions on already analyzed processes can be a precedent for current processes that refer to similar topics. Identifying similar topics is an open, yet essential task for identifying similarities between several processes. Since the actual amount of topics is considerably large, it is tedious and error-prone to identify topics using a pure manual approach. This paper presents a tool based on Machine Learning and Natural Language Processing to assists in building a labeled dataset. The tool relies on Topic Modelling with Latent Dirichlet Allocation to find the topics underlying a document followed by Jensen Shannon distance metric to generate a probability of similarity between documents pairs. Furthermore, in a case study with a corpus of decisions of the Rio de Janeiro State Court of Accounts, it was noted that data pre-processing plays an essential role in modeling relevant topics. Also, the combination of topic modeling and a calculated distance metric over document represented among generated topics has been proved useful in helping to construct a labeled base of similar and non-similar document pairs.

Keywords: courts of accounts, data labelling, document similarity, topic modeling

Procedia PDF Downloads 184

Authors: Kshitij Ansingkar, Yohei Hoshino, Liangliang Yang

Abstract:

Advances in algorithms related to autonomous systems have made it possible to research on improving the accuracy of a train’s location. This has the capability of increasing the throughput of a railway network without the need for the creation of additional infrastructure. To develop such a system, the railway industry requires data to test sensor fusion theories or implement simultaneous localization and mapping (SLAM) algorithms. Though such simulation data and ground truth datasets are available for testing automation algorithms of vehicles, however, due to regulations and economic considerations, there is a dearth of such datasets in the railway industry. Thus, there is a need for the creation of a simulation environment that can generate realistic synthetic datasets. This paper proposes (1) to leverage the capabilities of open-source 3D graphic rendering software to create a visualization of the environment. (2) to utilize open-source 3D geospatial data for accurate visualization and (3) to integrate the graphic rendering software with a programming language and numerical computing platform. To develop such an integrated platform, this paper utilizes the computing platform’s advanced sensor models like LIDAR, camera, IMU or GPS and merges it with the 3D rendering of the game engine to generate high-quality synthetic data. Further, these datasets can be used to train Railway models and improve the accuracy of a train’s location.

Keywords: 3D game engine, 3D geospatial data, dataset generation, railway simulator, sensor fusion, SLAM

Procedia PDF Downloads 18

Authors: Md. Mahadi Hasan Sany, Mumenunnessa Keya, Sharun Khushbu, Sheikh Abujar

Abstract:

Since the beginning of China's economic reform, trade between the U.S. and China has grown rapidly, and has increased since China's accession to the World Trade Organization in 2001. The US imports more than it exports from China, reducing the trade war between China and the U.S. for the 2019 trade deficit, but in 2020, the opposite happens. In international and U.S. trade, Washington launched a full-scale trade war against China in March 2016, which occurred a catastrophic epidemic. The main goal of our study is to measure and predict trade relations between China and the U.S., before and after the arrival of the COVID epidemic. The ML model uses different data as input but has no time dimension that is present in the time series models and is only able to predict the future from previously observed data. The LSTM (a well-known Recurrent Neural Network) model is applied as the best time series model for trading forecasting. We have been able to create a sustainable forecasting system in trade between China and the US by closely monitoring a dataset published by the State Website NZ Tatauranga Aotearoa from January 1, 2015, to April 30, 2021. Throughout the survey, we provided a 180-day forecast that outlined what would happen to trade between China and the US during COVID-19. In addition, we have illustrated that the LSTM model provides outstanding outcome in time series data analysis rather than RFR and SVR (e.g., both ML models). The study looks at how the current Covid outbreak affects China-US trade. As a comparative study, RMSE transmission rate is calculated for LSTM, RFR and SVR. From our time series analysis, it can be said that the LSTM model has given very favorable thoughts in terms of China-US trade on the future export situation.

Keywords: RFR, China-U.S. trade war, SVR, LSTM, deep learning, Covid-19, export value, forecasting, time series analysis

Procedia PDF Downloads 202

Authors: Jianfeng Hu

Abstract:

Personal authentication based on electroencephalography (EEG) signals is one of the important field for the biometric technology. More and more researchers have used EEG signals as data source for biometric. However, there are some disadvantages for biometrics based on EEG signals. The proposed method employs entropy measures for feature extraction from EEG signals. Four type of entropies measures, sample entropy (SE), fuzzy entropy (FE), approximate entropy (AE) and spectral entropy (PE), were deployed as feature set. In a silhouettes calculation, the distance from each data point in a cluster to all another point within the same cluster and to all other data points in the closest cluster are determined. Thus silhouettes provide a measure of how well a data point was classified when it was assigned to a cluster and the separation between them. This feature renders silhouettes potentially well suited for assessing cluster quality in personal authentication methods. In this study, “silhouettes scores” was used for assessing the cluster quality of k-means clustering algorithm is well suited for comparing the performance of each EEG dataset. The main goals of this study are: (1) to represent each target as a tuple of multiple feature sets, (2) to assign a suitable measure to each feature set, (3) to combine different feature sets, (4) to determine the optimal feature weighting. Using precision/recall evaluations, the effectiveness of feature weighting in clustering was analyzed. EEG data from 22 subjects were collected. Results showed that: (1) It is possible to use fewer electrodes (3-4) for personal authentication. (2) There was the difference between each electrode for personal authentication (p<0.01). (3) There is no significant difference for authentication performance among feature sets (except feature PE). Conclusion: The combination of k-means clustering algorithm and silhouette approach proved to be an accurate method for personal authentication based on EEG signals.

Keywords: personal authentication, K-mean clustering, electroencephalogram, EEG, silhouettes

Procedia PDF Downloads 289

Authors: Derejaw Lake Melie, Alemu Kumlachew Tegegne

Abstract:

The increasing prevalence of social media in Ethiopia has exacerbated societal challenges by fostering the proliferation of negative emotional posts and comments. Illicit use of social media has further exacerbated divisions among the population. Addressing these issues through manual identification and aggregation of emotions from millions of users for swift decision-making poses significant challenges, particularly given the rapid growth of Amharic language usage on social platforms. Consequently, there is a critical need to develop an intelligent system capable of automatically detecting and categorizing negative emotional content into social, religious, and political categories while also filtering out toxic online content. This paper aims to leverage sentiment analysis techniques to achieve automatic detection and filtering of negative emotional content from Amharic social media texts, employing a comparative study of deep learning algorithms. The study utilized a dataset comprising 29,962 comments collected from social media platforms using comment exporter software. Data pre-processing techniques were applied to enhance data quality, followed by the implementation of deep learning methods for training, testing, and evaluation. The results showed that CNN, GRU, LSTM, and Bi-LSTM classification models achieved accuracies of 83%, 50%, 84%, and 86%, respectively. Among these models, Bi-LSTM demonstrated the highest accuracy of 86% in the experiment.

Keywords: negative emotion, emotion detection, social media filtering sentiment analysis, deep learning.

Procedia PDF Downloads 40

Authors: Haidy S. Ghali, Engy Serag, A. Samer Ezeldin

Abstract:

Copper is an essential raw material used in the construction industry. During the year 2021 and the first half of 2022, the global market suffered from a significant fluctuation in copper raw material prices due to the aftermath of both the COVID-19 pandemic and the Russia-Ukraine war, which exposed its consumers to an unexpected financial risk. Thereto, this paper aims to develop two ANN-LSTM price prediction models, using Python, that can forecast the average monthly copper prices traded in the London Metal Exchange; the first model is a multivariate model that forecasts the copper price of the next 1-month and the second is a univariate model that predicts the copper prices of the upcoming three months. Historical data of average monthly London Metal Exchange copper prices are collected from January 2009 till July 2022, and potential external factors are identified and employed in the multivariate model. These factors lie under three main categories: energy prices and economic indicators of the three major exporting countries of copper, depending on the data availability. Before developing the LSTM models, the collected external parameters are analyzed with respect to the copper prices using correlation and multicollinearity tests in R software; then, the parameters are further screened to select the parameters that influence the copper prices. Then, the two LSTM models are developed, and the dataset is divided into training, validation, and testing sets. The results show that the performance of the 3-Month prediction model is better than the 1-Month prediction model, but still, both models can act as predicting tools for diverse economic situations.

Keywords: copper prices, prediction model, neural network, time series forecasting

Procedia PDF Downloads 117

Authors: Aditya Patil, Arun Patil, Vaishali Patil, Sudhir Chitnis, Anjum Patel

Abstract:

Rapid growth has given e-commerce platforms a lot of client behavior and spending data. To maximize their strategy, businesses must understand how customers utilize online shopping platforms and what influences their purchases. Our research focuses on e-commerce user behavior and purchasing trends. This extensive study examines spending and user behavior. Regression and grouping disclose relevant data from the dataset. We can understand user spending trends via multilevel regression. We can analyze how pricing, user demographics, and product categories affect customer purchase decisions with this technique. Clustering groups consumers by spending. Important information was found. Purchase habits vary by user group. Our analysis illuminates the complex world of e-commerce consumer behavior and purchase trends. Understanding user behavior helps create effective e-commerce marketing strategies. This market can benefit from K-means clustering. This study focuses on tailoring strategies to user groups and improving product and price effectiveness. Customer buying behaviors across categories were shown via K-means clusters. Average spending is highest in Cluster 4 and lowest in Cluster 3. Clothing is less popular than gadgets and appliances around the holidays. Cluster spending distribution is examined using average variables. Our research enhances e-commerce analytics. Companies can improve customer service and decision-making with this data.

Keywords: e-commerce, regression, clustering, k-means

Procedia PDF Downloads 28

Authors: Talha A. Al-Dubai, Rashad A. Bantan, Ramadan H. Abu-Zied, Brian G. Jones, Aaid G. Al-Zubieri

Abstract:

Contemporary foraminiferal samples sediments were collected from the intertidal sabkha of Al-Kharrar Lagoon, Saudi Arabia, to study the vertical distribution of Foraminifera and, based on a modern training set, their potential to develop a predictor of former sea-level changes in the area. Based on hierarchical cluster analysis, the intertidal sabkha is divided into three vertical zones (A, B & C) represented by three foraminiferal assemblages, where agglutinated species occupied Zone A and calcareous species occupied the other two zones. In Zone A (high intertidal), Agglutinella compressa, Clavulina angularis and C. multicamerata are dominant species with a minor presence of Peneroplis planatus, Coscinospira hemprichii, Sorites orbiculus, Quinqueloculina lamarckiana, Q. seminula, Ammonia convexa and A. tepida. In contrast, in Zone B (middle intertidal) the most abundant species are P. planatus, C. hemprichii, S. orbiculus, Q. lamarckiana, Q. seminula and Q. laevigata, while Zone C (low intertidal) is characterised by C. hemprichii, Q. costata, S. orbiculus, P. planatus, A. convexa, A. tepida, Spiroloculina communis and S. costigera. A transfer function for sea-level reconstruction was developed using a modern dataset of 75 contemporary sediment samples and 99 species collected from several transects across the sabkha. The model provided an error of 0.12m, suggesting that intertidal foraminifers are able to predict the past sea-level changes with high precision in Al-Kharrar Lagoon, and thus the future prediction of those changes in the area.

Keywords: Lagoonal foraminifers, intertidal sabkha, vertical zonation, transfer function, sea level

Procedia PDF Downloads 174

Authors: Zsanett Bahor, Cristina Nunes-Fonseca, Gillian L. Currie, Emily S. Sena, Lindsay D.G. Thomson, Malcolm R. Macleod

Abstract:

Psychosis is ranked as the third most disabling medical condition in the world by the World Health Organization. Despite a substantial amount of research in recent years, available treatments are not universally effective and have a wide range of adverse side effects. Since many clinical drug candidates are identified through in vivo modelling, a deeper understanding of these models, and their strengths and limitations, might help us understand reasons for difficulties in psychosis drug development. To provide an unbiased summary of the preclinical psychosis literature we performed a systematic electronic search of PubMed for publications modelling a psychotic disorder in vivo, identifying 14,721 relevant studies. Double screening of 11,000 publications from this dataset so far established 2403 animal studies of psychosis, with the most common model being schizophrenia (95%). 61% of these models are induced using pharmacological agents. For all the models only 56% of publications test a therapeutic treatment. We propose a systematic review of these studies to assess the prevalence of reporting of measures to reduce risk of bias, and a meta-analysis to assess the internal and external validity of these animal models. Our findings are likely to be relevant to future preclinical studies of psychosis as this generation of strong empirical evidence has the potential to identify weaknesses, areas for improvement and make suggestions on refinement of experimental design. Such a detailed understanding of the data which inform what we think we know will help improve the current attrition rate between bench and bedside in psychosis research.

Keywords: animal models, psychosis, systematic review, schizophrenia

Procedia PDF Downloads 294

Authors: Kara Gianina D. Rosas, Jade Emily L. Tong

Abstract:

This study assesses the vulnerability of households to food price shocks. Using the Philippines as a case study, the researchers aim to understand how such shocks can cause food insecurity in different types of households. This paper measures the impact of actual food price changes during the food crisis of 2006-2009 on poverty in relation to their spatial location. Households are classified as rural or urban and agricultural or non-agricultural. By treating food prices and consumption patterns as heterogeneous, this study differs from conventional poverty analysis as actual prices are used. Merging the Family, Income and Expenditure Survey (FIES) with the Consumer Price Index dataset (CPI), the researchers were able to determine the effects on poverty measures, specifically, headcount index, poverty gap, and poverty severity. The study finds that, without other interventions, food inflation would lead to a significant increase in the number of households that fall below the poverty threshold, except for households whose income is derived from agricultural activities. It also finds that much of the inflation during these years was fueled by the rise in staple food prices. Essentially, this paper aims to broaden the economic perspective of policymakers with regard to the heterogeneity of impacts of inflation through analyzing the deeper microeconomic levels of different subgroups. In hopes of finding a solution to lessen the inequality gap of poverty between the rural and urban poor, this paper aims to aid policymakers in creating projects targeted towards food insecurity.

Keywords: poverty, food inflation, agricultural households, non-agricultural households, net consumption ratio, urban poor, rural poor, head count index, poverty gap, poverty severity

Procedia PDF Downloads 255

Authors: Yasuyo Makido, Vivek Shandas, David J. Sailor, M. Salim Ferwati

Abstract:

Mitigating extreme urban heat is challenging in a desert climate such as Doha, Qatar, since outdoor daytime temperature area often too high for the human body to tolerate. Recent studies demonstrate that cities in arid and semiarid areas can exhibit ‘urban cool islands’ - urban areas that are cooler than the surrounding desert. However, the variation of temperatures as a result of the time of day and factors leading to temperature change remain at the question. To address these questions, we examined the spatial and temporal variation of air temperature in Doha, Qatar by conducting multiple vehicle-base local temperature observations. We also employed three statistical approaches to model surface temperatures using relevant predictors: (1) Ordinary Least Squares, (2) Regression Tree Analysis and (3) Random Forest for three time periods. Although the most important determinant factors varied by day and time, distance to the coast was the significant determinant at midday. A 70%/30% holdout method was used to create a testing dataset to validate the results through Pearson’s correlation coefficient. The Pearson’s analysis suggests that the Random Forest model more accurately predicts the surface temperatures than the other methods. We conclude with recommendations about the types of development patterns that show the greatest potential for reducing extreme heat in air climates.

Keywords: desert cities, tree-structure regression model, urban cool Island, vehicle temperature traverse

Procedia PDF Downloads 395

Authors: Joseph George, Anne Kotteswara Roa

Abstract:

Skin disease is one of the most common and popular kinds of health issues faced by people nowadays. Skin cancer (SC) is one among them, and its detection relies on the skin biopsy outputs and the expertise of the doctors, but it consumes more time and some inaccurate results. At the early stage, skin cancer detection is a challenging task, and it easily spreads to the whole body and leads to an increase in the mortality rate. Skin cancer is curable when it is detected at an early stage. In order to classify correct and accurate skin cancer, the critical task is skin cancer identification and classification, and it is more based on the cancer disease features such as shape, size, color, symmetry and etc. More similar characteristics are present in many skin diseases; hence it makes it a challenging issue to select important features from a skin cancer dataset images. Hence, the skin cancer diagnostic accuracy is improved by requiring an automated skin cancer detection and classification framework; thereby, the human expert’s scarcity is handled. Recently, the deep learning techniques like Convolutional neural network (CNN), Deep belief neural network (DBN), Artificial neural network (ANN), Recurrent neural network (RNN), and Long and short term memory (LSTM) have been widely used for the identification and classification of skin cancers. This survey reviews different DL techniques for skin cancer identification and classification. The performance metrics such as precision, recall, accuracy, sensitivity, specificity, and F-measures are used to evaluate the effectiveness of SC identification using DL techniques. By using these DL techniques, the classification accuracy increases along with the mitigation of computational complexities and time consumption.

Keywords: skin cancer, deep learning, performance measures, accuracy, datasets

Procedia PDF Downloads 137

Authors: Jiangxia Han, Liang Xue, Keda Chen

Abstract:

Despite the significant progress over the last decades in reservoir simulation using numerical discretization, meshing is complex. Moreover, the high degree of freedom of the space-time flow field makes the solution process very time-consuming. Therefore, we present Physics-Informed Convolutional Neural Networks(PICNN) as a hybrid scientific theory and data method for reservoir modeling. Besides labeled data, the model is driven by the scientific theories of the underlying problem, such as governing equations, boundary conditions, and initial conditions. PICNN integrates governing equations and boundary conditions into the network architecture in the form of a customized convolution kernel. The loss function is composed of data matching, initial conditions, and other measurable prior knowledge. By customizing the convolution kernel and minimizing the loss function, the neural network parameters not only fit the data but also honor the governing equation. The PICNN provides a methodology to model and history-match flow and transport problems in porous media. Numerical results demonstrate that the proposed PICNN can provide an accurate physical solution from a limited dataset. We show how this method can be applied in the context of a forward simulation for continuous problems. Furthermore, several complex scenarios are tested, including the existence of data noise, different work schedules, and different good patterns.

Keywords: convolutional neural networks, deep learning, flow and transport in porous media, physics-informed neural networks, reservoir simulation

Procedia PDF Downloads 152

Authors: Ankit Sharma, Mudit Sachan, Anurag Prakash

Abstract:

Rapid growth in data acquisition technologies through drones, have led to advances and interests in collecting high-resolution images of geological fields. Being advantageous in capturing high volume of data in short flights, a number of challenges have to overcome for efficient analysis of this data, especially while data acquisition, image interpretation and processing. We introduce a method that allows effective mapping of geological fields using photogrammetric data of surfaces, drainage area, water bodies etc, which will be captured by airborne vehicles like UAVs, we are not taking satellite images because of problems in adequate resolution, time when it is captured may be 1 yr back, availability problem, difficult to capture exact image, then night vision etc. This method includes advanced automated image interpretation technology and human data interaction to model structures and. First Geological structures will be detected from the primary photographic dataset and the equivalent three dimensional structures would then be identified by digital elevation model. We can calculate dip and its direction by using the above information. The structural map will be generated by adopting a specified methodology starting from choosing the appropriate camera, camera’s mounting system, UAVs design ( based on the area and application), Challenge in air borne systems like Errors in image orientation, payload problem, mosaicing and geo referencing and registering of different images to applying DEM. The paper shows the potential of using our method for accurate and efficient modeling of geological structures, capture particularly from remote, of inaccessible and hazardous sites.

Keywords: digital elevation model, mapping, photogrammetric data analysis, geological structures

Procedia PDF Downloads 689

Authors: Andrea Treviño Gavito, Diego Klabjan, Sanjiv J. Shah

Abstract:

Echocardiograms serve as pivotal resources for clinicians in diagnosing cardiac conditions, offering non-invasive insights into a heart’s structure and function. When echocardiographic studies are conducted, no standardized labeling of the acquired views is performed. Employing machine learning algorithms for automated echocardiogram view detection has emerged as a promising solution to enhance efficiency in echocardiogram use for diagnosis. However, existing approaches predominantly rely on supervised learning, necessitating labor-intensive expert labeling. In this paper, we introduce a fully unsupervised echocardiographic view detection framework that leverages convolutional autoencoders to obtain lower dimensional representations and the K-means algorithm for clustering them into view-related groups. Our approach focuses on discriminative patches from echocardiographic frames. Additionally, we propose a trainable inverse average layer to optimize decoding of average operations. By integrating both public and proprietary datasets, we obtain a marked improvement in model performance when compared to utilizing a proprietary dataset alone. Our experiments show boosts of 15.5% in accuracy and 9.0% in the F-1 score for frame-based clustering, and 25.9% in accuracy and 19.8% in the F-1 score for view-based clustering. Our research highlights the potential of unsupervised learning methodologies and the utilization of open-sourced data in addressing the complexities of echocardiogram interpretation, paving the way for more accurate and efficient cardiac diagnoses.

Keywords: artificial intelligence, echocardiographic view detection, echocardiography, machine learning, self-supervised representation learning, unsupervised learning

Procedia PDF Downloads 46

Authors: Bhogendra Mishra, Andy Nelson, Mirco Boschetti, Lorenzo Busetto, Alice Laborte

Abstract:

Rice is grown on over 100 million hectares in almost every country of Asia. It is the most important staple crop for food security and has high economic and cultural importance in Asian societies. The combination of genetic diversity and management options, coupled with the large geographic extent means that there is a large variation in seasonality (when it is grown) and cropping intensity (how often it is grown per year on the same plot of land), even over relatively small distances. Seasonality and intensity can and do change over time depending on climatic, environmental and economic factors. Detecting where and when these changes happen can provide information to better understand trends in regional and even global rice production. Remote sensing offers a unique opportunity to estimate these trends. We apply the recently published PhenoRice algorithm to 14 years of moderate resolution remote sensing (MODIS) data (utilizing 250m resolution 16 day composites from Terra and Aqua) to estimate seasonality and cropping intensity per year and changes over time. We compare the results to the surveyed data collected by International Rice Research Institute (IRRI). The study results in a unique and validated dataset on the extent and change of extent, the seasonality and change in seasonality and the cropping intensity and change in cropping intensity between 2003 and 2016 for the Philippines. Observed trends and their implications for food security and trade policies are also discussed.

Keywords: rice, cropping intensity, moderate resolution remote sensing (MODIS), phenology, seasonality

Procedia PDF Downloads 314

Authors: Parisa Mansour

Abstract:

Cystic fibrosis (CF) is one of the most common autosomal recessive diseases among whites. It mostly affects the lungs, causing infections and inflammation that account for 90% of deaths in CF patients. Because of this high variability in clinical presentation and organ involvement, investigating treatment responses and evaluating lung changes over time is critical to preventing CF progression. High-resolution computed tomography (HRCT) greatly facilitates the assessment of lung disease progression in CF patients. Recently, artificial intelligence was used to analyze chest CT scans of CF patients. In this paper, we propose a convolutional neural network (CNN) approach to classify CF lung patterns in HRCT images. The proposed network consists of two convolutional layers with 3 × 3 kernels and maximally connected in each layer, followed by two dense layers with 1024 and 10 neurons, respectively. The softmax layer prepares a predicted output probability distribution between classes. This layer has three exits corresponding to the categories of normal (healthy), bronchitis and inflammation. To train and evaluate the network, we constructed a patch-based dataset extracted from more than 1100 lung HRCT slices obtained from 45 CF patients. Comparative evaluation showed the effectiveness of the proposed CNN compared to its close peers. Classification accuracy, average sensitivity and specificity of 93.64%, 93.47% and 96.61% were achieved, indicating the potential of CNNs in analyzing lung CF patterns and monitoring lung health. In addition, the visual features extracted by our proposed method can be useful for automatic measurement and finally evaluation of the severity of CF patterns in lung HRCT images.

Keywords: HRCT, CF, cystic fibrosis, chest CT, artificial intelligence

Procedia PDF Downloads 71

Authors: C. Xavier Mendieta, J. J McArthur

Abstract:

Governments are playing an increasingly active role in reducing carbon emissions, and a key strategy has been the introduction of mandatory energy disclosure policies. These policies have resulted in a significant amount of publicly available data, providing researchers with a unique opportunity to develop location-specific energy and carbon emission benchmarks from this data set, which can then be used to develop building archetypes and used to inform urban energy models. This study presents the development of such a benchmark using the public reporting data. The data from Ontario’s Ministry of Energy for Post-Secondary Educational Institutions are being used to develop a series of building archetype dynamic building loads and energy benchmarks to fill a gap in the currently available building database. This paper presents the development of a benchmark for college and university residences within ASHRAE climate zone 6 areas in Ontario using the mandatory disclosure energy and greenhouse gas emissions data. The methodology presented includes data cleaning, statistical analysis, and benchmark development, and lessons learned from this investigation are presented and discussed to inform the development of future energy benchmarks from this larger data set. The key findings from this initial benchmarking study are: (1) the importance of careful data screening and outlier identification to develop a valid dataset; (2) the key features used to develop a model of the data are building age, size, and occupancy schedules and these can be used to estimate energy consumption; and (3) policy changes affecting the primary energy generation significantly affected greenhouse gas emissions, and consideration of these factors was critical to evaluate the validity of the reported data.

Keywords: building archetypes, data analysis, energy benchmarks, GHG emissions

Procedia PDF Downloads 309

Authors: Ngoc-Hieu Vu

Abstract:

A study was conducted to estimate the record keeping, genetic selection, educational experience, and farm management effect on monthly milk yield per farm, average milk yield per cow, monthly milk revenue per farm, and monthly milk revenue per cow of dairy farms in the Southern region of Vietnam. The dataset contained 5448 monthly record collected from January 2013 to May 2015. Results showed that longer experience increased (P < 0.001) monthly milk yields and revenues. Better educated farmers produced more monthly milk per farm and monthly milk per cow and revenues (P < 0.001) than lower educated farmers. Farm that kept records on individual animals had higher (P < 0.001) for monthly milk yields and revenues than farms that did not. Farms that used hired people produced the highest (p < 0.05) monthly milk yield per farm, milk yield per cow and revenues, followed by farms that used both hire and family members, and lowest values were for farms that used family members only. Farms that used crosses Holstein in herd were higher performance (p < 0.001) for all traits than farms that used purebred Holstein and other breeds. Farms that used genetic information and phenotypes when selecting sires were higher (p < 0.05) for all traits than farms that used only phenotypes and personal option. Farms that received help from Vet, organization staff, or government officials had higher monthly milk yield and revenues than those that decided by owner. These findings suggest that dairy farmers should be training in systematic, must be considered and continuous support to improve farm milk production and revenues, to increase the likelihood of adoption on a sustainable way.

Keywords: dairy farming, education, milk yield, Southern Vietnam

Procedia PDF Downloads 334

Authors: Bliss Singhal

Abstract:

Machine learning (ML) is a branch of Artificial Intelligence (AI) where computers analyze data and find patterns in the data. The study focuses on the detection of metastatic cancer using ML. Metastatic cancer is the stage where cancer has spread to other parts of the body and is the cause of approximately 90% of cancer-related deaths. Normally, pathologists spend hours each day to manually classifying whether tumors are benign or malignant. This tedious task contributes to mislabeling metastasis being over 60% of the time and emphasizes the importance of being aware of human error and other inefficiencies. ML is a good candidate to improve the correct identification of metastatic cancer, saving thousands of lives and can also improve the speed and efficiency of the process, thereby taking fewer resources and time. So far, the deep learning methodology of AI has been used in research to detect cancer. This study is a novel approach to determining the potential of using preprocessing algorithms combined with classification algorithms in detecting metastatic cancer. The study used two preprocessing algorithms: principal component analysis (PCA) and the genetic algorithm, to reduce the dimensionality of the dataset and then used three classification algorithms: logistic regression, decision tree classifier, and k-nearest neighbors to detect metastatic cancer in the pathology scans. The highest accuracy of 71.14% was produced by the ML pipeline comprising of PCA, the genetic algorithm, and the k-nearest neighbor algorithm, suggesting that preprocessing and classification algorithms have great potential for detecting metastatic cancer.

Keywords: breast cancer, principal component analysis, genetic algorithm, k-nearest neighbors, decision tree classifier, logistic regression

Procedia PDF Downloads 91

Authors: Mohammad R. Kavian Nezhad, Carlos F. Lange, Brian A. Fleck

Abstract:

This research presents the validation study of a computational fluid dynamics (CFD) model developed to simulate the scalar dispersion emitted from rooftop sources around the buildings at the University of Alberta North Campus. The ANSYS CFX code was used to perform the numerical simulation of the wind regime and pollutant dispersion by solving the 3D steady Reynolds-averaged Navier-Stokes (RANS) equations on a building-scale high-resolution grid. The validation study was performed in two steps. First, the CFD model performance in 24 cases (eight wind directions and three wind speeds) was evaluated by comparing the predicted flow fields with the available data from the previous measurement campaign designed at the North Campus, using the standard deviation method (SDM), while the estimated results of the numerical model showed maximum average percent errors of approximately 53% and 37% for wind incidents from the North and Northwest, respectively. Good agreement with the measurements was observed for the other six directions, with an average error of less than 30%. In the second step, the reliability of the implemented turbulence model, numerical algorithm, modeling techniques, and the grid generation scheme was further evaluated using the Mock Urban Setting Test (MUST) dispersion dataset. Different statistical measures, including the fractional bias (FB), the geometric mean bias (MG), and the normalized mean square error (NMSE), were used to assess the accuracy of the predicted dispersion field. Our CFD results are in very good agreement with the field measurements.

Keywords: CFD, plume dispersion, complex urban geometry, validation study, wind flow

Procedia PDF Downloads 141

Authors: Ping Tan, Xiaomeng Su, Yi Shen

Abstract:

The motion intention in the motor imagery braincomputer interface is identified by classifying the event-related desynchronization (ERD) and event-related synchronization ERS characteristics of sensorimotor rhythm (SMR) in EEG signals. When the subject imagines different limbs or different parts moving, the rhythm components and bandwidth will change, which varies from person to person. How to find the effective sensorimotor frequency band of subjects is directly related to the classification accuracy of brain-computer interface. To solve this problem, this paper proposes a Minimum Distance to Riemannian Mean Classification method based on Non-Uniform Filter Banks. During the training phase, the EEG signals are decomposed into multiple different bandwidt signals by using multiple band-pass filters firstly; Then the spatial covariance characteristics of each frequency band signal are computered to be as the feature vectors. these feature vectors will be classified by the MDRM (Minimum Distance to Riemannian Mean) method, and cross validation is employed to obtain the effective sensorimotor frequency bands. During the test phase, the test signals are filtered by the bandpass filter of the effective sensorimotor frequency bands, and the extracted spatial covariance feature vectors will be classified by using the MDRM. Experiments on the BCI competition IV 2a dataset show that the proposed method is superior to other classification methods.

Keywords: non-uniform filter banks, motor imagery, brain-computer interface, minimum distance to Riemannian mean

Procedia PDF Downloads 130

Authors: Rahman Ali, Muhammad Sajjad, Farkhund Iqbal, Muhammad Sadiq Hassan Zada, Mohammed Hussain

Abstract:

Emission of Carbon Dioxide (CO₂) has adversely affected the environment. One of the major sources of CO₂ emission is transportation. In the last few decades, the increase in mobility of people using vehicles has enormously increased the emission of CO₂ in the environment. To reduce CO₂ emission, sustainable transportation system is required in which smart parking is one of the important measures that need to be established. To contribute to the issue of reducing the amount of CO₂ emission, this research proposes a smart parking system. A cloud-based solution is provided to the drivers which automatically searches and recommends the most preferred parking slots. To determine preferences of the parking areas, this methodology exploits a number of unique parking features which ultimately results in the selection of a parking that leads to minimum level of CO₂ emission from the current position of the vehicle. To realize the methodology, a scenario-based implementation is considered. During the implementation, a mobile application with GPS signals, vehicles with a number of vehicle features and a list of parking areas with parking features are used by sorting, multi-level filtering, exploratory data analysis (EDA, Analytical Hierarchy Process (AHP)) and weighted sum model (WSM) to rank the parking areas and recommend the drivers with top-k most preferred parking areas. In the EDA process, “2020testcar-2020-03-03”, a freely available dataset is used to estimate CO₂ emission of a particular vehicle. To evaluate the system, results of the proposed system are compared with the conventional approach, which reveal that the proposed methodology supersedes the conventional one in reducing the emission of CO₂ into the atmosphere.

Keywords: car parking, Co2, Co2 reduction, IoT, merge sort, number plate recognition, smart car parking

Procedia PDF Downloads 149

Authors: Daohong Yang, Xiang Zhang, Lei Li, Wanting Zhou

Abstract:

Visual Simultaneous Localization and Mapping (VSLAM) is a technology that obtains information in the environment for self-positioning and mapping. It is widely used in computer vision, robotics and other fields. Many visual SLAM systems, such as OBSLAM3, employ a constant-speed motion model that provides the initial pose of the current frame to improve the speed and accuracy of feature matching. However, in actual situations, the constant velocity motion model is often difficult to be satisfied, which may lead to a large deviation between the obtained initial pose and the real value, and may lead to errors in nonlinear optimization results. Therefore, this paper proposed a motion model based on acceleration, which can be applied on most SLAM systems. In order to better describe the acceleration of the camera pose, we decoupled the pose transformation matrix, and calculated the rotation matrix and the translation vector respectively, where the rotation matrix is represented by rotation vector. We assume that, in a short period of time, the changes of rotating angular velocity and translation vector remain the same. Based on this assumption, the initial pose of the current frame is estimated. In addition, the error of constant velocity model was analyzed theoretically. Finally, we applied our proposed approach to the ORBSLAM3 system and evaluated two sets of sequences on the TUM dataset. The results showed that our proposed method had a more accurate initial pose estimation and the accuracy of ORBSLAM3 system is improved by 6.61% and 6.46% respectively on the two test sequences.

Keywords: error estimation, constant acceleration motion model, pose estimation, visual SLAM

Procedia PDF Downloads 99

Authors: Mona Hejazi, Ali Motie Nasrabadi

Abstract:

Epilepsy is a persistent neurological disorder that affects more than 50 million people worldwide. Hence, there is a necessity to introduce an efficient prediction model for making a correct diagnosis of the epileptic seizure and accurate prediction of its type. In this study we consider how the Effective Connectivity (EC) patterns obtained from intracranial Electroencephalographic (EEG) recordings reveal information about the dynamics of the epileptic brain and can be used to predict imminent seizures, as this will enable the patients (and caregivers) to take appropriate precautions. We use this definition because we believe that effective connectivity near seizures begin to change, so we can predict seizures according to this feature. Results are reported on the standard Freiburg EEG dataset which contains data from 21 patients suffering from medically intractable focal epilepsy. Six channels of EEG from each patients are considered and effective connectivity using Directed Transfer Function (DTF) and Granger Causality (GC) methods is estimated. We concentrate on effective connectivity standard deviation over time and feature changes in five brain frequency sub-bands (Alpha, Beta, Theta, Delta, and Gamma) are compared. The performance obtained for the proposed scheme in predicting seizures is: average prediction time is 50 minutes before seizure onset, the maximum sensitivity is approximate ~80% and the false positive rate is 0.33 FP/h. DTF method is more acceptable to predict epileptic seizures and generally we can observe that the greater results are in gamma and beta sub-bands. The research of this paper is significantly helpful for clinical applications, especially for the exploitation of online portable devices.

Keywords: effective connectivity, Granger causality, directed transfer function, epilepsy seizure prediction, EEG

Procedia PDF Downloads 472

Authors: Mohammed Bnebaghdad, Feriel Betouche, Malika Semmani

Abstract:

Image processing become more sophisticated over time due to technological advances, especially artificial intelligence (AI) technology. Currently, AI image processing is used in many areas, including surveillance, industry, science, and medicine. AI in medical image processing can help doctors diagnose diseases faster, with minimal mistakes, and with less effort. Among these diseases is malaria, which remains a major public health challenge in many parts of the world. It affects millions of people every year, particularly in tropical and subtropical regions. Early detection of malaria is essential to prevent serious complications and reduce the burden of the disease. In this paper, we propose and implement a scheme based on AI image processing to enhance malaria disease diagnosis through automated analysis of blood smear images. The scheme is based on the convolutional neural network (CNN) method. So, we have developed a model that classifies infected and uninfected single red cells using images available on Kaggle, as well as real blood smear images obtained from the Central Laboratory of Medical Biology EHS Laadi Flici (formerly El Kettar) in Algeria. The real images were segmented into individual cells using the watershed algorithm in order to match the images from the Kaagle dataset. The model was trained and tested, achieving an accuracy of 99% and 97% accuracy for new real images. This validates that the model performs well with new real images, although with slightly lower accuracy. Additionally, the model has been embedded in a Raspberry Pi4, and a graphical user interface (GUI) was developed to visualize the malaria diagnostic results and facilitate user interaction.

Keywords: medical image processing, malaria parasite, classification, CNN, artificial intelligence

Procedia PDF Downloads 25

Authors: Gerardo Arias, Richard Wall, Jamie Stevens

Abstract:

Lucilia bufonivora Moniez, is an obligate parasite of toads and frogs widely distributed in Europe. Its sister taxon Lucilia silvarum Meigen behaves mainly as a carrion breeder in Europe, however it has been reported as a facultative parasite of amphibians. These two closely related species are morphologically almost identical, which has led to misidentification, and in fact, it has been suggested that the amphibian myiasis cases by L. silvarum reported in Europe should be attributed to L. bufonivora. Both species remain poorly studied and their taxonomic relationships are still unclear. The identification of the larval specimens involved in amphibian myiasis with molecular tools and phylogenetic analysis of these two closely related species may resolve this problem. In this work seventeen unidentified larval specimens extracted from toad myiasis cases of the UK, the Netherlands and Switzerland were obtained, their COX1 (mtDNA) and EF1-α (Nuclear DNA) gene regions were amplified and then sequenced. The 17 larval samples were identified with both molecular markers as L. bufonivora. Phylogenetic analysis was carried out with 10 other blowfly species, including L. silvarum samples from the UK and USA. Bayesian Inference trees of COX1 and a combined-gene dataset suggested that L. silvarum and L. bufonivora are separate sister species. However, the nuclear gene EF1-α does not appear to resolve their relationships, suggesting that the rates of evolution of the mtDNA are much faster than those of the nuclear DNA. This work provides the molecular evidence for successful identification of L. bufonivora and a molecular analysis of the populations of this obligate parasite from different locations across Europe. The relationships with L. silvarum are discussed.

Keywords: calliphoridae, molecular evolution, myiasis, obligate parasitism

Procedia PDF Downloads 243

Authors: Qingqing Feng, S. Thomas Ng, Frank Xu

Abstract:

Energy consumption in domestic sector has been increasing rapidly in China all along these years. Confronted with environmental challenges, the international society has made a concerted effort by setting the Paris Agreement, the Sustainable Development Goals, and the New Urban Agenda. Thus it’s very important for China to put forward reasonable countermeasures to boost building energy conservation which necessitates looking into the actuality of residential energy end-use profile and its influence factors. In this study, questionnaire surveys have been conducted in Zhuhai city in China, a typical city in hot summer warm winter climate zone. The data solicited mainly include the occupancy schedule, building’s information, residents’ information, household energy uses, the type, quantity and use patterns of appliances and occupants’ satisfaction. Over 200 valid samples have been collected through face-to-face interviews. Descriptive analysis, clustering analysis, correlation analysis and sensitivity analysis were then conducted on the dataset to understand the energy end-use profile. The findings identify: 1) several typical clusters of occupancy patterns and appliances utilization patterns; 2) the top three sensitive factors influencing energy consumption; 3) the correlations between satisfaction and energy consumption. For China with many different climates zones, it’s difficult to find a silver bullet on energy conservation. The aim of this paper is to provide a theoretical basis for multi-stakeholders including policy makers, residents, and academic communities to formulate reasonable energy saving blueprints for hot and humid urban residential buildings in China.

Keywords: residential building, energy end-use profile, questionnaire survey, sustainability

Procedia PDF Downloads 137

Authors: Jacob Green, Cecilia Cabrera, Maximilian Jakobs, Andrea Dimitracopoulos, Mark van der Wilk, Ryan Greenhalgh

Abstract:

Predicting drug properties is key in drug discovery to enable de-risking of assets before expensive clinical trials and to find highly active compounds faster. Interest from the machine learning community has led to the release of a variety of benchmark datasets and proposed methods. However, it remains unclear for practitioners which method or approach is most suitable, as different papers benchmark on different datasets and methods, leading to varying conclusions that are not easily compared. Our large-scale empirical study links together numerous earlier works on different datasets and methods, thus offering a comprehensive overview of the existing property classes, datasets, and their interactions with different methods. We emphasise the importance of uncertainty quantification and the time and, therefore, cost of applying these methods in the drug development decision-making cycle. To the best of the author's knowledge, it has been observed that the optimal approach varies depending on the dataset and that engineered features with classical machine learning methods often outperform deep learning. Specifically, QSAR datasets are typically best analysed with classical methods such as Gaussian Processes, while ADMET datasets are sometimes better described by Trees or deep learning methods such as Graph Neural Networks or language models. Our work highlights that practitioners do not yet have a straightforward, black-box procedure to rely on and sets a precedent for creating practitioner-relevant benchmarks. Deep learning approaches must be proven on these benchmarks to become the practical method of choice in drug property prediction.

Keywords: activity (QSAR), ADMET, classical methods, drug property prediction, empirical study, machine learning

Procedia PDF Downloads 86