Computer Engineering Commons^™

Improving Question Answering Retrieval System Through Multi-Result Ranking Model, Danupat Khamnuansin

Chulalongkorn University Theses and Dissertations (Chula ETD)

Recent trends in various industries involve integrating artificial intelligence (AI) systems to enhance operational efficiency. Among these advancements, AI-supported question-answering (QA) systems have gained significant attention. These systems typically employ a two-stage process, integrating QA system capabilities with Information Retrieval (IR) methods. The introduction of the Retrieval Question Answering (ReQA) has further refined this process, offering a more practical solution, aiming to improve real-world applicability. Considering the wide availability of diverse QA retrieval models, employing a combination of multiple systems presents as a viable solution. However, the approach to combining QA retrieval systems remains relatively limited. We propose a method …

การออกแบบและพัฒนาระบบการจัดการความรู้สำหรับการพัฒนาซอฟต์แวร์แบบสกรัมตามมาตรฐาน Iso/Iec 12207, กีรติกา ห่อเกียรติ

Chulalongkorn University Theses and Dissertations (Chula ETD)

โครงงานมหาบัณฑิตนี้มีวัตถุประสงค์ในการปรับปรุงกระบวนการจัดการความรู้ของหน่วยงานแห่งหนึ่งให้สอดคล้องกับมาตรฐานไอเอสโอ/ไออีซี 12207 กระบวนการจัดการความรู้นี้นำเสนอสำหรับโครงการที่ใช้ระเบียบวิธีแบบเอจายล์ตามกรอบงานสกรัม พร้อมทั้งได้ออกแบบและพัฒนาระบบการจัดการความรู้ เพื่อสนับสนุนการประยุกต์ใช้กระบวนการจัดการความรู้ภายในองค์กร ระบบที่พัฒนาสามารถให้บริการในการจัดเก็บ ค้นหา และแบ่งปันความรู้จากโครงการต่าง ๆ ขององค์กร อีกทั้งสนับสนุนให้มีการแบ่งปันความรู้เพื่อให้มีการพัฒนาซอฟต์แวร์ที่ความสอดคล้องตามความต้องการของผู้ใช้ และช่วยเพิ่มสมรรถนะในการทำงานของทีมพัฒนาซอฟต์แวร์ การดำเนินงานเริ่มจากการศึกษาแนวคิดและมาตรฐานที่เกี่ยวข้องกับการจัดการความรู้ รวมถึงการวิเคราะห์มาตรฐาน ไอเอสโอ/ไออีซี 12207 เพื่อประเมินและปรับปรุงกระบวนการจัดการความรู้ในปัจจุบันของ 2 กิจกรรม ได้แก่ “กิจกรรมที่ 3) การแบ่งปันสินทรัพย์ความรู้ทั่วทั้งองค์กร” และ “กิจกรรมที่ 4) การจัดการความรู้ ทักษะ และสินทรัพย์ความรู้” ในการปรับปรุงกระบวนการนั้นได้นำเสนอ โครงสร้างพื้นฐานกระบวนการ และการนิยามกระบวนการที่ปรับปรุง จากนั้นได้ทำการพัฒนาระบบการจัดการความรู้ ที่สามารถรองรับกระบวนการจัดการความรู้ที่ปรับปรุงแล้ว รวมถึงทวนสอบระบบกับความต้องการเชิงฟังก์ชัน และเพื่อยืนยันว่าระบบสามารถสนับสนุนกระบวนการทำงานของทีมสกรัมตามการนิยามกระบวนการได้อย่างมีประสิทธิภาพ ผลลัพธ์ของโครงงานนี้แสดงว่า ได้ช่วยเพิ่มประสิทธิภาพในการจัดการความรู้ภายในองค์กรตัวอย่าง และสามารถใช้เป็นแนวทางสำหรับองค์กรอื่น ๆ ที่ต้องการปรับปรุงกระบวนการจัดการความรู้ในบริบทของการพัฒนาซอฟต์แวร์แบบสกรัม

Enhancing Multilingual Sentence Representation Learning For The Job Recruitment Domain, Napat Laosaengpha

Chulalongkorn University Theses and Dissertations (Chula ETD)

With the advancement in natural language processing (NLP), there has been significant development in multilingual pretraining sentence encoder. Typically, these pretraining models are trained on large-scale datasets that consist of general text data from various sources such as Wikipedia. However, the general proposed models aren't enough to understand contexts in such a domain-specific, especially in the job recruitment domain. It is due to its niche nature and the lack of readily available related information. To enhance the existing multilingual pretraining sentence encoder and mitigate the aforementioned problems, we first propose multi-task dual-encoder framework to improve the sentence encoder for general-purpose …

Benchmarking And Enhancing Generalization In Multilingual Speech Emotion Recognition, Mohamed Osman Ismael

Theses and Dissertations

Speech Emotion Recognition (SER) is pivotal in advancing human-computer interaction by enabling machines to understand and respond to human emotions. Despite significant progress with self-supervised learning models, SER systems often struggle with generalization across diverse languages and unseen data distributions, limiting their real-world applicability. This thesis addresses these challenges by first introducing a large-scale benchmark to evaluate the robustness and adaptability of state-of-the-art SER models in both in-domain and out-of-domain settings. The benchmark includes a diverse set of multilingual datasets, emphasizing cross-lingual and out-of-domain evaluations to assess model generalization. Surprisingly, we find that the Whisper model, originally designed for automatic …

Predicting Remaining Useful Life Of Turbofan Engines On Cmapss And N-Cmapss Using Deep Recurrent Neural Networks, Samaikya Tippareddy

Master's Projects

Aircraft engines are susceptible to failure at multiple points over their lifespan and need replacement or repairs. The ability to proactively determine how long an engine will function helps avoid fatalities and build a reliable prognostic system. To accomplish this, predictive models are being developed using various approaches like physics-based and data-driven techniques. Physics-based models need huge computing power for simulations and domain knowledge for understanding and implementing the models. Alternatively, if we have substantial data for prediction, data-driven models can be used. In this research, we use data-driven approach for engine Remaining-Useful-Life (RUL) prediction on the NASA Commercial Modular …

Multimodal Techniques For Malware Classification, Jonathan Jiang

Master's Projects

The threat of malware has remained a serious concern for computer networks and systems, highlighting the need for accurate classification techniques. This research adopted the structured nature of PE files incorporated with a multi-modal machinelearning approach, to classify malware types. Features extracted from the PE headers were used to train an LSTM model. Features extracted from the PE sections were used to train a CNN model. Probabilities produced from these two models were then concatenated and fed into an SVM classifier. This multi-modal approach demonstrated high accuracy by experimenting with and verifying the approach on a large and labeled dataset. …

Explainable Reinforcement Learning For Network Routing Optimization, Yu Xiu

Master's Projects

Software-defined Networking (SDN) provides a solution for configuring multiple network devices by offering a centralized controller architecture. Network routing is one of the most crucial problems in network configuration. In particular, the surging network traffic demands require efficient routing techniques to load balance communication links. In order to optimize the communication path’s utilization and reduce request blocking, this project utilizes Reinforcement Learning (RL) to decide the routes for given network requests. Furthermore, we adopt Explainable Reinforcement Learning (XRL) to explain the RL learning agent’s decision-making process to enhance the trustworthiness of our approach. In particular, we focus on Feature Importance …

A Review Of Textmining Techniques: Trends, And Applications In Various Domains, Hiba J. Aleqabie, Mais Saad Sfoq, Rand Abdulwahid Albeer, Enaam Hadi Abd

Iraqi Journal for Computer Science and Mathematics

Text mining, a subfield of natural language processing (NLP), has received considerable attention in recent years due to its ability to extract valuable insights from large volumes of unstructured textual data. This review aims to provide a comprehensive evaluation of the applicability of text mining techniques across various domains and industries. The reviewstarts off with a dialogue of the basic ideas and methodologies that are concerned with textual content mining together with preprocessing, feature extraction, and machine learning algorithms. Furthermore, this survey highlights the challenges faced at some stage in implementing textual content mining strategies. Additionally, the review explores emerging …

Magnetometer-Less State-Estimation Of A Mobile Robot Using Cascaded Kalman Filters, Tommy Le

Graduate Research Theses & Dissertations

Localization, or state-estimation algorithms, are one of the most important aspects inthe development of autonomous mobile robots. Typical localization requires an IMU (Inertial Measurement Unit) along with an external reference, such as GPS (Global Positioning System) for outdoor applications. In indoor applications, the GPS data is not accessible so many mobile robot implementations turn to magnetometers to provide additional pose information. However, in the context of miniaturizing robotic systems, magnetometers are not always reliable due to their proximity to motors and other electronics, causing magnetic distortion and in turn, incorrect pose information. To address this issue, this thesis proposes a …

Using Ontological Methods To Compare Cybersecurity Maturity Model Certification 2.0 And Cobit 19, Aaron Marshall Ramey

CCAC Theses and Dissertations

Cybersecurity frameworks developed by a variety of organizations and implemented by a much larger collection of organizations differ in their focus and application. Whether designed by a private or government organization, the primary goal is to provide a framework to assess and reduce risk. The Department of Defense (DoD) has recently implemented the second version of the Cybersecurity Maturity Model Certification (CMMC 2.0). In some situations, compliance with CMMC 2.0 has already become mandatory for the Defense Industrial Base (DIB). Compliance will soon be required for all Large Businesses (LB) and Small Businesses (SB) within the DIB. While COBIT 19 …

Combating Disinformation On Social Media Networks With Media And Information Literacy Training For Social Media Network Users, Oscar Kwok Chao Ho

CCAC Theses and Dissertations

In the Internet age, social media networks (SMNs), such as Facebook (FB), Instagram (IG), and Twitter (TW), have gained popularity and become an essential part of human life. SMNs provide ease of connection to family, friends, and communities; however, they increase the chances social media network users (SMN users) will disclose private information (PI), causing critical harm to SMN users’ information privacy (IP). Furthermore, SMN users are exposed to significant amounts of disinformation, misinformation, or fake news, which they share without realizing the information is untrustworthy.

The goal of this developmental research was to investigate, examine, and understand the effects …

A Technique For Visualization Of Multivariate Categorical Data, Janice James

CCAC Theses and Dissertations

Multivariate Categorical Data (MCD) plays a significant role in many industries, and the ability to understand the data is critical for insight and decision making. Visualization is a key tool for understanding the data. This dissertation designed and implemented a novel technique for visualizing MCD called Pivoting Parallel Charts (PPC). The design of PPC was informed by studying several existing MCD visualization techniques.

PPC visualizes MCD as a sequence of parallel axes with affixed bar charts. A user-specified axis, called the pivot, acts as the crucial point of consideration for all data relationships. The bar charts are color-coded by the …

Enhancing Iot Security: Optimizing Anomaly Detection Through Machine Learning, Maria Balega, Waleed Farag, Xin-Wen Wu, Soundarararjan Ezekiel, Zaryn Good

Computer Science Articles

As the Internet of Things (IoT) continues to evolve, securing IoT networks and devices remains a continuing challenge. Anomaly detection is a crucial procedure in protecting the IoT. A promising way to perform anomaly detection in the IoT is through the use of machine learning (ML) algorithms. There is a lack of studies in the literature identifying optimal (with regard to both effectiveness and efficiency) anomaly detection models for the IoT. To fill the gap, this work thoroughly investigated the effectiveness and efficiency of IoT anomaly detection enabled by several representative machine learning models, namely Extreme Gradient Boosting (XGBoost), Support …

Social Media Bot Detection Using Dropout-Gan, Anant Shukla

Master's Projects

Bot activity on social media platforms is a pervasive problem, undermining the credibility of online discourse and potentially leading to cybercrime. We propose an approach to bot detection using Generative Adversarial Networks (GAN). We discuss how we overcome the issue of mode collapse by utilizing multiple discriminators to train against one generator, while decoupling the discriminator to perform social media bot detection and utilizing the generator for data augmentation. We demonstrate that our approach outperforms---in terms of accuracy---the state-of-the-art techniques in this field. We also show how the generator in the GAN can be used to evade such a classification …

Gradual Typing For Information Flow Control In Typescript Using Es Lint, Ashish Agarwal

Master's Projects

Current state-of-the-art systems tackle data security threats by incorporating information flow control (IFC) to ensure that a piece of information reaches only its intended recipient. However, most IFC implementations introduce a custom language built on top of a well-known language. Adaptations of such languages are limited due to limited support and updates, along with difficulty in learning new syntaxes. Implementations without a custom language offer incomplete IFC support. We present a comprehensive framework by leveraging Typescript, in conjunction with ESLint and NodeJS, aiming to resolve some of the limitations of IFC and intending to facilitate acceptance by a wide range …

Zookeeper Through The Ages: A Comparative Performance Analysis Of Evolutionary Versions, Ashish Khanchandani

Master's Projects

The rise in the need for scalable, fault-tolerant, and high-performance systems is the primary factor driving the developments in distributed computing. However, the proliferation of distributed computing creates significant difficulties. In an internet- scale setting where errors and network delays are frequent, coordinating distributed applications presents difficulties that ZooKeeper attempts to solve. It ensures several crucial characteristics that guarantee reliability and consistency. However, performance and latency play a huge part when it comes to dealing with systems built for handling very high loads. ZooKeeper has very low latency for read-heavy workloads, making it suitable for real-time applications. These factors contribute …

A Censorship Resistant News Website Using The Ethereum Blockchain, Hoang Lai

Master's Projects

Misleading information, false claims, and fabricated news articles not only misguide readers but also undermine the trustworthiness of the news platforms themselves. The blockchain provides decentralized, immutable data storage and offers a promising solution to prevent censorship on news websites. Compared to traditional news websites, a decentralized application (dApp) offers benefits such as greater stability and resistance to information manipulation. A decentralized web app is harder to attack than centralized servers since the database is stored across a blockchain network. Moreover, blockchain prevents censorship by letting readers check data across all blocks in the Blockchain, which is good for a …

Echo: A Browser Extension That Runs Experimental Javascript, Prayuj Pillai

Master's Projects

Narcissus is a JavaScript interpreter written in JavaScript. While it is a good engine for experimenting with JavaScript’s design, it does not integrate easily into the browser. This project introduces ‘‘Echo’’, a browser add-on designed to execute Narcissus JavaScript files and scripts within web browsers. The project explores the performance of the Narcissus interpreter against native browser JavaScript engines and benchmarks the results, showcasing the trade-offs in running an experimental engine—the Narcissus interpreter—on the browser versus native JavaScript. Additionally, as a proof of concept, we implement taint tracking, a capability meant to boost security by preventing sensitive data from being …

Domain Switch On Sentiment Analysis Using Gradient Reversal Layer, Hemish Veeraboina

Master's Projects

Switching domains in sentiment analysis presents the challenge of transferring learned knowledge from one context to another without the need to label data. Traditional methods often struggle when dealing with differences in data distribution a problem known as the domain shift issue. To tackle this using Gradient Reversal Layers (GRL) has emerged as a solution for adapting to different domains in an unsupervised learning setting. This study introduces an enhancement to the standard GRL approach by incorporating a sigmoid function that gradually adjusts how intensely domain adaptation occurs during training. This upgraded GRL technique ensures controlled learning outcomes making it …

Scalable Container Caching Optimization With Action Masking For Serverless Edge Computing, Manikanta Sanjay Veera

Master's Projects

Serverless edge computing is an emerging technology that realizes the low latency and resource-efficient function calls for responsive computing. In cloud-based serverless computing, it is a common practice to cache sufficiently many function containers for future reuse to reduce the overhead of container initiation. In contrast, the capacity limitation of edge nodes poses a complex problem to the caching strategy in serverless edge computing of selecting an appropriate set of container caches based on the request distribution. Deep Reinforcement Learning (DRL) can play a crucial role in optimizing the caching decisions under dynamic request arrivals. In this paper, we propose …

Regional Sea Level Rise Prediction In Monterey Bay With Lstms And Vertical Land Motion, Branden Lopez

Master's Projects

Earth system data is vast in volume and variety, and is used to forecast weather,

hurricanes, floods, and sea level. Sea Level Rise (SLR) impacts various sectors, espe- cially ecosystems, food production, industry, population, health, and the availability of

clean water. Because of its broad impact, describing the behavior and forecasting SLR is an important topic. Traditional Machine Learning (ML) models vary in use, but many are not capable of capturing all the non-linear spatial and temporal properties of SLR factors. Deep learning models efficaciously handle complex time series data, noise, and high dimensional spaces, making them a focus of …

Skin Cancer Detection Using Reinforcement Learning, Vikas Chercadu

Master's Projects

Advances in medical diagnostics have increasingly harnessed the power of artificial

intelligence, offering substantial improvements in early and accurate disease identifi- cation. This paper elaborates on a novel integration of Multi-Agent Reinforcement

Learning (MARL) with Deep Learning for the early detection of skin cancer, one of the most prevalent and lethal forms of cancer when left unchecked. Our project capitalizes on the sophisticated VGG16 Network for the extraction of detailed features from the widely-utilized HAM10000 dermatoscopic dataset, enhancing these features with additional color, texture, and shape analysis. Utilizing a custom-designed MARL environment, we facilitate a collaboration among multiple intelligent agents, …

Mitigating Learning Bias In Healthcare Datasets, Samyak Jagdish Kumbhalwar

Master's Projects

CVDs have been a major cause of deaths worldwide with WHO reporting 17.9 million deaths annually. Although there are advancements in the treatment of these diseases, most of the fatalities are a result of untimely diagnosis. Active research is going on to collect data points and risk factors related to these diseases, which can enable early diagnosis. Of the datasets available, many researchers have employed different ML models to predict/detect the prevalence of heart diseases. Many employed Tree based, regression models [3, 6]. Few also tried ensemble approaches [1, 2, 4]. These healthcare datasets are generally found to be imbalanced. …

Understanding Distracted Driving And Gaze Patterns In A Driving Simulator: Simple Vs. Complex Challenges, Rishi Prabhat Narayana Sannala

Master's Projects

Distracted driving has grown in criticality over the recent years, given the numerous distractions that drivers face today, and which have further been magnified by the proliferation of in-vehicle technologies and mobile devices. Such distractions can seriously compromise a driver's ability to be fully focused on the road and to carry out timely responses and informed decisions that are key in minimizing the risks of a crash and maximizing road safety. The general aim of the research is to observe how simple versus complex distractors affect driving performance and gaze patterns. The eyetracker used is the Tobii Pro Fusion, synchronized …

Hindi Image Captioning Using Indictrans2 And Encoder-Decoder Architecture, Anahita Vayalombrone Dinesh

Master's Projects

One of the most prominent tasks that lie on the conjunction of Natural Language Processing (NLP) and computer vision, is image captioning. Image captioning is the generative task of achieving textual descriptions from images. Its application finds use in many real-world scenarios like aiding the visually impaired, editing applications, recommendation systems, and medical imaging. This research focus lies in Hindi image captioning, the official language of India, as it has not been explored as far as its need. Several challenges such as the lack of substantial Hindi text data for training models, the need for human annotators to verify the …

A Rule-Based Hybrid Translation Model For Context-Aware Speech To Indian Sign Language Tasks, Dania Jaison

Master's Projects

In India, one of the most significant tools to communicate with the Deaf and Hard of Hearing (DHH) communities is Indian Sign Language(ISL). The issue of a scarcity of computational resources for ISL is even more pronounced when it comes to the translation of written or spoken English into ISL. This paper proposes a rule-based model for translation where the spoken english sentences are translated into ISL by focusing on the specific syntactic and grammatical differences between these two languages. ISL uses a simplified syntax unlike the nuanced sentence structures in English — we omit most of the auxiliary verbs …

Multimodal Emotion Detection In Conversations And Dialogues: A Fusion Model Approach, Abhinay Jatoth

Master's Projects

Emotion recognition is gaining traction due to its wide range of potential applications across different fields. With the rise of social media, chat platforms, and voice assistants, there is a vast increase in data through which humans implicitly and explicitly carry emotional cues. With new algorithms being developed for understanding the nuances of human language and emotion, businesses can tailor more personalized and empathetic service. Sentiment analysis, expresses a positive, negative, or neutral viewpoint laid the foundation of Emotion classification. Emotion classification in conversations represents the most advanced stage of classification. It is also challenging due to the existence and …

Lexigen: Lexical-Driven Image Generation, Sangram Prashant Chincholkar

Master's Projects

This research project proposes a novel approach to user-driven image editing via natural language descriptions. The aim is an accurate change of certain features of an image with respect to the descriptive text while maintaining, with equal concern, the integrity of the remaining parts of the image not affected by the description. The task is particularly relevant for fields like content creation, personalized design, and automated image editing that require both coherence of a visual scene and textual description. We propose a generative model, LexiGen, which perfectly integrates natural language descriptions with their corresponding visual changes within an image. The …

Personalizing Image Generation From Prompts Using Generative Ai, Mit Ramesh Jain

Master's Projects

The rapid advancements in Generative AI, particularly Text-to-Image (T2I) models, have opened up new possibilities for personalized image generation. Finetuning large T2I models for specific downstream tasks is a key approach to achieving tailored outputs. In recent years, Parameter-Efficient Fine-Tuning (PEFT) techniques have gained significant attention as a cost-effective and efficient solution for fine-tuning large models. Initially developed for fine-tuning large language models (LLMs), PEFT techniques have been extensively studied and compared in the context of language tasks. However, regarding the T2I domain, there is a lack of similarly exhaustive and detailed literature on PEFT. This research project, in the …

Load Balancing For Cloud-Based Applications, Nitish Ranjan

Master's Projects

Effective load balancing is critical in ensuring optimal resource utilization, reducing latency, and improving the overall performance of distributed systems. This report commences with a comprehensive literature review on existing load-balancing algorithms, examining their methodologies, strengths, and limitations within various computing environments, including cloud computing, data centers, and network traffic management. Despite significant advancements in this field, the dynamic nature of distributed systems, coupled with the ever-increasing demand for efficient data processing, poses ongoing challenges. In response, this study proposes a novel load-balancing algorithm to address these contemporary challenges. The approach leverages dynamic and hybrid load balancing, distinguishing it from …