Computer Engineering Commons^™

Detecting And Correcting Automatic Speech Recognition Errors With A New Model, Recep Si̇nan Arslan, Necaatti̇n Barişçi, Nursal Arici, Sabri̇ Koçer

Turkish Journal of Electrical Engineering and Computer Sciences

The purpose of automatic speech recognition (ASR) systems is to recognize speech signals obtained from people and convert them into text so that they can be processed by a computer. Although many ASR applications are versatile and widely used in the real world, they still generate relatively inaccurate results. They tend to generate spelling errors in recognized words, especially in noisy environments, in situations where the vocabulary size is increased, and at times when the input speech is of poor quality. The permanent presence of errors in ASR systems has led to the need to find alternative methods for automatic …

Exploring The Attention Process Differentiation Of Attention Deficit Hyperactivity Disorder (Adhd) Symptomatic Adults Using Artificial Intelligence Onelectroencephalography (Eeg) Signals, Gökhan Güney, Esra Kisacik, Canan Kalaycioğlu, Görkem Saygili

Turkish Journal of Electrical Engineering and Computer Sciences

Attention deficit and hyperactivity disorder (ADHD) onset in childhood and its symptoms can last up till adulthood. Recently, electroencephalography (EEG) has emerged as a tool to investigate the neurophysiological connection of ADHD and the brain. In this study, we investigated the differentiation of attention process of healthy subjects with or without ADHD symptoms under visual continuous performance test (VCPT). In our experiments, artificial neural network (ANN) algorithm achieved 98.4% classification accuracy with 0.98 sensitivity when P2 event related potential (ERP) was used. Additionally, our experimental results showed that fronto-central channels were the most contributing. Overall, we conclude that the attention …

Privacy Preserving Hybrid Recommender System Based On Deep Learning, Sangeetha Selvaraj, Sudha Sadasivam Gangadharan

Turkish Journal of Electrical Engineering and Computer Sciences

Deep learning models are widely being used to provide relevant recommendations in hybrid recommender systems. These hybrid systems combine the advantages of both content based and collaborative filtering approaches. However, these learning systems hamper the user privacy and disclose sensitive information. This paper proposes a privacy preserving deep learning based hybrid recommender system. In hybrid deep neural network, user?s side information such as age, location, occupation, zip code along with user rating is embedded and provided as input. These embedding?s pose a severe threat to individual privacy. In order to eliminate this breach of privacy, we have proposed a private …

Performance Evaluation Of Hht And Wt For Detection Of Hif And Ct Saturationin Smart Grids, Saeid Heidari, Saeed Asgharigovar, Pouya Pourghasem, Heresh Seyedi, Ömer Usta

Turkish Journal of Electrical Engineering and Computer Sciences

Hilbert-Huang transform (HHT), continuous wavelet transform (CWT) and discrete wavelet transform (DWT) are well-known signal processing methods that are widely utilized for feature extraction and fault detection by protection systems in smart grids. In this paper, we assess the performances of these methods encountering challenging situations in distribution networks, i.e. high impedance arcing fault (HIF) and current transformer (CT) saturation. Low fault current amplitude in HIF case causes the overcurrent protection, which is the predominant protection method in distribution grids, to fail. Furthermore, some faults may lead to CT saturation, which may result in delayed operation of the relay. To …

Prediction Of Long-Term Physical Properties Of Low Density Polyethylene (Ldpe)Cable Insulation Materials By Artificial Neural Network Modeling Approach Underenvironmental Constraints, Ferhat Slimani, Abdallah Hedir, Mustapha Moudoud, Ali̇ Durmuş, Mounir Amir, Mohamed Megherbi

Turkish Journal of Electrical Engineering and Computer Sciences

This study quantifies long-term physical properties of low density polyethylene (LDPE) cables insulations exposed to environmental constraints such as UV radiation and temperature via both experimental measurements and mathematical modeling approach. For this purpose, tensile test and electrical breakdown test were carried out to determine elongation at break, tensile strength, and dielectric strength of unaged and aged specimens, respectively. Experimental results showed that both UV and temperature exposures affected the LDPE properties, significantly. A supervised artificial neural network (ANN) trained by the Levenberg?Marquardt algorithm was designed for predicting the long-term characteristics of specimens and also for minimizing the experimental procedures. …

Real-Time Motion Tracking Enhancement Via Data-Fusion Based Particle Filter, Tuğrul Taşci, Numan Çelebi̇

Turkish Journal of Electrical Engineering and Computer Sciences

Motion tracking is a well-defined yet application-specific problem of computer vision field, mostly entailing real-time constraints. Methods addressing such problems are expected also to ensure achievements such as high accuracy and robustness. A probabilistic estimation-based approach is proposed in this paper, in order to enhance the real-time motion tracking process of an RGB-Depth device, in terms of accuracy. A novel method is presented for tracking handpalm of a moving human subject to this end, under a sequence of assumptions such as indoor environment, single object, smooth movement and stable illumination. Tracking accuracy is improved within a particle filter framework by …

A New Distributed Anomaly Detection Approach For Log Ids Management Based Ondeep Learning, Murat Koca, Muhammed Ali̇ Aydin, Ahmet Sertbaş, Abdül Hali̇m Zai̇m

Turkish Journal of Electrical Engineering and Computer Sciences

Today, with the rapid increase of data, the security of big data has become more important than ever for managers. However, traditional infrastructure systems cannot cope with increasingly big data that is created like an avalanche. In addition, as the existing database systems increase licensing costs per transaction, organizations using information technologies are shifting to free and open source solutions. For this reason, we propose an anomaly attack detection model on Apache Hadoop distributed file system (HDFS), which stands out in open source big data analytics, and Apache Spark, which stands out with its speed performance in analysis to reduce …

On The Closed-Form Evaluation Of The Po Integral Using The Radon Transforminterpretation For Linear Triangles, Aslihan Aktepe, Hüseyi̇n Arda Ülkü

Turkish Journal of Electrical Engineering and Computer Sciences

This letter presents the complete mathematical formulation for the closed-form evaluation of the time domain physical optics (PO) integral on linear triangular patches using Radon transform (RT) interpretation. The incident field is assumed to be an impulsively excited plane wave and scattered fields are observed at far-zone. The PO integral is evaluated in closed-form as the intersection of the triangle and the plane formed by the incident and observation directions. In addition, a formula is suggested for the special case, which occurs if there is no intersection of the plane and all scatterer. Accuracy of the closed-form expressions is demonstrated …

A New Model For Minimizing The Electric Vehicle Battery Capacity In Electrictravelling Salesman Problem With Time Windows, Kazim Erdoğdu, Korhan Karabulut

Turkish Journal of Electrical Engineering and Computer Sciences

The growing pollution in the environment and the negative shift in the global climate compel authorities to take action to protect the environment and human health. Transportation is one of the major contributors to this environmental decay. The harmful gases released to the air by the vehicles using petroleum fuel increase each day. One of the solutions is to make a gradual transition to electric vehicles. A major part of manufacturing an electric vehicle is to produce an efficient electric motor and battery for it. Reducing the manufacturing and operating costs of these components will result in reducing the overall …

Multiagent Q-Learning Based Uav Trajectory Planning For Effective Situationalawareness, Erdal Akin, Kubi̇lay Demi̇r, Hali̇l Yetgi̇n

Turkish Journal of Electrical Engineering and Computer Sciences

In the event of a natural disaster, arrival time of the search and rescue (SAR) teams to the affected areas is of vital importance to save the life of the victims. In particular, when an earthquake occurs in a geographically large area, reconnaissance of the debris within a short-time is critical for conducting successful SAR missions. An effective and quick situational awareness in postdisaster scenarios can be provided via the help of unmanned aerial vehicles (UAVs). However, off-the-shelf UAVs suffer from the limited communication range as well as the limited airborne duration due to battery constraints. If telecommunication infrastructure is …

Learning Prototypes For Multiple Instance Learning, Özgür Emre Si̇vri̇kaya, Mert Yüksekgönül, Mustafa Gökçe Baydoğan

Turkish Journal of Electrical Engineering and Computer Sciences

Multiple instance learning (MIL) is a weakly supervised learning method that works on the labeled bag of instances data. A prototypical network is a popular embedding approach in MIL. They overcome the common problems that other MIL approaches may have to deal with including dimensionality, loss of instance-level information, and complexity. They demonstrate competitive performance in classification. This work proposes a simple model that provides a permutation invariant prototype generator from a given MIL data set. We aim to find out prototypes in the feature space to map the collection of instances (i.e. bags) to a distance feature space and …

Automated Waterloo Rubric For Concept Map Grading, Shresht Bhatia, Sajal Bhatia, Irfan Ahmed

School of Computer Science & Engineering Faculty Publications

Concept mapping is a well-known pedagogical tool to help students organize, represent, and develop an understanding of a topic. The grading of concept maps is typically manual, time-consuming, and tedious, especially for a large class. Existing research mostly focuses on topological scoring based-on structural features of concept maps. However, the scoring does not achieve comparable accuracy to well-defined rubrics for manual analysis on the quality of content in a concept map. This paper presents Kastor, a new method to automate the Waterloo Rubric of scoring concept maps by quantifying the rubric’s quality assessment parameters. The evaluation is performed on a …

Cybersecurity Analysis Of Load Frequency Control In Power Systems: A Survey, Sahaj Saxena, Sajal Bhatia, Rahul Gupta

School of Computer Science & Engineering Faculty Publications

Today, power systems have transformed considerably and taken a new shape of geographically distributed systems from the locally centralized systems thereby leading to a new infrastructure in the framework of networked control cyber-physical system (CPS). Among the different important operations to be performed for smooth generation, transmission, and distribution of power, maintaining the scheduled frequency, against any perturbations, is an important one. The load frequency control (LFC) operation actually governs this frequency regulation activity after the primary control. Due to CPS nature, the LFC operation is vulnerable to attacks, both from physical and cyber standpoints. The cyber-attack strategies ranges from …

Improving A Network Intrusion Detection System’S Efficiency Using Model-Based Data Augmentation, Vinicius Waterkemper Lodetti

Dissertations

A network intrusion detection system (NIDS) is one important element to mitigate cybersecurity risks, the NIDS allow for detecting anomalies in a network which may be a cyberattack to a corporate network environment. A NIDS can be seen as a classification problem where the ultimate goal is to distinguish between malicious traffic among a majority of benign traffic. Researches on NIDS are often performed using outdated datasets that don’t represent the actual cyberspace. Datasets such as the CICIDS2018 address this gap by being generated from attacks and an infrastructure that reflects an up-to-date scenario.

A problem may arise when machine …

An Evaluation On The Performance Of Code Generated With Webassembly Compilers, Raymond Phelan

Dissertations

WebAssembly is a new technology that is revolutionizing the web. Essentially it is a low-level binary instruction set that can be run on browsers, servers or stand-alone environments. Many programming languages either currently have, or are working on, compilers that will compile the language into WebAssembly. This means that applications written in languages like C++ or Rust can now be run on the web, directly in a browser or other environment. However, as we will highlight in this research, the quality of code generated by the different WebAssembly compilers varies and causes performance issues. This research paper aims to evaluate …

A Hybrid Neural Network For Stock Price Direction Forecasting, Daniel Devine

Dissertations

The volatility of stock markets makes them notoriously difficult to predict and is the reason that many investors sell out at the wrong time. Contrary to the efficient market hypothesis (EMH) and the random walk theory, contribution to the study of machine learning models for stock price forecasting has shown evidence of stock markets predictability with varying degrees of success. Contemporary approaches have sought to use a hybrid of convolutional neural network (CNN) for its feature extraction capabilities and long short-term memory (LSTM) neural network for its time series prediction. This comparative study aims to determine the predictability of stock …

Identifying Significant Features For Player Evaluation In Nfl Comparing Anns And Traditional Models, Ronan Walsh

Dissertations

The evaluation of player performance in sports is popular and important in modern sports, enabling teams to use real data in the construction of their rosters. This dissertation proposes to apply machine learning algorithms to predicting the player evaluations from a leading NFL analytics company who use a combination of statistics and expert evaluation. In addition, it will investigate what features are significant in the evaluation of a position. Data for the dissertation is obtained from multiple online sources - Pro Football Reference and Pro Football Focus (the the NFL analytics company). These data sets are combined and analysed before …

Identifying Roles Of Software Developers From Their Answers On Stack Overflow, Dean Power

Dissertations

Stack Overflow is the world’s largest community of software developers. Users ask and answer questions on various tagged topics of software development. The set of questions a site user answers is representative of their knowledge base, or “wheelhouse”. It is proposed that clustering users by their wheelhouse yields communities of similar software developers by skill-set. These communities represent the different roles within software development and could be used as the basis to define roles at any point in time in an ever-evolving landscape of software development. A network graph of site users, linked if they answered questions on the same …

Can Generative Adversarial Networks Help Us Fight Financial Fraud?, Sean Mciver

Dissertations

Transactional fraud datasets exhibit extreme class imbalance. Learners cannot make accurate generalizations without sufficient data. Researchers can account for imbalance at the data level, algorithmic level or both. This paper focuses on techniques at the data level. We evaluate the evidence of the optimal technique and potential enhancements. Global fraud losses totalled more than 80 % of the UK’s GDP in 2019. The improvement of preprocessing is inherently valuable in fighting these losses. Synthetic minority oversampling technique (SMOTE) and extensions of SMOTE are currently the most common preprocessing strategies. SMOTE oversamples the minority classes by randomly generating a point between …

Performance Comparison Between A Distributed Particle Swarm Algorithm And A Centralised Algorithm, Ciarán O’Loughlin

Dissertations

Particle Swarm optimisation (PSO) is a particular form of swarm intelligence, which itself is an innovative intelligent paradigm for solving optimization problems. PSO is generally used to find a global optimum in a single optimisation function. This typically occurs on one node(machine) but there has been a significant body of research into creating distributed implementations of the PSO algorithm. Such research has often focused on the creation and performance of the distributed implementation in an isolated manner or compared to different distributed algorithms.

This research piece aims to bridge a gap in the existing literature, by testing a distributed implementation …

Voice Impersonation For Thai Speech Using Cyclegan Over Prosody, Chatri Chuanngulueam

Chulalongkorn University Theses and Dissertations (Chula ETD)

No abstract provided.

Integrating The First Person View And The Third Person View Using A Connected Vr-Mr System For Pilot Training, Chang-Geun Oh, Kwanghee Lee, Myunghoon Oh

Journal of Aviation/Aerospace Education & Research

Virtual reality (VR)-based flight simulator provides pilots the enhanced reality from the first-person view. Mixed reality (MR) technology generates effective 3D graphics. The users who wear the MR headset can walk around the 3D graphics to see all its 360 degrees of vertical and horizontal aspects maintaining the consciousness of real space. A VR flight simulator and an MR application were connected to create the capability of both first-person view and third-person view for a comprehensive pilot training system. This system provided users the capability to monitor the aircraft progress along the planned path from the third-person view as well …

Converting Optical Videos To Infrared Videos Using Attention Gan And Its Impact On Target Detection And Classification Performance, Mohammad Shahab Uddin, Reshad Hoque, Kazi Aminul Islam, Chiman Kwan, David Gribben, Jiang Li

Electrical & Computer Engineering Faculty Publications

To apply powerful deep-learning-based algorithms for object detection and classification in infrared videos, it is necessary to have more training data in order to build high-performance models. However, in many surveillance applications, one can have a lot more optical videos than infrared videos. This lack of IR video datasets can be mitigated if optical-to-infrared video conversion is possible. In this paper, we present a new approach for converting optical videos to infrared videos using deep learning. The basic idea is to focus on target areas using attention generative adversarial network (attention GAN), which will preserve the fidelity of target areas. …

Semantics Of The Black-Box: Can Knowledge Graphs Help Make Deep Learning Systems More Interpretable And Explainable?, Manas Gaur, Keyur Faldu, Amit Sheth

Publications

The recent series of innovations in deep learning (DL) have shown enormous potential to impact individuals and society, both positively and negatively. The DL models utilizing massive computing power and enormous datasets have significantly outperformed prior historical benchmarks on increasingly difficult, well-defined research tasks across technology domains such as computer vision, natural language processing, signal processing, and human-computer interactions. However, the Black-Box nature of DL models and their over-reliance on massive amounts of data condensed into labels and dense representations poses challenges for interpretability and explainability of the system. Furthermore, DLs have not yet been proven in their ability to …

Green Underwater Wireless Communications Using Hybrid Optical-Acoustic Technologies, Kazi Y. Islam, Iftekhar Ahmad, Daryoush Habibi, M. Ishtiaque A. Zahed, Joarder Kamruzzaman

Research outputs 2014 to 2021

Underwater wireless communication is a rapidly growing field, especially with the recent emergence of technologies such as autonomous underwater vehicles (AUVs) and remotely operated vehicles (ROVs). To support the high-bandwidth applications using these technologies, underwater optics has attracted significant attention, alongside its complementary technology – underwater acoustics. In this paper, we propose a hybrid opto-acoustic underwater wireless communication model that reduces network power consumption and supports high-data rate underwater applications by selecting appropriate communication links in response to varying traffic loads and dynamic weather conditions. Underwater optics offers high data rates and consumes less power. However, due to the severe …

Visualization For Solving Non-Image Problems And Saliency Mapping, Divya Chandrika Kalla

All Master's Theses

High-dimensional data play an important role in knowledge discovery and data science. Integration of visualization, visual analytics, machine learning (ML), and data mining (DM) are the key aspects of data science research for high-dimensional data. This thesis is to explore the efficiency of a new algorithm to convert non-images data into raster images by visualizing data using heatmap in the collocated paired coordinates (CPC). These images are called the CPC-R images and the algorithm that produces them is called the CPC-R algorithm. Powerful deep learning methods open an opportunity to solve non-image ML/DM problems by transforming non-image ML problems into …

Ieee Access Special Section Editorial: Trends And Advances In Bio-Inspired Image-Based Deep Learning Methodologies And Applications, Peter Peer, Carlos M. Travieso-Gonzalez, Vijayan K. Asari, Malay Kishore Dutta

Electrical and Computer Engineering Faculty Publications

Many of the technological advances we enjoy today have been inspired by biological systems due to their ease of operation and outstanding efficiency. Designing technological solutions based on biological inspiration has become a cornerstone of research in a variety of areas ranging from control theory and optimization to computer vision, machine learning, and artificial intelligence. Especially in the latter few areas, biologically relevant solutions are becoming increasingly important as we look for new ways to make artificial systems more efficient, intelligent, and overall effective.

On The Mandelbrot Set For I**2 = ±1 And Imaginary Higgs Fields, Jonathan Blackledge

Articles

We consider the consequence of breaking with a fundamental result in complex analysisby lettingi2=±1wherei=√−1is the basic unit of all imaginary numbers. An analysis of theMandelbrot set for this case shows that a demarcation between a Fractal and a Euclidean object ispossible based oni2=−1andi2= +1, respectively. Further, we consider the transient behaviourassociated with the two cases to produce a range of non-standard sets in which a Fractal geometricstructure is transformed into a Euclidean object. In the case of the Mandelbrot set, the Euclideanobject is a square whose properties are investigate. Coupled with the associated Julia sets and othercomplex plane mappings, this …

การพยากรณ์ปริมาณและความเข้มข้นสารฟลอกคูแลต์ในกระบวนการพักใสสำหรับอุตสาหกรรมการผลิตน้ำตาลจากอ้อย, สิงหดิศร์ จันทรักษ์

Chulalongkorn University Theses and Dissertations (Chula ETD)

กระบวนการพักใสเป็นกระบวนการที่สำคัญในกระบวนการผลิตน้ำตาล ซึ่งกระบวนการมีการทำงานเพื่อแยกระหว่างตะกอนกับน้ำอ้อยออกจากกันโดยใช้สารฟลอกคูแลนต์ โดยในการใส่ปริมาณและความเข้มข้นสารฟลอกคูแลนต์ลงไปในน้ำอ้อยทำให้ส่งผลกระทบโดยตรงต่อความเร็วการตกตะกอนและค่าความขุ่นของน้ำอ้อย วิทยานิพนธ์เล่มนี้เสนอวิธีการพยากรณ์ปริมาณและความเข้มข้นสารฟลอกคูแลต์ในกระบวนการพักใสสำหรับอุตสาหกรรมการผลิตน้ำตาลจากอ้อย โดยใช้โครงข่ายประสาทเทียมแบบวนซ้ำชนิดพิเศษ Long Short-Term Memory โดยข้อมูลที่นำมาใช้เป็นข้อมูลขาเข้าสำหรับการสร้างโมเดลได้แก่ ปริมาณอ้อยสด, ปริมาณอ้อยเผา, ความขุ่นของน้ำอ้อย และปริมาณน้ำฝน และข้อมูลขาออกได้แก่ ปริมาณและความเข้มข้นของสารฟลอกคูแลนต์ ทั้งนี้ข้อมูลที่ได้นำมาจากโรงงานผลิตน้ำตาลแห่งหนึ่งในประเทศไทย ผลการทดลองแสดงให้เห็นถึงประสิทธิภาพของโมเดลที่ได้นำเสนอ LSTM โดยการเปรียบเทียบกับโมเดลอื่นๆ ได้แก่ Autoregressive Integrated Moving Average (ARIMA), Recurrent Neural Network (RNN) และ Gated Recurrent Unit (GRU) โดยใช้ตัวแปร RMSE และ MAPE เป็นตัววัดประสิทธิภาพของโมเดล พบว่าโมเดลที่นำเสนอมีประสิทธิภาพที่สุดในการพยากรณ์ปริมาณและความเข้มข้นของสารฟลอกคูแลนต์

การจำแนกปัญหาของเทคโนโลยีฐานข้อมูลในชุมชนถามตอบออนไลน์, ณัฐนัย สุวรรณชูชิต

Chulalongkorn University Theses and Dissertations (Chula ETD)

วิทยานิพนธ์นี้นำเสนอแนวทางการสร้างเครื่องมือการทำงานอัตโนมัติเพื่อจำแนกคำถามบนเว็บไซต์สแต็กโอเวอร์โฟลว์ โดยเฉพาะที่เกี่ยวกับชนิดของผลิตภัณฑ์ฐานข้อมูล ซึ่งถือเป็นข้อมูลที่มีค่าสำหรับเจ้าของผลิตภัณฑ์ฐานข้อมูลในการนำไปปรับปรุงผลิตภัณฑ์ หมวดหมู่ของคำถามกำหนดไว้เป็นสองระดับได้แก่ ระดับปัญหา และ ปัญหาย่อย โดยที่ระดับปัญหาประกอบด้วย การพัฒนา การติดตั้ง และ การปรับปรุงประสิทธิภาพ ในขณะที่ ปัญหาย่อย ประกอบด้วย การออกแบบ ข้อจำกัด และการอภิปรายปัญหา ด้วยการรวมทั้งสองระดับเข้าด้วยกัน คำถามจะถูกจำแนกออกเป็นเก้าหมวดของปัญหา-ปัญหาย่อย การประมวลผลภาษาธรรมชาติและการจำแนกข้อความถูกนำมาใช้ โดยใช้อัลกอริทึมการเรียนรู้ของเครื่องที่หลากหลาย โมเดลการจำแนกประเภทที่มีประสิทธิภาพดีที่สุดจะถูกนำมาใช้ในเว็บแอปพลิเคชัน เพื่อจำแนกแต่ละคำถามโดยใช้แท็กปัญหา-ปัญหาย่อย นอกจากนี้คำถามที่ถูกจำแนกออกตามหมวดแล้ว สามารถนำมาวิเคราะห์เพิ่มเติมโดยใช้อัลกอริทึมการสร้างแบบจำลองหัวข้อ เพื่อให้ทราบว่าคำถามในแต่ละหมวดนั้นกล่าวถึงหัวข้อใดบ้าง ซึ่งจะเป็นข้อมูลเพิ่มเติมให้กับเจ้าของผลิตภัณฑ์ฐานข้อมูลในการทำความเข้าใจถึงปัญหาของผลิตภัณฑ์เพื่อจะได้ทำการปรับปรุงต่อไป