Commenced in January 2007
Frequency: Monthly
Edition: International
Paper Count: 2597

World Academy of Science, Engineering and Technology

[Computer and Information Engineering]

Online ISSN : 1307-6892

2597 Convolutional Neural Networks for Feature Extraction and Automated Target Recognition in Synthetic Aperture Radar Images

Authors: Ying Zhao, John Geldmacher, Christopher Yerkes

Abstract:

Advances in the development of deep neural networks and other machine learning algorithms combined with ever more powerful hardware and the huge amount of data available on the internet has led to a revolution in ML research and applications. These advances present massive potential and opportunity for the military applications such as the analysis of Synthetic Aperture Radar (SAR) imagery. Synthetic Aperture Radar imagery is a useful tool capable of capturing high resolution images regardless of cloud coverage and at night. However, there is a limited amount of publicly available SAR data to train a machine learning model. This paper shows how to successfully dissect, modify, and re-architect cross-domain object recognition models such as the VGG-16 model, transfer learning models from the ImageNet, and the k-nearest neighbor (kNN) classifier. The paper demonstrates that the combinations of these factors can significantly and effectively improve the automated object recognition (ATR) of SAR clean and noisy images. The paper shows a potentially inexpensive, accurate, transfer and unsurpervised learning SAR ATR system when data labels are scarce and data are noisy, simplifying the whole recognition for the tactical operation requirements in the area of SAR ATR.

Keywords: Deep learning, Transfer Learning, kNN, SAR images, k-nearest neighbor, synthetic aperture radar images, VGG-16

Procedia PDF Downloads 1
2596 Face Recognition Using Body-Worn Camera: Dataset and Baseline Algorithms

Authors: Ali Almadan, Anoop Krishnan, Ajita Rattani

Abstract:

Facial recognition is a widely adopted technology in surveillance, border control, healthcare, banking services, and lately, in mobile user authentication with Apple introducing “Face ID” moniker with iPhone X. A lot of research has been conducted in the area of face recognition on datasets captured by surveillance cameras, DSLR, and mobile devices. Recently, face recognition technology has also been deployed on body-worn cameras to keep officers safe, enabling situational awareness and providing evidence for trial. However, limited academic research has been conducted on this topic so far, without the availability of any publicly available datasets with a sufficient sample size. This paper aims to advance research in the area of face recognition using body-worn cameras. To this aim, the contribution of this work is two-fold: (1) collection of a dataset consisting of a total of 136,939 facial images of 102 subjects captured using body-worn cameras in in-door and daylight conditions and (2) evaluation of various deep-learning architectures for face identification on the collected dataset. Experimental results suggest a maximum True Positive Rate(TPR) of 99.86% at False Positive Rate(FPR) of 0.000 obtained by SphereFace based deep learning architecture in daylight condition. The collected dataset and the baseline algorithms will promote further research and development. A downloadable link of the dataset and the algorithms is available by contacting the authors.

Keywords: Face Recognition, Deep learning, Person Identification, body-worn cameras

Procedia PDF Downloads 1
2595 The Vertex Degree Distance of One Vertex Union of the Cycle and the Star

Authors: Ying Wang, Haiyan Xie, Aoming Zhang

Abstract:

The degree distance of a graph is a graph invariant that is more sensitive than the Wiener index. In this paper, we calculate the vertex degree distances of one vertex union of the cycle and the star, and the degree distance of one vertex union of the cycle and the star. These results lay a foundation for further study on the extreme value of the vertex degree distances, and the distribution of the vertices with the extreme value in one vertex union of the cycle and the star.

Keywords: graph, degree distance, vertex-degree-distance, one vertex union of a cycle and a star

Procedia PDF Downloads 1
2594 Intrusion Detection Based on Graph Oriented Big Data Analytics

Authors: Ahlem Abid, Farah Jemili

Abstract:

Intrusion detection has been the subject of numerous studies in industry and academia, but cyber security analysts always want greater precision and global threat analysis to secure their systems in cyberspace. To improve intrusion detection system, the visualisation of the security events in form of graphs and diagrams is important to improve the accuracy of alerts. In this paper, we propose an approach of an IDS based on cloud computing, big data technique and using a machine learning graph algorithm which can detect in real time different attacks as early as possible. We use the MAWILab intrusion detection dataset . We choose Microsoft Azure as a unified cloud environment to load our dataset on. We implement the k2 algorithm which is a graphical machine learning algorithm to classify attacks. Our system showed a good performance due to the graphical machine learning algorithm and spark structured streaming engine.

Keywords: Machine Learning, Intrusion Detection, graph, Apache Spark Streaming, k2 algorithm, MAWILab, Microsoft Azure Cloud

Procedia PDF Downloads 1
2593 Green Thumb Engineering - Explainable Artificial Intelligence for Managing IoT Enabled Houseplants

Authors: Antti Nurminen, Avleen Malhi

Abstract:

Significant progress in intelligent systems in combination with exceedingly wide application domains having machine learning as the core technology are usually opaque, non-intuitive, and commonly complex for human users. We use innovative IoT technology which monitors and analyzes moisture, humidity, luminosity and temperature levels to assist end users for optimization of environmental conditions for their houseplants. For plant health monitoring, we construct a system yielding the Normalized Difference Vegetation Index (NDVI), supported by visual validation by users. We run the system for a selected plant, basil, in varying environmental conditions to cater for typical home conditions, and bootstrap our AI with the acquired data. For end users, we implement a web based user interface which provides both instructions and explanations.

Keywords: Intelligent Agent, IoT, NDVI, explainable artificial intelligence

Procedia PDF Downloads 1
2592 Scalable Action Mining for Recommendations to Reduce Hospital Readmission

Authors: Angelina Tzacheva, Arunkumar Bagavathi, Apurwa Apurwa

Abstract:

Hospital re-admission problem is one of the longtime issues of healthcares in USA. Unplanned re-admissions to hospitals not only increase cost for patients, but also for hospitals and taxpayers. Action mining is one of the data mining approaches to recommend actions to undertake for an organization or individual to achieve required condition or status. Undertaking such actionable recommendations incur some form of cost to users. The actionable recommendation system fails when the recommended actions are cost wise unendurable or non-profitable and uninteresting to the end user. Finding low cost actionable patterns in larger datasets is a time consuming and requires a scalable approach. In this work, we propose a scalable action mining method to recommend hospitals and taxpayers on what actions would potentially reduce patient readmission to hospitals at lowest costs. Most importantly we incorporated graph search methods to extract low cost actionable patterns. We use the Healthcare Cost and Utilization Project(HCUP) databases to evaluate our approach. All our proposed scalable approaches are cloud based and use Apache Spark to handle data processing and to make recommendations.

Keywords: Data Mining, Scalable, action mining, hospital readmission

Procedia PDF Downloads 1
2591 Churn Prediction for Telecommunication Industry Using Artificial Neural Networks

Authors: Ulas Vural

Abstract:

Telecommunication service providers demand accurate and precise prediction of customer churn probabilities to increase the effectiveness of their customer relation services. The large amount of customer data owned by the service providers is suitable for analysis by machine learning methods. In this study, expenditure data of customers are analyzed by using an artificial neural network (ANN). The ANN model is applied to the data of customers with different billing duration. The proposed model successfully predicts the churn probabilities at 83\% accuracy for only three months expenditure data and the prediction accuracy increases up to 89\% when the nine month data is used. The experiments also show that the accuracy of ANN model increases on an extended feature set with information of the changes on the bill amounts.

Keywords: Artificial Neural Networks, Deep learning, Customer Relationship Management, telecom industry, churn prediction

Procedia PDF Downloads 1
2590 A Review of Polymorphic Malware Detection Techniques

Authors: Joma Alrzini, Diane Pennington

Abstract:

Despite the continuous updating of anti-detection systems for malicious programs (malware), malware has moved to an abnormal threat level; it is being generated and spread faster than before. One of the most serious challenges faced by anti-detection malware programs is an automatic mutation in the code; this is called polymorphic malware via the polymorphic engine. In this case, it is difficult to block the impact of signature-based detection. Hence new techniques have to be used to analyse modern malware. One of these techniques is machine learning algorithms in a virtual machine (VM) that can run the packed malicious file and analyse it dynamically through automated testing of the code. Moreover, recent research used image processing techniques with deep learning framework as a hybrid method with two analysis types and extracting a feature engineering approach in the analysis process to detect polymorphic malware efficiently. This paper presents a brief review of the latest applied techniques against this type of malware with more focus on the machine learning method for analysing and detecting polymorphic malware. It will discuss briefly the merits and demerits of it.

Keywords: anti-detection, polymorphic automated testing, abnormal threats, packed malicious file

Procedia PDF Downloads 1
2589 Efficient Positioning of Data Aggregation Point for Wireless Sensor Network

Authors: Sifat Rahman Ahona, Rifat Tasnim, Naima Hassan

Abstract:

Data aggregation is a helpful technique for reducing the data communication overhead in wireless sensor network. One of the important tasks of data aggregation is positioning of the aggregator points. There are a lot of works done on data aggregation. But, efficient positioning of the aggregators points is not focused so much. In this paper, authors are focusing on the positioning or the placement of the aggregation points in wireless sensor network. Authors proposed an algorithm to select the aggregators positions for a scenario where aggregator nodes are more powerful than sensor nodes.

Keywords: Data Communication, Wireless Sensor Network, data aggregation, aggregation point

Procedia PDF Downloads 1
2588 Performance Testing of Deep Learning Networks on Printed Odia Characters

Authors: Sanjibani Pattanayak, Sateesh Pradhan, Ramesh Mallick

Abstract:

Deep machine learning includes a series of layers to mimic the working of the human brain for taking a decision. Deep learning networks have shown good results in character recognition in the past. This paper evaluates the performance of different deep learning networks like Convolutional Neural Network(CNN), Long Short-Term Memory(LSTM) based recurrent neural network and Convolutional LSTM on printed Odia characters. The Odia character database contains more than 24,000 images of printed Odia characters out of which 23,857 nos. of images are chosen for this work. Here, Convolutional LSTM is showing better results in terms of error rate, accuracy and no. of epochs in comparison to the other two. It gives 83% accuracy. Different pre-processing steps like binarization, size-normalization, blurring, interpolation, etc. are involved before passing the images to the deep neural networks.

Keywords: Character Recognition, convolutional neural network, long short-term memory, convolutional LSTM, Odia character database

Procedia PDF Downloads 1
2587 Machine Learning Strategies for Data Extraction from Unstructured Documents in Financial Services

Authors: Delphine Vendryes, Dushyanth Sekhar, Baojia Tong, Matthew Theisen, Chester Curme

Abstract:

Much of the data that inform the decisions of governments, corporations and individuals are harvested from unstructured documents. Data extraction is defined here as a process that turns non-machine-readable information into a machine-readable format that can be stored, for instance, in a database. In financial services, introducing more automation in data extraction pipelines is a major challenge. Information sought by financial data consumers is often buried within vast bodies of unstructured documents, which have historically required thorough manual extraction. Automated solutions provide faster access to non-machine-readable datasets, in a context where untimely information quickly becomes irrelevant. Data quality standards cannot be compromised, so automation requires high data integrity. This multifaceted task is broken down into smaller steps: ingestion, table parsing (detection and structure recognition), text analysis (entity detection and disambiguation), schema-based record extraction, user feedback incorporation. Selected intermediary steps are phrased as machine learning problems. Solutions leveraging cutting-edge approaches from the fields of computer vision (e.g. table detection) and natural language processing (e.g. entity detection and disambiguation) are proposed.

Keywords: Computer Vision, Finance, Machine Learning, Information Retrieval, natural language processing, Entity Recognition

Procedia PDF Downloads 1
2586 Use of Interpretable Evolved Search Query Classifiers for Sinhala Documents

Authors: Prasanna Haddela

Abstract:

Document analysis is a well matured yet still active research field, partly as a result of the intricate nature of building computational tools but also due to the inherent problems arising from the variety and complexity of human languages. Breaking down language barriers is vital in enabling access to a number of recent technologies. This paper investigates the application of document classification methods to new Sinhalese datasets. This language is geographically isolated and rich with many of its own unique features. We will examine the interpretability of the classification models with a particular focus on the use of evolved Lucene search queries generated using a Genetic Algorithm (GA) as a method of document classification. We will compare the accuracy and interpretability of these search queries with other popular classifiers. The results are promising and are roughly in line with previous work on English language datasets.

Keywords: Genetic Algorithm, evolved search queries, Sinhala document classification, Lucene Sinhala analyzer, interpretable text classification

Procedia PDF Downloads 1
2585 Keyloggers Prevention with Time-sensitive Obfuscation

Authors: Chien-Wei Hung, Fu-Hau Hsu, Chuan-Sheng Wang, Chia-Hao Lee

Abstract:

Nowadays, the abuse of keyloggers is one of the most widespread approaches to steal sensitive information. In this paper, we propose an On-Screen Prompts Approach to Keyloggers (OSPAK) and its analysis. OSPAK utilizes a canvas to cue users when their keystrokes are going to logged or ignored by OSPAK. This approach can protect computers against recoding sensitive inputs, which obfuscates keyloggers with letters inserted among users' keystrokes. Experimental results made by 95 volunteers show that OSPAK is a simple approach and easy to learn for users.

Keywords: Privacy, Computer Security, Authentication, keylogger, information leakage

Procedia PDF Downloads 1
2584 An Embarrassingly Simple Semi-supervised Approach to Increase Recall in Online Shopping Domain to Match Structured Data with Unstructured Data

Authors: Sachin Nagargoje

Abstract:

Complete labeled data is often difficult to obtain in a practical scenario. Even if one manages to obtain the data, the quality of the data is always in question. In shopping vertical, offers are the input data, which is given by advertiser with or without a good quality of information. In this paper, an author investigated the possibility of using a very simple Semi-supervised learning approach to increase the recall of unhealthy offers (has badly written Offer Title or partial product details) in shopping vertical domain. The author found that the semisupervised learning method had improved the recall in the Smart Phone category by 30% on A=B testing on 10% traffic and increased the YoY (Year over Year) number of impressions per month by 33% at production. This also made a significant increase in Revenue, but that cannot be publicly disclosed.

Keywords: Clustering, semi-supervised learning, coverage, recall

Procedia PDF Downloads 1
2583 Case Study about Women Driving in Saudi Arabia Announced in 2018: Netnographic and Data Mining Study

Authors: Majdah Alnefaie

Abstract:

The ‘netnographic study’ and data mining have been used to monitor the public interaction on Social Media Sites (SMSs) to understand what the motivational factors influence the Saudi intentions regarding allowing women driving in Saudi Arabia in 2018. The netnographic study monitored the publics’ textual and visual communications in Twitter, Snapchat, and YouTube. SMSs users’ communications method is also known as electronic word of mouth (eWOM). Netnography methodology is still in its initial stages as it depends on manual extraction, reading and classification of SMSs users text. On the other hand, data mining is come from the computer and physical sciences background, therefore it is much harder to extract meaning from unstructured qualitative data. In addition, the new development in data mining software does not support the Arabic text, especially local slang in Saudi Arabia. Therefore, collaborations between social and computer scientists such as ‘netnographic study’ and data mining will enhance the efficiency of this study methodology leading to comprehensive research outcome. The eWOM communications between individuals on SMSs can promote a sense that sharing their preferences and experiences regarding politics and social government regulations is a part of their daily life, highlighting the importance of using SMSs as assistance in promoting participation in political and social. Therefore, public interactions on SMSs are important tools to comprehend people’s intentions regarding the new government regulations in the country. This study aims to answer this question, "What factors influence the Saudi Arabians' intentions of Saudi female's car-driving in 2018". The study utilized qualitative method known as netnographic study. The study used R studio to collect and analyses 27000 Saudi users’ comments from 25th May until 25th June 2018. The study has developed data collection model that support importing and analysing the Arabic text in the local slang. The data collection model in this study has been clustered based on different type of social networks, gender and the study main factors. The social network analysis was employed to collect comments from SMSs owned by governments’ originations, celebrities, vloggers, social activist and news SMSs accounts. The comments were collected from both males and females SMSs users. The sentiment analysis shows that the total number of positive comments Saudi females car driving was higher than negative comments. The data have provided the most important factors influenced the Saudi Arabians’ intention of Saudi females car driving including, culture and environment, freedom of choice, equal opportunities, security and safety. The most interesting finding indicted that women driving would play a role in increasing the individual freedom of choice. Saudi female will be able to drive cars to fulfill her daily life and family needs without being stressed due to the lack of transportation. The study outcome will help Saudi government to improve woman quality of life by increasing the ability to find more jobs and studies, increasing income through decreasing the spending on transport means such as taxi and having more freedom of choice in woman daily life needs. The study enhances the importance of using use marketing research to measure the public opinions on the new government regulations in the country. The study has explained the limitations and suggestions for future research.

Keywords: Social Media, Data Mining, Saudi Arabia, netnographic study, female driving

Procedia PDF Downloads 1
2582 Visualization and Performance Measure to Determine Number of Topics in Twitter Data Clustering Using Hybrid Topic Modeling

Authors: Moulana Mohammed

Abstract:

Topic models are widely used in building clusters of documents for more than a decade, yet problems occurring in choosing optimal number of topics. The main problem is the lack of a stable metric of the quality of topics obtained during the construction of topic models. The authors analyzed from previous works, most of the models used in determining the number of topics are non-parametric and quality of topics determined by using perplexity and coherence measures and concluded that they are not applicable in solving this problem. In this paper, we used the parametric method, which is an extension of the traditional topic model with visual access tendency for visualization of the number of topics (clusters) to complement clustering and to choose optimal number of topics based on results of cluster validity indices. Developed hybrid topic models are demonstrated with different Twitter datasets on various topics in obtaining the optimal number of topics and in measuring the quality of clusters. The experimental results showed that the Visual Non-negative Matrix Factorization (VNMF) topic model performs well in determining the optimal number of topics with interactive visualization and in performance measure of the quality of clusters with validity indices.

Keywords: Interactive Visualization, visual mon-negative matrix factorization model, optimal number of topics, cluster validity indices, Twitter data clustering

Procedia PDF Downloads 1
2581 Camera Model Identification for Mi Pad 4, Oppo A37f, Samsung M20, and Oppo f9

Authors: Ulrich Wake, Eniman Syamsuddin

Abstract:

The model for camera model identificaiton is trained using pretrained model ResNet43 and ResNet50. The dataset consists of 500 photos of each phone. Dataset is divided into 1280 photos for training, 320 photos for validation and 400 photos for testing. The model is trained using One Cycle Policy Method and tested using Test-Time Augmentation. Furthermore, the model is trained for 50 epoch using regularization such as drop out and early stopping. The result is 90% accuracy for validation set and above 85% for Test-Time Augmentation using ResNet50. Every model is also trained by slightly updating the pretrained model’s weights

Keywords: ​ One Cycle Policy, ResNet34, ResNet50, Test-Time Agumentation

Procedia PDF Downloads 1
2580 Electronic Resources and Information Literacy in Higher Education Library

Authors: Rajesh Kumar, Nirmal Singh

Abstract:

Abstract- Information literacy aims to develop both critical understanding and active participation in scholars. It enables scholars to interpret and make informed judgments as users of information sources, and it also enables them to become producers of information in their own right, and thereby to become more powerful participants in society. Information literacy is about developing people‘s critical and creative abilities. Digital media – and particularly the Internet – significantly increase the potential for such active participation of the individual, provided scholars have the means and training to effectively access and use them. This paper provides definition, standards and importance of information literacy (IL). Keywords: Information literacy, Digital Media, Training, Communications Technologies.

Keywords: training, Digital Media, Information literacy, Communications Technologies

Procedia PDF Downloads 1
2579 Estimation of Noise Barriers for Arterial Roads of Delhi

Authors: Sourabh Jain, Parul Madan

Abstract:

Traffic noise pollution has become a challenging problem for all metro cities of India due to rapid urbanization, growing population and rising number of vehicles and transport development. In Delhi the prime source of noise pollution is vehicular traffic. In Delhi it is found that the ambient noise level (Leq) is exceeding the standard permissible value at all the locations. Noise barriers or enclosures are definitely useful in obtaining effective deduction of traffic noise disturbances in urbanized areas. US’s Federal Highway Administration Model (FHWA) and Calculation of Road Traffic Noise (CORTN) of UK are used to develop spread sheets for noise prediction. Spread sheets are also developed for evaluating effectiveness of existing boundary walls abutting houses in mitigating noise, redesigning them as noise barriers. Study was also carried out to examine the changes in noise level due to designed noise barrier by using both models FHWA and CORTN respectively. During the collection of various data it is found that receivers are located far away from road at Rithala and Moolchand sites and hence extra barrier height needed to meet prescribed limits was less as seen from calculations and most of the noise diminishes by propagation effect.On the basis of overall study and data analysis, it is concluded that FHWA and CORTN models under estimate noise levels. FHWA model predicted noise levels with an average percentage error of -7.33 and CORTN predicted with an average percentage error of -8.5. It was observed that at all sites noise levels at receivers were exceeding the standard limit of 55 dB. It was seen from calculations that existing walls are reducing noise levels. Average noise reduction due to walls at Rithala was 7.41 dB and at Panchsheel was 7.20 dB and lower amount of noise reduction was observed at Friend colony which was only 5.88. It was observed from analysis that Friends colony sites need much greater height of barrier. This was because of residential buildings abutting the road. At friends colony great amount of traffic was observed since it is national highway. At this site diminishing of noise due to propagation effect was very less.As FHWA and CORTN models were developed in excel programme, it eliminates laborious calculations of noise. There was no reflection correction in FHWA models as like in CORTN model.

Keywords: IFHWA, CORTN, Noise Sources, Noise Barriers

Procedia PDF Downloads 1
2578 Amharic Text News Classification Using Supervised Learning

Authors: Misrak Assefa

Abstract:

The Amharic language is the second most widely spoken Semitic language in the world. There are several new overloaded on the web. Searching some useful documents from the web on a specific topic, which is written in the Amharic language, is a challenging task. Hence, document categorization is required for managing and filtering important information. In the classification of Amharic text news, there is still a gap in the domain of information that needs to be launch. This study attempts to design an automatic Amharic news classification using a supervised learning mechanism on four un-touch classes. To achieve this research, 4,182 news articles were used. Naive Bayes (NB) and Decision tree (j48) algorithms were used to classify the given Amharic dataset. In this paper, k-fold cross-validation is used to estimate the accuracy of the classifier. As a result, it shows those algorithms can be applicable in Amharic news categorization. The best average accuracy result is achieved by j48 decision tree and naïve Bayes is 95.2345 %, and 94.6245 % respectively using three categories. This research indicated that a typical decision tree algorithm is more applicable to Amharic news categorization.

Keywords: Text categorization, Decision Tree, naive Bayes, supervised machine learning

Procedia PDF Downloads 1
2577 Early Gastric Cancer Prediction from Diet and Epidemiological Data Using Machine Learning in Mizoram Population

Authors: Brindha Senthil Kumar, Payel Chakraborty, Senthil Kumar Nachimuthu, Arindam Maitra, Prem Nath

Abstract:

Gastric cancer is predominantly caused by demographic and diet factors as compared to other cancer types. The aim of the study is to predict Early Gastric Cancer (ECG) from diet and lifestyle factors using supervised machine learning algorithms. For this study, 160 healthy individual and 80 cases were selected who had been followed for 3 years (2016-2019), at Civil Hospital, Aizawl, Mizoram. A dataset containing 11 features that are core risk factors for the gastric cancer were extracted. Supervised machine algorithms: Logistic Regression, Naive Bayes, Support Vector Machine (SVM), Multilayer perceptron, and Random Forest were used to analyze the dataset using Python Jupyter Notebook Version 3. The obtained classified results had been evaluated using metrics parameters: minimum_false_positives, brier_score, accuracy, precision, recall, F1_score, and Receiver Operating Characteristics (ROC) curve. Data analysis results showed Naive Bayes - 88, 0.11; Random Forest - 83, 0.16; SVM - 77, 0.22; Logistic Regression - 75, 0.25 and Multilayer perceptron - 72, 0.27 with respect to accuracy and brier_score in percent. Naive Bayes algorithm out performs with very low false positive rates as well as brier_score and good accuracy. Naive Bayes algorithm classification results in predicting ECG showed very satisfactory results using only diet cum lifestyle factors which will be very helpful for the physicians to educate the patients and public, thereby mortality of gastric cancer can be reduced/avoided with this knowledge mining work.

Keywords: Machine Learning, diet, Early Gastric cancer, Lifestyle Characteristics

Procedia PDF Downloads 1
2576 Fairness in Recommendations Ranking: From Pairwise Approach to Listwise Approach

Authors: Patik Joslin Kenfack, Polyakov Vladimir Mikhailovich

Abstract:

Machine Learning (ML) systems are trained using human generated data that could be biased by implicitly containing racist, sexist, or discriminating data. ML models learn those biases or even amplify them. Recent research in work on has begun to consider issues of fairness. The concept of fairness is extended to recommendation. A recommender system will be considered fair if it doesn’t under rank items of protected group (gender, race, demographic...). Several metrics for evaluating fairness concerns in recommendation systems have been proposed, which take pairs of items as ‘instances’ in fairness evaluation. It doesn’t take in account the fact that the fairness should be evaluated across a list of items. The paper explores a probabilistic approach that generalize pairwise metric by using a list k (listwise) of items as ‘instances’ in fairness evaluation, parametrized by k. We also explore new regularization method based on this metric to improve fairness ranking during model training.

Keywords: Recommender System, ranking, fairness, Listwise Approach

Procedia PDF Downloads 1
2575 Multi-Class Text Classification Using Ensembles of Classifiers

Authors: Syed Basit Ali Shah Bukhari, Yan Qiang, Saad Abdul Rauf, Syed Saqlaina Bukhari

Abstract:

Text Classification is the methodology to classify any given text into the respective category from a given set of categories. It is highly important and vital to use proper set of pre-processing , feature selection and classification techniques to achieve this purpose. In this paper we have used different ensemble techniques along with variance in feature selection parameters to see the change in overall accuracy of the result and also on some other individual class based features which include precision value of each individual category of the text. After subjecting our data through pre-processing and feature selection techniques , different individual classifiers were tested first and after that classifiers were combined to form ensembles to increase their accuracy. Later we also studied the impact of decreasing the classification categories on over all accuracy of data. Text classification is highly used in sentiment analysis on social media sites such as twitter for realizing people’s opinions about any cause or it is also used to analyze customer’s reviews about certain products or services. Opinion mining is a vital task in data mining and text categorization is a back-bone to opinion mining.

Keywords: natural language processing, adaboost, Ensemble Classifier, Bagging Classifier

Procedia PDF Downloads 1
2574 Detecting Manipulated Media Using Deep Capsule Network

Authors: Joseph Uzuazomaro Oju

Abstract:

The ease at which manipulated media can be created, and the increasing difficulty in identifying fake media makes it a great threat. Most of the applications used for the creation of these high-quality fake videos and images are built with deep learning. Hence, the use of deep learning in creating a detection mechanism cannot be overemphasized. Any successful fake media that is being detected before it reached the populace will save people from the self-doubt of either a content is genuine or fake and will ensure the credibility of videos and images. The methodology introduced in this paper approaches the manipulated media detection challenge using a combo of VGG-19 and a deep capsule network. In the case of videos, they are converted into frames, which, in turn, are resized and cropped to the face region. These preprocessed images/videos are fed to the VGG-19 network to extract the latent features. The extracted latent features are inputted into a deep capsule network enhanced with a 3D -convolution dynamic routing agreement. The 3D –convolution dynamic routing agreement algorithm helps to reduce the linkages between capsules networks. Thereby limiting the poor learning shortcoming of multiple capsule network layers. The resultant output from the deep capsule network will indicate a media to be either genuine or fake.

Keywords: dynamic routing, deep capsule network, fake media detection, manipulated media

Procedia PDF Downloads 1
2573 An Improved Convolution Deep Learning Model for Predicting Trip Mode Scheduling

Authors: Amin Nezarat, Naeime Seifadini

Abstract:

Trip mode selection is a behavioral characteristic of passengers with immense importance for travel demand analysis, transportation planning, and traffic management. Identification of trip mode distribution will allow transportation authorities to adopt appropriate strategies to reduce travel time, traffic and air pollution. The majority of existing trip mode inference models operate based on human selected features and traditional machine learning algorithms. However, human selected features are sensitive to changes in traffic and environmental conditions and susceptible to personal biases, which can make them inefficient. One way to overcome these problems is to use neural networks capable of extracting high-level features from raw input. In this study, the convolutional neural network (CNN) architecture is used to predict the trip mode distribution based on raw GPS trajectory data. The key innovation of this paper is the design of the layout of the input layer of CNN as well as normalization operation, in a way that is not only compatible with the CNN architecture but can also represent the fundamental features of motion including speed, acceleration, jerk, and Bearing rate. The highest prediction accuracy achieved with the proposed configuration for the convolutional neural network with batch normalization is 85.26%.

Keywords: Neural Network, Deep learning, Predicting, urban trip

Procedia PDF Downloads 1
2572 An Information Quality Framework for University Website

Authors: Joseph Elliot, Daniel Berleant

Abstract:

Prospective students use university websites as the primary information resource when choosing universities to apply to for admission. Hence, the Information Quality (IQ) of a university website should play a major role in the decision process for prospective students when selecting a university for their higher education. In this first part of a two-part series, we identify university website Information Quality Dimensions relevant for a prospective student when deciding whether or not to apply or enroll at a university. We discuss the rationales for identifying these IQ dimensions and propose a Website Information Quality (WebIQ) Framework to quantify the individual IQ dimensions as well as the strategy for specifying a composite IQ for such a website. The outcome of this research could provide insight into a university when planning their student recruitment strategy.

Keywords: world wide web, Data Quality, Information Quality, dimensional IQ, composite IQ

Procedia PDF Downloads 1
2571 Intelligent Recognition Tools for Industrial Automation

Authors: Amin Nazerzadeh, Afsaneh Nouri Houshyar, Azadeh Noori Hoshyar

Abstract:

With the rapid growing of information technology, the industry and manufacturing systems are becoming more automated. Therefore, achieving the highly accurate automatic systems with reliable security is becoming more critical. Biometrics that refers to identifying individual based on physiological or behavioral traits are unique identifiers provide high reliability and security in different industrial systems. As biometric cannot easily be transferred between individuals or copied, it has been receiving extensive attention. Due to the importance of security applications, this paper provides an overview on biometrics and discuss about background, types and applications of biometric as an effective tool for the industrial applications.

Keywords: Information Technology, Industial and manufacturing applications, intelligence and security, recognition; security technology; biometrics

Procedia PDF Downloads 1
2570 Organization Culture: Mediator of Information Technology Competence and IT Governance Effectiveness

Authors: Sonny Nyeko, Moses Niwe

Abstract:

Purpose: This research paper examined the mediation effect of organization culture in the relationship between information technology (IT) competence and IT governance effectiveness in Ugandan public universities. The purpose of the research paper is to examine the role of organizational culture in the relationship between IT competence and IT governance effectiveness. Design/methodology/approach: The paper adopted the MedGraph program, Sobel tests and Kenny and Baron Approach for testing the mediation effects. Findings: It is impeccable that IT competence and organization culture are true drivers of IT governance effectiveness in Ugandan public universities. However, organizational culture reveals partial mediation in the IT competence and IT governance effectiveness relationship. Research limitations/implications: The empirical investigation in this research depends profoundly on public universities. Future research in Ugandan private universities could be undertaken to compare results. Practical implications: To effectively achieve IT governance effectiveness, it means senior management requires IT knowledge which is a vital ingredient of IT competence. Moreover, organizations today ought to adopt cultures that are intended to have them competitive in their businesses, with IT operations not in isolation. Originality/value: Spending thousands of dollars on IT resources in advanced institutes of learning necessitates IT control. Preliminary studies in Ugandan public universities have revealed the ineffective utilization of IT resources. Besides, IT governance issues with IT competence and organization culture remain outstanding. Thus, it’s a new study testing the mediating outcome of organization culture in the association between IT competence and IT governance effectiveness in the Ugandan universities.

Keywords: Universities, Effectiveness, IT governance, Uganda, organization culture, mediating effect, IT competence

Procedia PDF Downloads 1
2569 Kalman Filter Gain Elimination

Authors: Nicholas Assimakis

Abstract:

The conventional Kalman filter uses, in every iteration, the Kalman filter gain in order to produce estimation and prediction of the n-dimensional state vector using the m-dimensional measurement vector. The computation of the Kalman filter gain requires the inversion of an mxm matrix in every iteration. In this paper, a variation of the Kalman filter eliminating the Kalman filter gain is proposed. In the time-varying case, the elimination of the Kalman filter gain requires the inversion of an nxn matrix and the inversion of an mxm matrix in every iteration. In the time invariant case, the elimination of the Kalman filter gain requires the inversion of an nxn matrix in every iteration. Hence, the proposed Kalman filter gain elimination algorithm may be faster than the conventional Kalman filter, depending on the model dimensions.

Keywords: estimation, Kalman Filter, discrete time, Kalman filter gain

Procedia PDF Downloads 1
2568 Project Management Process Implementation in Data Centers

Authors: F. Nasiri, M. Fadaeefath Abadi, F. Haghighat

Abstract:

Data Centers (DCs) are one of the most critical infrastructures because of their complexity, having important components and providing essential online services to the world. Many Information Technology (IT) companies, governmental departments, and agencies around the world must have reliable DCs to protect and maintain significant amounts of data for their users and customers. Project Management, which manages every process in a system is also highly involved in a DC design project and should be addressed and considered. DC Project Management's main elements are planning, scheduling and providing safety and security for the DC infrastructure. It is also responsible for managing various operations and assets such as telecommunications networks, data storage, and processing systems and available equipment, which are servers, network switches, etc. In this paper, the application of Project Management in DCs has been evaluated considering important quality standards. The main parameters and facts based on recent research works and advancements have been presented and discussed to identify issues, problems, and challenges in terms of applying Project Management standards in DCs. This study will also allow better clarification and understanding of DC Project Management metrics to assist DC project managers and practitioners in performance monitoring.

Keywords: Project Management, data center (DC), information technology (IT) infrastructure

Procedia PDF Downloads 1