Search results for: feature selection

Authors: Yildiz Stella Dak, Jale Tezcan

Abstract:

Ground motion models that relate a strong motion parameter of interest to a set of predictive seismological variables describing the earthquake source, the propagation path of the seismic wave, and the local site conditions constitute a critical component of seismic hazard analyses. When a sufficient number of strong motion records are available, ground motion relations are developed using statistical analysis of the recorded ground motion data. In regions lacking a sufficient number of recordings, a synthetic database is developed using stochastic, theoretical or hybrid approaches. Regardless of the manner the database was developed, ground motion relations are developed using regression analysis. Development of a ground motion relation is a challenging process which inevitably requires the modeler to make subjective decisions regarding the inclusion criteria of the recordings, the functional form of the model and the set of seismological variables to be included in the model. Because these decisions are critically important to the validity and the applicability of the model, there is a continuous interest on procedures that will facilitate the development of ground motion models. This paper proposes the use of the Least Absolute Shrinkage and Selection Operator (LASSO) in selecting the set predictive seismological variables to be used in developing a ground motion relation. The LASSO can be described as a penalized regression technique with a built-in capability of variable selection. Similar to the ridge regression, the LASSO is based on the idea of shrinking the regression coefficients to reduce the variance of the model. Unlike ridge regression, where the coefficients are shrunk but never set equal to zero, the LASSO sets some of the coefficients exactly to zero, effectively performing variable selection. Given a set of candidate input variables and the output variable of interest, LASSO allows ranking the input variables in terms of their relative importance, thereby facilitating the selection of the set of variables to be included in the model. Because the risk of overfitting increases as the ratio of the number of predictors to the number of recordings increases, selection of a compact set of variables is important in cases where a small number of recordings are available. In addition, identification of a small set of variables can improve the interpretability of the resulting model, especially when there is a large number of candidate predictors. A practical application of the proposed approach is presented, using more than 600 recordings from the National Geospatial-Intelligence Agency (NGA) database, where the effect of a set of seismological predictors on the 5% damped maximum direction spectral acceleration is investigated. The set of candidate predictors considered are Magnitude, Rrup, Vs30. Using LASSO, the relative importance of the candidate predictors has been ranked. Regression models with increasing levels of complexity were constructed using one, two, three, and four best predictors, and the models’ ability to explain the observed variance in the target variable have been compared. The bias-variance trade-off in the context of model selection is discussed.

Keywords: ground motion modeling, least absolute shrinkage and selection operator, penalized regression, variable selection

Procedia PDF Downloads 305

Authors: Benjamin Beltzung, Marie Pelé, Julien P. Renoult, Cédric Sueur

Abstract:

To this day, it is not known if drawing is specifically human behavior or if this behavior finds its origins in ancestor species. An interesting window to enlighten this question is to analyze the drawing behavior in genetically close to human species, such as non-human primate species. A good candidate for this approach is the orangutan, who shares 97% of our genes and exhibits multiple human-like behaviors. Focusing on figurative aspects may not be suitable for orangutans’ drawings, which may appear as scribbles but may have meaning. A manual feature selection would lead to an anthropocentric bias, as the features selected by humans may not match with those relevant for orangutans. In the present study, we used deep learning to analyze the drawings of a female orangutan named Molly († in 2011), who has produced 1,299 drawings in her last five years as part of a behavioral enrichment program at the Tama Zoo in Japan. We investigate multiple ways to decipher Molly’s drawings. First, we demonstrate the existence of differences between seasons by training a deep learning model to classify Molly’s drawings according to the seasons. Then, to understand and interpret these seasonal differences, we analyze how the information spreads within the network, from shallow to deep layers, where early layers encode simple local features and deep layers encode more complex and global information. More precisely, we investigate the impact of feature complexity on classification accuracy through features extraction fed to a Support Vector Machine. Last, we leverage style transfer to dissociate features associated with drawing style from those describing the representational content and analyze the relative importance of these two types of features in explaining seasonal variation. Content features were relevant for the classification, showing the presence of meaning in these non-figurative drawings and the ability of deep learning to decipher these differences. The style of the drawings was also relevant, as style features encoded enough information to have a classification better than random. The accuracy of style features was higher for deeper layers, demonstrating and highlighting the variation of style between seasons in Molly’s drawings. Through this study, we demonstrate how deep learning can help at finding meanings in non-figurative drawings and interpret these differences.

Keywords: cognition, deep learning, drawing behavior, interpretability

Procedia PDF Downloads 130

Authors: Teklu Urgessa, Wookjae Maeng, Joong Seek Lee

Abstract:

Application of five implementations of three data mining classification techniques was experimented for extracting important insights from tourism data. The aim was to find out the best performing algorithm among the compared ones for tourism knowledge discovery. Knowledge discovery process from data was used as a process model. 10-fold cross validation method is used for testing purpose. Various data preprocessing activities were performed to get the final dataset for model building. Classification models of the selected algorithms were built with different scenarios on the preprocessed dataset. The outperformed algorithm tourism dataset was Random Forest (76%) before applying information gain based attribute selection and J48 (C4.5) (75%) after selection of top relevant attributes to the class (target) attribute. In terms of time for model building, attribute selection improves the efficiency of all algorithms. Artificial Neural Network (multilayer perceptron) showed the highest improvement (90%). The rules extracted from the decision tree model are presented, which showed intricate, non-trivial knowledge/insight that would otherwise not be discovered by simple statistical analysis with mediocre accuracy of the machine using classification algorithms.

Keywords: classification algorithms, data mining, knowledge discovery, tourism

Procedia PDF Downloads 269

Authors: Jhuang Jin-Jhong, Hsieh Wei-Fan

Abstract:

Climate change causes natural disasters persistently. Therefore, nowadays environmental planning objective tends to the issues of respecting nature and coexisting with nature. As a result, the natural environment analysis, e.g., the analysis of topography, soil, hydrology, climate, vegetation, is highly emphasized. On the other hand, Feng Shui has been a criterion of site selection for residence in Eastern since the ancient times and has had farther influence on site selection for castles and even for temples and tombs. The primary criterion of site selection is judging the quality of Long: mountain range, Sha: nearby mountains, Shui: hydrology, Xue: foundation, Xiang: aspect, which are similar to the environmental variables of mountain range, topography, hydrology and aspect. For the reason, a lot researchers attempt to probe into the connection between the criterion of Feng Shui and environmental planning factors. Most researches only discussed with the composition and theory of space of Feng Shui, but there is no research which explained Feng Shui through the environmental field. Consequently, this study reviewed the theory of Feng Shui through the environmental planning viewpoint and assembled essential composition factors of Feng Shui. The results of this study point. From literature review and comparison of theoretical meanings, we find that the ideal principles for planning the Feng Shui environment can also be used for environmental planning. Therefore, this article uses 12 ideal environmental features used in Feng Shui to contrast the natural aspects of the environment and make comparisons with previous research and classifies the environmental factors into climate, topography, hydrology, vegetation, and soil.

Keywords: the composition of Feng Shui, environmental planning, site selection, main components of the Feng Shui environment

Procedia PDF Downloads 486

Authors: Bisi S. Olawoyin

Abstract:

This study attempts an exploration into behavioural finance in which the traditional assumptions of expected utility maximization with rational investors in efficient markets are dropped. It reviews prior research and evidence about how psychological biases affect investors behaviour and stock selection. This study examined the relationship between demographic variables and financial behaviour biases among public sector retirees who invested in the Nigerian Stock Exchange prior to their retirement. By using questionnaire survey method, a total of 214 valid convenient samples were collected in order to determine how specific demographic and psychological trait affect stock selection between dividend paying and non-dividend paying stocks. Descriptive statistics and OLS were used to analyse the results. Findings showed that most of the retirees prefer dividend paying stocks in few years preceding their retirement but still hold on to their non-dividend paying stock on retirement. A significant difference also exists between senior and junior retirees in preference for non-dividend paying stocks. These findings are consistent with the clientele theories of dividend.

Keywords: behavioural finance, clientele theories, dividend paying stocks, stock selection

Procedia PDF Downloads 114

Authors: Vanaja Rao

Abstract:

In the recent times, the need for the integration of Renewable Energy Sources (RES) in a Smart Grid is on the rise. As a result of this, associated energy storage systems are known to play important roles in sustaining the efficient operation of such RES like wind power and solar power. This paper investigates the importance of location selection of Energy Storage Systems (ESSs) in a Smart Grid. Three scenarios of ESS location is studied and analyzed in a Smart Grid, which are – 1. Near the generation/source, 2. In the middle of the Grid and, 3. Near the demand/consumption. This is explained with the aim of assisting any Distribution Network Operator (DNO) in deploying the ESSs in a power network, which will significantly help reduce the costs and time of planning and avoid any damages incurred as a result of installing them at an incorrect location of a Smart Grid. To do this, the outlined scenarios mentioned above are modelled and analyzed with the National Grid’s datasets of energy generation and consumption in the UK power network. As a result, the outcome of this analysis aims to provide a better overview for the location selection of the ESSs in a Smart Grid. This ensures power system stability and security along with the optimum usage of the ESSs.

Keywords: distribution networks, energy storage system, energy security, location planning, power stability, smart grid

Procedia PDF Downloads 277

Authors: Ali Asghar Memon

Abstract:

A cautious and astute selection of switching devices used in power electronic converters of a switched reluctance (SR) motor is required. It is a matter of choice of best switching devices with respect to their switching ability rather than fulfilling the number of switches. This paper highlights the computational determination of switching losses comprising of switch-on, switch-off and conduction losses respectively by using experimental data in simulation model of a SR machine. The finding of this research is helpful for proper selection of electronic switches and suitable converter topology for switched reluctance motor.

Keywords: converter, operating modes, switched reluctance motor, switching losses

Procedia PDF Downloads 477

Authors: Fatima Akhmedova, Simon Liao

Abstract:

One of the most critical decision points in the design of a face recognition system is the choice of an appropriate face representation. Effective feature descriptors are expected to convey sufficient, invariant and non-redundant facial information. In this work, we propose a set of Hahn moments as a new approach for feature description. Hahn moments have been widely used in image analysis due to their invariance, non-redundancy and the ability to extract features either globally and locally. To assess the applicability of Hahn moments to Face Recognition we conduct two experiments on the Olivetti Research Laboratory (ORL) database and University of Notre-Dame (UND) X1 biometric collection. Fusion of the global features along with the features from local facial regions are used as an input for the conventional k-NN classifier. The method reaches an accuracy of 93% of correctly recognized subjects for the ORL database and 94% for the UND database.

Keywords: face recognition, Hahn moments, recognition-by-parts, time-lapse

Procedia PDF Downloads 341

Authors: Nisa Ertoy Inci, Cengiz Toker

Abstract:

Winter or autumn-sown faba bean (Vicia faba L.) is one the most efficient ways to overcome drought since faba bean is usually grown under rainfed where drought and high-temperature stresses are the main growth constraints. The objectives of this study were assessment of (i) relationships between cold tolerance and agro-morphological traits, and (ii) the most suitable agro-morphological trait(s) under cold conditions. Three species of the genus Vicia L. includes 109 genotypes of faba bean (Vicia faba L.), three genotypes of narbon bean (V. narbonensis L.) and two genotypes of V. montbretii Fisch. & C.A. Mey. Davis and Plitmann were sown in autumn at highland of Mediterranean region of Turkey. All relatives of faba bean were more cold-tolerant than the faba bean genotypes. Three faba bean genotypes, ACV-42, ACV-84 and ACV-88, were selected as sources of cold tolerance under field conditions. Path and correlation coefficients and factor and principal component analyses indicated that biological yield should be evaluated in selection for cold tolerance under cold conditions ahead of many agro-morphological traits. The seed weight should be considered for selection in early breeding generations because they had the highest heritability.

Keywords: cold tolerance, faba bean, narbon bean, selection

Procedia PDF Downloads 365

Authors: S. Kherchaoui, A. Houacine

Abstract:

This paper presents a facial expression recognition system. It performs identification and classification of the seven basic expressions; happy, surprise, fear, disgust, sadness, anger, and neutral states. It consists of three main parts. The first one is the detection of a face and the corresponding facial features to extract the most expressive portion of the face, followed by a normalization of the region of interest. Then calculus of curvelet coefficients is performed with dimensionality reduction through principal component analysis. The resulting coefficients are combined with two ratios; mouth ratio and face edge ratio to constitute the whole feature vector. The third step is the classification of the emotional state using the SVM method in the feature space.

Keywords: facial expression identification, curvelet coefficient, support vector machine (SVM), recognition system

Procedia PDF Downloads 213

Authors: Oshin Anand, Atanu Rakshit

Abstract:

The vast increment in the e-commerce ventures makes this area a prominent research stream. Besides several quantified parameters, the textual content of reviews is a storehouse of many information that can educate companies and help them earn profit. This study is an attempt in this direction. The article attempts to categorize data based on a computed metric that quantifies the influencing capacity of reviews rendering two categories of high and low influential reviews. Further, each of these document is studied to conclude several product feature categories. Each of these categories along with the computed metric is converted to linguistic identifiers and are used in an association mining model. The article makes a novel attempt to combine feature attraction with quantified metric to categorize review text and finally provide frequent patterns that depict customer preferences. Frequent mentions in a highly influential score depict customer likes or preferred features in the product whereas prominent pattern in low influencing reviews highlights what is not important for customers. This is achieved using a hybrid approach of text mining for feature and term extraction, sentiment analysis, multicriteria decision-making technique and association mining model.

Keywords: association mining, customer preference, frequent pattern, online reviews, text mining

Procedia PDF Downloads 365

Authors: Anitha Narasimhan, Aarthy Anandan, Madhan Karky, C. N. Subalalitha

Abstract:

Games can be the excellent tools for teaching a language. There are few e-learning games in Indian languages like word scrabble, cross word, quiz games etc., which were developed mainly for educational purposes. This paper proposes a Tamil word game called, “Porul”, which focuses on education as well as on players’ thinking and decision-making skills. Porul is a multiple choice based quiz game, in which the players attempt to answer questions correctly from the given multiple options that are generated using a unique algorithm called the Option Selection algorithm which explores the semantics of the question in various dimensions namely, synonym, rhyme and Universal Networking Language semantic category. This kind of semantic exploration of the question not only increases the complexity of the game but also makes it more interesting. The paper also proposes a Scoring Algorithm which allots a score based on the popularity score of the question word. The proposed game has been tested using 20,000 Tamil words.

Keywords: Porul game, Tamil word game, option selection, flash card, scoring, algorithm

Procedia PDF Downloads 386

Authors: Rawan A. Abdelrazeq, Ahmed M. Khalafallah, Nabil A. Kartam

Abstract:

Selection of competent contractors for construction projects is usually accomplished through competitive bidding or negotiated contracting in which the contract bid price is the basic criterion for selection. The evaluation of contractor’s safety performance is still not a typical criterion in the selection process, despite the existence of various safety prequalification procedures. There is a critical need for practical and automated systems that enable owners and decision makers to evaluate contractor safety performance, among other important contractor selection criteria. These systems should ultimately favor safety-conscious contractors to be selected by the virtue of their past good safety records and current safety programs. This paper presents an exploratory sequential mixed-methods approach to develop a framework for an automated decision support system that evaluates contractor safety performance based on a multitude of indicators and metrics that have been identified through a comprehensive review of construction safety research, and a survey distributed to domain experts. The framework is developed in three phases: (1) determining the indicators that depict contractor current and past safety performance; (2) soliciting input from construction safety experts regarding the identified indicators, their metrics, and relative significance; and (3) designing a decision support system using relational database models to integrate the identified indicators and metrics into a system that assesses and rates the safety performance of contractors. The proposed automated system is expected to hold several advantages including: (1) reducing the likelihood of selecting contractors with poor safety records; (2) enhancing the odds of completing the project safely; and (3) encouraging contractors to exert more efforts to improve their safety performance and practices in order to increase their bid winning opportunities which can lead to significant safety improvements in the construction industry. This should prove useful to decision makers and researchers, alike, and should help improve the safety record of the construction industry.

Keywords: construction safety, contractor selection, decision support system, relational database

Procedia PDF Downloads 253

Authors: Hanlin Liu, Hua Li, Yintan AI

Abstract:

Network performance prediction (NPP) is essential for the management and optimization of software-defined networking (SDN) and contributes to improving the quality of service (QoS) in SDN to meet the requirements of users. Although current deep learning-based methods can achieve high effectiveness, they still suffer from some problems, such as difficulty in capturing global information of the network, inefficiency in modeling end-to-end network performance, and inadequate graph feature extraction. To cope with these issues, our proposed Graphormer-based architecture for NPP leverages the powerful graph representation ability of Graphormer to effectively model the graph structure data, and a node-edge transformation algorithm is designed to transfer the feature extraction object from nodes to edges, thereby effectively extracting the end-to-end performance characteristics of the network. Moreover, routing oriented centrality measure coefficient for nodes and edges is proposed respectively to assess their importance and influence within the graph. Based on this coefficient, an enhanced feature extraction method and an advanced centrality encoding strategy are derived to fully extract the structural information of the graph. Experimental results on three public datasets demonstrate that the proposed GraphNPP architecture can achieve state-of-the-art results compared to current NPP methods.

Keywords: software-defined networking, network performance prediction, Graphormer, graph neural network

Procedia PDF Downloads 16

Authors: Tao Song, Hao Cheng, Guangfeng Tian, Chuang Xu

Abstract:

An autonomous emergency braking system is an advanced driving assistance system that enables vehicle collision avoidance and pedestrian collision avoidance to improve vehicle safety. At present, because the pedestrian target is small, and the mobility is large, the pedestrian AEB system is faced with more technical difficulties and higher functional requirements. In this paper, a method of pedestrian target selection based on a variable width funnel is proposed. Based on the current position and predicted position of pedestrians, the relative position of vehicle and pedestrian at the time of collision is calculated, and different braking strategies are adopted according to the hazard level of pedestrian collisions. In the CNCAP standard operating conditions, comparing the method of considering only the current position of pedestrians and the method of considering pedestrian prediction position, as well as the method based on fixed width funnel and variable width funnel, the results show that, based on variable width funnel, the choice of pedestrian target will be more accurate and the opportunity of the intervention of AEB system will be more reasonable by considering the predicted position of the pedestrian target and vehicle's lateral motion.

Keywords: automatic emergency braking system, pedestrian target selection, TTC, variable width funnel

Procedia PDF Downloads 134

Authors: Caleb B. Sagwa, Tobias O. Okeno, Alexander K. Kahi

Abstract:

Dairy cattle production plays an important role in the Kenyan economy. However, high incidences of mastitis is a major setback to the productivity in this industry. The current dairy cattle breeding objective in Kenya does not include mastitis resistance, mainly because the economic value of mastitis resistance has not been determined. Therefore this study aimed at estimating the economic value of mastitis resistance in dairy cattle in Kenya. Initial input parameters were obtained from literature on dairy cattle production systems in the tropics. Selection index methodology was used to derive the economic value of mastitis resistance. Somatic cell count (SCC) was used an indicator trait for mastitis resistance. The economic value was estimated relative to milk yield (MY). Economic values were assigned to SCC in a selection index such that the overall gain in the breeding goal trait was maximized. The option of estimating the economic value for SCC by equating the response in the trait of interest to its index response was considered. The economic value of mastitis resistance was US $23.64 while maximum response to selection for MY was US $66.01. The findings of this study provide vital information that is a pre-requisite for the inclusion of mastitis resistance in the current dairy cattle breeding goal in Kenya.

Keywords: somatic cell count, milk quality, payment system, breeding goal

Procedia PDF Downloads 231

Authors: Jiaqi Yan, Yani Shi

Abstract:

With the growing popularity of bitcoin transactions, criminals have exploited the bitcoin like cryptocurrencies, and cybercriminals such as money laundering have thrived. Unlike traditional currencies, the Internet-based virtual currencies can be used anonymously via the blockchain technology underpinning. In this paper, we analyze the double-edged sword features of blockchain technology in the context of money laundering. In particular, the traceability feature of blockchain-based system facilitates a level of governance, while the decentralization feature of blockchain-based system may bring governing difficulties. Based on the analysis, we propose guidelines for policy makers in governing blockchain-based cryptocurrency systems.

Keywords: cryptocurrency, money laundering, blockchain, decentralization, traceability

Procedia PDF Downloads 165

Authors: R. S. Vikaash, S. Vinodh, T. S. Sai Prashanth

Abstract:

Six sigma is a defect reduction strategy enabling modern organizations to achieve business prosperity. The practitioners are in need to select best six sigma project among the available alternatives to achieve customer satisfaction. In this circumstance, this article presents a study in which six sigma project selection is formulated as Multi-Criteria Decision-Making(MCDM) problem and the best project has been found using AHP. Five main governing criteria and 14 sub criteria are being formulated. The decision maker’s inputs were gathered and computations were performed. The project with the high values from the set of projects is selected as the best project. Based on calculations, Project “P1”is found to be the best and further deployment actions have been undertaken in the organization.

Keywords: six Sigma, project selection, MCDM, analytic hierarchy process, business prosperity

Procedia PDF Downloads 321

Authors: Jenq-Shyong Chen, Ben-Fong Yu

Abstract:

CNC(computer numerical control) machining centers have been widely used for machining different metal components for various industries. For a specific CNC machine, its everyday job is assigned to cut different products with quite different attributes such as material type, workpiece weight, geometry, tooling, and cutting conditions. Theoretically, the dynamic characteristics of the CNC machine should be properly tuned match each machining job in order to get the optimal machining performance. However, most of the CNC machines are set with only a standard set of CNC parameters. In this study, we have developed an auto-tuning system which can automatically change the CNC parameters and in hence change the machine dynamic characteristics according to the selection of machining modes which are set by the mixed combination of three machine performance indexes: the HO (high surface quality) index, HP (high precision) index and HS (high speed) index. The acceleration, jerk, corner error tolerance, oscillation and dynamic bandwidth of machine’s feed axes have been changed according to the selection of the machine performance indexes. The proposed auto-tuning system of the CNC parameters has been implemented on a PC-based CNC controller and a three-axis machining center. The measured experimental result have shown the promising of our proposed auto-tuning system.

Keywords: auto-tuning, CNC parameters, machining mode, high speed, high accuracy, high surface quality

Procedia PDF Downloads 360

Authors: Bernd Kiefer, Abraham Gebru Tesfay, Dietrich Klakow

Abstract:

The motivation of our work is to detect different terrain types traversed by a robot based on acoustic data from the robot-terrain interaction. Different acoustic features and classifiers were investigated, such as Mel-frequency cepstral coefficient and Gamma-tone frequency cepstral coefficient for the feature extraction, and Gaussian mixture model and Feed forward neural network for the classification. We analyze the system’s performance by comparing our proposed techniques with some other features surveyed from distinct related works. We achieve precision and recall values between 87% and 100% per class, and an average accuracy at 95.2%. We also study the effect of varying audio chunk size in the application phase of the models and find only a mild impact on performance.

Keywords: acoustic features, autonomous robots, feature extraction, terrain classification

Procedia PDF Downloads 338

Authors: Belaynesh Chekol, Numan Çelebi

Abstract:

The main purpose of this work is to recognize individual characters extracted from natural scene images using scale invariant feature transform (SIFT) features as an input to K-nearest neighbor (KNN); a classification learner algorithm. For this task, 1,068 and 78 images of English alphabet characters taken from Chars74k data set is used to train and test the classifier respectively. For each character image, We have generated describing features by using SIFT algorithm. This set of features is fed to the learner so that it can recognize and label new images of English characters. Two types of KNN (fine KNN and weighted KNN) were trained and the resulted classification accuracy is 56.9% and 56.5% respectively. The training time taken was the same for both fine and weighted KNN.

Keywords: character recognition, KNN, natural scene image, SIFT

Procedia PDF Downloads 257

Authors: A. Degale Desta, Cheng Jian

Abstract:

Deep learning application in computer vision is rapidly advancing, giving it the ability to monitor the public and quickly identify potentially anomalous behaviour from crowd scenes. Therefore, the purpose of the current work is to improve the performance of safety of people in crowd events from panic behaviour through introducing the innovative idea of Aggregation of Ensembles (AOE), which makes use of the pre-trained ConvNets and a pool of classifiers to find anomalies in video data with packed scenes. According to the theory of algorithms that applied K-means, KNN, CNN, SVD, and Faster-CNN, YOLOv5 architectures learn different levels of semantic representation from crowd videos; the proposed approach leverages an ensemble of various fine-tuned convolutional neural networks (CNN), allowing for the extraction of enriched feature sets. In addition to the above algorithms, a long short-term memory neural network to forecast future feature values and a handmade feature that takes into consideration the peculiarities of the crowd to understand human behavior. On well-known datasets of panic situations, experiments are run to assess the effectiveness and precision of the suggested method. Results reveal that, compared to state-of-the-art methodologies, the system produces better and more promising results in terms of accuracy and processing speed.

Keywords: action recognition, computer vision, crowd detecting and tracking, deep learning

Procedia PDF Downloads 124

Authors: Nitin Shravan, Sudarsun Santhiappan, B. Sivaselvan

Abstract:

Subsection identification, in the context of Electronic Health Records (EHRs), is identifying the important sections for down-stream tasks like auto-coding. In this work, we classify the text present in EHRs according to their information, using machine learning and deep learning techniques. We initially describe briefly about the problem and formulate it as a text classification problem. Then, we discuss upon the methods from the literature. We try two approaches - traditional feature extraction based machine learning methods and deep learning methods. Through experiments on a private dataset, we establish that the deep learning methods perform better than the feature extraction based Machine Learning Models.

Keywords: deep learning, machine learning, semantic clinical classification, subsection identification, text classification

Procedia PDF Downloads 183

Authors: Abeer Amayri, Akif A. Bulgak

Abstract:

Due to today’s globalization as well as outsourcing practices of the companies, the Supply Chain (SC) performances have become more dependent on the efficient movement of material among places that are geographically dispersed, where there is more chance for disruptions. One such disruption is the quality and delivery uncertainties of outsourcing. These uncertainties could lead the products to be unsafe and, as is the case in a number of recent examples, companies may have to end up in recalling their products. As a result of these problems, there is a need to develop a methodology for selecting suppliers globally in view of risks associated with low quality and late delivery. Accordingly, we developed a two-stage stochastic model that captures the risks associated with uncertainty in quality and delivery as well as a solution procedure for the model. The stochastic model developed simultaneously optimizes supplier selection and purchase quantities under price discounts over a time horizon. In particular, our target is the study of global organizations with multiple sites and multiple overseas suppliers, where the pricing is offered in suppliers’ local currencies. Our proposed methodology is applied to a case study for a US automotive company having two assembly plants and four potential global suppliers to illustrate how the proposed model works in practice.

Keywords: global supply chains, quality, stochastic programming, supplier selection

Procedia PDF Downloads 431

Authors: Syeda Mainaaz Unnisa, Deepa Suresh

Abstract:

Due to the increase in usage of public domain channels, such as the internet, and communication technology, there is a concern about the protection of intellectual property and security threats. This interest has led to growth in researching and implementing techniques for information hiding. Steganography is the art and science of hiding information in a private manner such that its existence cannot be recognized. Communication using steganographic techniques makes not only the secret message but also the presence of hidden communication, invisible. Steganalysis is the art of detecting the presence of this hidden communication. Parallel to steganography, steganalysis is also gaining prominence, since the detection of hidden messages can prevent catastrophic security incidents from occurring. Steganalysis can also be incredibly helpful in identifying and revealing holes with the current steganographic techniques, which makes them vulnerable to attacks. Through the formulation of new effective steganalysis methods, further research to improve the resistance of tested steganography techniques can be developed. Feature-based steganalysis method for JPEG images calculates the features of an image using the L1 norm of the difference between a stego image and the calibrated version of the image. This calibration can help retrieve some of the parameters of the cover image, revealing the variations between the cover and stego image and enabling a more accurate detection. Applying this method to various steganographic schemes, experimental results were compared and evaluated to derive conclusions and principles for more protected JPEG steganography.

Keywords: cover image, feature-based steganalysis, information hiding, steganalysis, steganography

Procedia PDF Downloads 187

Authors: Nalini C. Iyer, Shraddha H., Anagha B. Varahamurthy, Dikshith C. S., Ishwar G. Kubasad, Vinayak I. Karalatti, Pavan B. Mulimani

Abstract:

The current scenario of the pandemic due to COVID-19 has led to the awareness among the people to avoid unneces-sary contact in public places. There is a need to avoid contact with physical objects to stop the spreading of infection. The contactless feature has to be included in the systems in public places wherever possible. For example, attendance monitoring systems with fingerprint biometric can be replaced with a contactless feature. One more important protocol followed in the current situation is temperature monitoring and screening. The paper describes an attendance system with a contactless feature and temperature screening for the university. The system displays a QR code to scan, which redirects to the student login web page only if the location is valid (the location where the student scans the QR code should be the location of the display of the QR code). Once the student logs in, the temperature of the student is scanned by the contactless temperature sensor (mlx90614) with an error of 0.5°C. If the temperature falls in the range of the desired value (range of normal body temperature), then the attendance of the student is marked as present, stored in the database, and the door opens automatically. The attendance is marked as absent in the other case, alerted with the display of temperature, and the door remains closed. The door is automated with the help of a servomotor. To avoid the proxy, IR sensors are used to count the number of students in the classroom. The hardware system consisting of a contactless temperature sensor and IR sensor is implemented on the microcontroller, NodeMCU.

Keywords: NodeMCU, IR sensor, attendance monitoring, contactless, temperature

Procedia PDF Downloads 158

Authors: Kilari Nikhil, Ankur Tibrewal, Srinivas Kruthiventi S. S.

Abstract:

Conventional speaker identification systems often fall short of capturing the diverse variations present in speech data due to fixed-scale architectures. In this research, we propose a CNN-based architecture, USENet, designed to overcome these limitations. Leveraging two key techniques, our approach achieves superior performance on the VoxCeleb 1 Dataset without any pre-training. Firstly, we adopt a U-net-inspired design to extract features at multiple scales, empowering our model to capture speech characteristics effectively. Secondly, we introduce the squeeze and excitation block to enhance spatial feature learning. The proposed architecture showcases significant advancements in speaker identification, outperforming existing methods, and holds promise for future research in this domain.

Keywords: multi-scale feature extraction, squeeze and excitation, VoxCeleb1 speaker identification, mel-spectrograms, USENet

Procedia PDF Downloads 45

Authors: Kun Xu, Yuan Xu, Jia Qiao

Abstract:

The document detection plays an important role in optical character recognition and text analysis. Because the traditional detection methods have weak generalization ability, and deep neural network has complex structure and large number of parameters, which cannot be well applied in mobile devices, this paper proposes a lightweight Document Corner Detection Network (DCDNet). DCDNet is a two-stage architecture. The first stage with Encoder-Decoder structure adopts depthwise separable convolution to greatly reduce the network parameters. After introducing the Feature Attention Union (FAU) module, the second stage enhances the feature information of spatial and channel dim and adaptively adjusts the size of receptive field to enhance the feature expression ability of the model. Aiming at solving the problem of the large difference in the number of pixel distribution between corner and non-corner, Weighted Binary Cross Entropy Loss (WBCE Loss) is proposed to define corner detection problem as a classification problem to make the training process more efficient. In order to make up for the lack of Dataset of document corner detection, a Dataset containing 6620 images named Document Corner Detection Dataset (DCDD) is made. Experimental results show that the proposed method can obtain fast, stable and accurate detection results on DCDD.

Keywords: document detection, corner detection, attention mechanism, lightweight

Procedia PDF Downloads 327

Authors: Fathi Kallel, Abdulelah Alabd Uljabbar, Abdulrahman Aldukhail, Abdulaziz Alomran

Abstract:

The brain is an important organ in our body since it is responsible about the majority actions such as vision, memory, etc. However, different diseases such as Alzheimer and tumors could affect the brain and conduct to a partial or full disorder. Regular diagnosis are necessary as a preventive measure and could help doctors to early detect a possible trouble and therefore taking the appropriate treatment, especially in the case of brain tumors. Different imaging modalities are proposed for diagnosis of brain tumor. The powerful and most used modality is the Magnetic Resonance Imaging (MRI). MRI images are analyzed by doctor in order to locate eventual tumor in the brain and describe the appropriate and needed treatment. Diverse image processing methods are also proposed for helping doctors in identifying and analyzing the tumor. In fact, a large Computer Aided Diagnostic (CAD) tools including developed image processing algorithms are proposed and exploited by doctors as a second opinion to analyze and identify the brain tumors. In this paper, we proposed a new advanced CAD for brain tumor identification, classification and feature extraction. Our proposed CAD includes three main parts. Firstly, we load the brain MRI. Secondly, a robust technique for brain tumor extraction is proposed. This technique is based on both Discrete Wavelet Transform (DWT) and Principal Component Analysis (PCA). DWT is characterized by its multiresolution analytic property, that’s why it was applied on MRI images with different decomposition levels for feature extraction. Nevertheless, this technique suffers from a main drawback since it necessitates a huge storage and is computationally expensive. To decrease the dimensions of the feature vector and the computing time, PCA technique is considered. In the last stage, according to different extracted features, the brain tumor is classified into either benign or malignant tumor using Support Vector Machine (SVM) algorithm. A CAD tool for brain tumor detection and classification, including all above-mentioned stages, is designed and developed using MATLAB guide user interface.

Keywords: MRI, brain tumor, CAD, feature extraction, DWT, PCA, classification, SVM

Procedia PDF Downloads 223

Authors: Jingting Luo, Yong Wang, Ying Wang

Abstract:

Pedestrian detection in image or video data is a very important and challenging task in security surveillance. The difficulty of this task is to locate and detect pedestrians of different scales in complex scenes accurately. To solve these problems, a deep neural network (RT-YOLOv3) is proposed to realize real-time pedestrian detection at different scales in security monitoring. RT-YOLOv3 improves the traditional YOLOv3 algorithm. Firstly, the deep residual network is added to extract vehicle features. Then six convolutional neural networks with different scales are designed and fused with the corresponding scale feature maps in the residual network to form the final feature pyramid to perform pedestrian detection tasks. This method can better characterize pedestrians. In order to further improve the accuracy and generalization ability of the model, a hybrid pedestrian data set training method is used to extract pedestrian data from the VOC data set and train with the INRIA pedestrian data set. Experiments show that the proposed RT-YOLOv3 method achieves 93.57% accuracy of mAP (mean average precision) and 46.52f/s (number of frames per second). In terms of accuracy, RT-YOLOv3 performs better than Fast R-CNN, Faster R-CNN, YOLO, SSD, YOLOv2, and YOLOv3. This method reduces the missed detection rate and false detection rate, improves the positioning accuracy, and meets the requirements of real-time detection of pedestrian objects.

Keywords: pedestrian detection, feature detection, convolutional neural network, real-time detection, YOLOv3

Procedia PDF Downloads 116