Digital Commons Network^™

การพยากรณ์ความตรงในกระบวนการกลึงซีเอ็นซีสำหรับชิ้นงานเหล็กและอะลูมิเนียมโดยการประยุกต์ใช้โครงข่ายประสาทเทียม, วริศรา หลายวัฒนไพศาล

Chulalongkorn University Theses and Dissertations (Chula ETD)

ในปัจจุบันเทคโนโลยีใหม่ ๆ ได้ถูกพัฒนาอย่างต่อเนื่องเพื่อปรับปรุงและรองรับการผลิตขั้นสูง เนื่องจากสถานการณ์การแข่งขันที่ดุเดือดระหว่างอุตสาหกรรมการผลิตต่าง ๆ ดังนั้นเครื่องจักรกลอัจฉริยะและระบบการผลิตแบบอัจฉริยะจึงถูกคาดหวังว่าจะมีบทบาทสำคัญในอนาคตอันใกล้ เครื่องกลึงซีเอ็นซีถูกใช้อย่างกว้างขวางในอุตสาหกรรมการผลิตขึ้นสูงที่หลากหลาย ความตรงเป็นพารามิเตอร์ที่มีความสำคัญในกระบวนการกลึง เนื่องจากส่งผลกระทบโดยตรงต่อการประกอบชิ้นงาน อย่างไรก็ตามการควบคุมและการตรวจติดตามความตรงขณะกลึงชิ้นงานทำได้ยาก นอกจากนี้เครื่องกลึงซีเอ็นซียังไม่สามารถปรับตั้งค่าได้แบบทันทีทันใดขณะกลึงชิ้นงานโดยไม่หยุดเครื่องจักร ดังนั้นงานวิจัยนี้จึงมีจุดประสงค์ในการพัฒนาแบบจำลองการพยากรณ์ความตรงในกระบวนการกลึงซีเอ็นซีสำหรับชิ้นงานเหล็กกล้าคาร์บอนและอะลูมิเนียมเพื่อที่จะปรับปรุงกระบวนการควบคุมและตรวจติดตามความตรงโดยการประยุกต์ใช้โครงข่ายประสาทเทียมสองชั้นแบบป้อนข้อมูลไปข้างหน้า ซึ่งถูกสอนด้วยอัลกอริทึมแบบแพร่ย้อนกลับของเลเวนเบิร์ก-มาร์คอร์ด อัตราส่วนแรงตัดถูกนำมาใช้ในการคำนวณความตรงภายใต้เงื่อนการตัดต่าง ๆ การแปลงเวฟเลทแบบดอเบชีส์ถูกใช้ในการแยกแรงตัดพลวัตออกเป็น 10 ระดับ เพื่อที่จะกำจัดสัญญาณรบกวนอื่น ๆ ทำให้แบบจำลองมีความแม่นยำมากยิ่งขึ้น ปัจจัยในการตัดที่เกี่ยวข้องประกอบไปด้วยความเร็วตัด อัตราป้อนตัด ความลึกตัด รัศมีจมูกมีดตัด และมุมคายเศษโลหะ แบบจำลองโครงข่ายประสาทเทียมสำหรับพยากรณ์ความตรงของชิ้นงานเหล็กกล้าคาร์บอนและอะลูมิเนียมที่พัฒนาขึ้นจะถูกนำมาเปรียบเทียบกับแบบจำลองการพยากรณ์ความตรงแบบอื่น ๆ คือ แบบจำลองโครงข่ายประสาทเทียมสำหรับพยากรณ์ความตรงของชิ้นงานเหล็กกล้าคาร์บอน แบบจำลองโครงข่ายประสาทเทียมสำหรับพยากรณ์ความตรงของชิ้นงานอะลูมิเนียม และวิธีการวิเคราะห์ด้วยการถดถอยแบบพหุคูณ จากผลการวิจัยพบว่า แบบจำลองโครงข่ายประสาทเทียมสำหรับพยากรณ์ความตรงของชิ้นงานเหล็กกล้าคาร์บอนและอะลูมิเนียมที่ถูกเสนอนั้นมีความแม่นยำที่ 76.27% สำหรับชิ้นงานเหล็กกล้าคาร์บอนและอะลูมิเนียม มีความแม่นยำที่ 82.57% สำหรับชิ้นงานเหล็กกล้าคาร์บอน และมีความแม่นยำที่ 69.97% สำหรับชิ้นงานอะลูมิเนียม ในขณะที่วิธีการวิเคราะห์ด้วยการถดถอยแบบพหุคูณมีความแม่นยำที่ 74.23% แม้ว่าแบบจำลองโครงข่ายประสาทเทียมสำหรับพยากรณ์ความตรงของชิ้นงานเหล็กกล้าคาร์บอนและแบบจำลองโครงข่ายประสาทเทียมสำหรับพยากรณ์ความตรงของชิ้นงานอะลูมิเนียมจะมีความแม่นยำที่มากกว่าที่ 86.53% และ 70.70% ตามลำดับ แต่ผลการพยากรณ์ความตรงของแบบจำลองโครงข่ายประสาทเทียมทั้งสามแบบจำลองไม่มีความแตกต่างกันอย่างมีนัยสำคัญเมื่อทดสอบใน Paired t-Test

A Robust System For Core Thai Natural Language Processing Technologies, Can Udomcharoenchaikit

Chulalongkorn University Theses and Dissertations (Chula ETD)

As the amount of unstructured textual data grows, it becomes increasingly important to build an intelligent system that can process it. Natural Language Processing (NLP) is a technology that allows a computer to exploit human languages to perform tasks. Deep learning models have shown excellent results across fundamental tasks in NLP, such as word segmentation, part-of-speech tagging, and named-entity recognition. However, in many situations, these proposed methods fail to perform well. For an NLP system to be robust, it must address issues such as out-of-vocabulary and spelling-mistakes. This thesis's research goal is to develop NLP models that can handle malformed …

Semi-Supervised Thai Sentence Segmentation Using Local And Distant Word Representations, Chanatip Saetia

Chulalongkorn University Theses and Dissertations (Chula ETD)

A sentence is typically treated as the minimal syntactic unit used for extracting valuable information from a longer piece of text. However, in written Thai, there are no explicit sentence markers. We proposed a deep learning model for the task of sentence segmentation that includes three main contributions. First, we integrate n-gram embedding as a local representation to capture word groups near sentence boundaries. Second, to focus on the keywords of dependent clauses, we combine the model with a distant representation obtained from self-attention modules. Finally, due to the scarcity of labeled data, for which annotation is difficult and time-consuming, …

Explainable Stock Price Prediction Using Technical Indicators With Short Thai Textual Information, Kittisak Prachyachuwong

Chulalongkorn University Theses and Dissertations (Chula ETD)

A stock trend prediction has been in the spotlight from the past to the present. Fortunately, there is an enormous amount of information available nowadays. There were prior attempts that have tried to forecast the trend using textual information; however, it can be further improved since they relied on fixed word embedding, and it depends on the sentiment of the whole market. In this paper, we propose a deep learning model to predict the Thailand Futures Exchange (TFEX) with the ability to analyze both numerical and textual information. We have used Thai economic news headlines from various online sources. To …

Using Automatic Speech Recognition To Assess Thai Speech Language Fluency In Montreal Cognitive Assessment (Moca), Pimarn Kantithammakorn

Chulalongkorn University Theses and Dissertations (Chula ETD)

The Montreal Cognitive Assessment (MoCA), a widely accepted screening tool for identifying patients with mild cognitive impairment (MCI), includes a language fluency test of verbal functioning where scores are based on the number of unique correct words produced by the test-taker. However, with different languages, it is possible that unique words may be counted differently. This study focuses on Thai as a language that differs from English in its type of word combination. We applied various automatic speech recognition (ASR) techniques to develop an assisted scoring system for the language fluency test of the MoCA with Thai language support. The …

Mathematical Models Of Cellular Signaling And Supramolecular Self-Assembly, Pratip Rana

Theses and Dissertations

Synthetic biologists endeavor to predict how the increasing complexity of multi-step signaling cascades impacts the fidelity of molecular signaling, whereby cellular state information is often transmitted with proteins diffusing by a pseudo-one-dimensional stochastic process. We address this problem by using a one-dimensional drift-diffusion model to derive an approximate lower bound on the degree of facilitation needed to achieve single-bit informational efficiency in signaling cascades as a function of their length. We find that a universal curve of the Shannon-Hartley form describes the information transmitted by a signaling chain of arbitrary length and depends upon only a small number of physically …

Alone: A Dataset For Toxic Behavior Among Adolescents On Twitter, Thilini Wijesiriwardene, Hale Inan, Ugur Kursuncu, Manas Gaur, Valerie L. Shalin, Krishnaprasad Thirunarayan, Amit P. Sheth, I. Budak Arpinar

Publications

The convenience of social media has also enabled its misuse, potentially resulting in toxic behavior. Nearly 66% of internet users have observed online harassment, and 41% claim personal experience, with 18% facing severe forms of online harassment. This toxic communication has a significant impact on the well-being of young individuals, affecting mental health and, in some cases, resulting in suicide. These communications exhibit complex linguistic and contextual characteristics, making recognition of such narratives challenging. In this paper, we provide a multimodal dataset of toxic social media interactions between confirmed high school students, called ALONE (AdoLescents ON twittEr), along with descriptive …

Knowledge-Infused Statistical Learning For Social Good, Kaushik Roy, Manas Gaur

Publications

Humans are able to provide symbolic knowledge in structured form for potential use by an AI system in learning human-desirable concepts. In clinical settings, for instance, prediction of patient outcomes by an AI can be guided by knowledge from patient history. This history contains concepts such as treatment information, observational and drug-related information, mental health conditions, and severity of disease/disorder. Additionally, there is also often a certain graphical structure to the knowledge among the concepts, for example, ”patient symptoms cause certain tests to be taken”, which in turn affects the prescription of medication. This type of structure between human interpretable …

Analysis Of Antenna Designs For The Maximum Power Transmission, Lauryn P. Smith

College of Graduate Studies: Theses & Dissertations

Since Nikola Tesla discovered wireless power transmission, it has become a very interesting topic of study in the antennas and wireless propagation community. Various aspects and applications for wireless power transmission are studied today, a few of which are investigated in this work. First, various antenna geometries are analyzed for radiative near-field wireless power transfer in terms of electrical field strength. It is determined that the meander antenna is ideal for maximum power transfer in its radiative near-field region, contrary to its far-field behavior. Next, in the application of radio frequency identification, a directive, UHF RFID tag antenna is designed …

Enhanced Pml Based On The Long Short Term Memory Network For The Fdtd Method, He Ming Yao, Lijun Jiang

Electrical and Computer Engineering Faculty Research & Creative Works

This paper proposes a new absorbing boundary condition (ABC) computation approach based on the deep learning technique. Benefited from the sequence dependence feature, the Long Short-Term Memory (LSTM) network is employed to replace the conventional perfectly matched layer (PML) ABC for the Finite-Difference Time-Domain (FDTD) solving process. The newly proposed LSTM based PML model is trained by the electromagnetic field data on the interface of the conventional PML. Different from the conventional PML, the newly proposed model only needs one cell layer as the boundary. Hence, the newly proposed method conveniently reduces both the algorithm's complexity and the area of …

The Development Of A Project Typology And Selection Tool To Improve Decision-Making In Sustainable Projects, Rakan Alyamani

Doctoral Dissertations

"Decision-making in sustainable projects is a complex and challenging process, especially during the initiating and planning phases of project development, due to influence from several external factors, as well as the uncertain environments surrounding their creation. It is essential to improve the decision-making process in sustainable projects during these two phases by relying on strong decision-making tools. The first contribution in this work identifies gaps in the literature of how institutionalization can impact sustainable projects through the effects of institutional isomorphisms from institutional theory. This helps decision makers better understand the relationship between institutionalization and sustainable projects. The second contribution …

Content-Based Filtering Recommendation Approach To Label Irish Legal Judgements, Sandesh Gangadhar

Dissertations

Machine learning approaches are applied across several domains to either simplify or automate tasks which directly result in saved time or cost. Text document labelling is one such task that requires immense human knowledge about the domain and efforts to review, understand and label the documents. The company Stare Decisis summarises legal judgements and labels them as they are made available on Irish public legal source www.courts.ie. This research presents a recommendation-based approach to reduce the time for solicitors at Stare Decisis by reducing many numbers of available labels to pick from to a concentrated few that potentially contains the …

Exploration Of Approaches To Arabic Named Entity Recognition, Husamelddin Balla, Sarah Jane Delany

Conference papers

Abstract. The Named Entity Recognition (NER) task has attracted significant attention in Natural Language Processing (NLP) as it can enhance the performance of many NLP applications. In this paper, we compare English NER with Arabic NER in an experimental way to investigate the impact of using different classifiers and sets of features including language-independent and language-specific features. We explore the features and classifiers on five different datasets. We compare deep neural network architectures for NER with more traditional machine learning approaches to NER. We discover that most of the techniques and features used for English NER perform well on Arabic …

Proactive Content Caching In Future Generation Communication Networks: Energy And Security Considerations, Muhammad Ishtiaque Aziz Zahed

Theses: Doctorates and Masters

The proliferation of hand-held devices and Internet of Things (IoT) applications has heightened demand for popular content download. A high volume of content streaming/downloading services during peak hours can cause network congestion. Proactive content caching has emerged as a prospective solution to tackle this congestion problem. In proactive content caching, data storage units are used to store popular content in helper nodes at the network edge. This contributes to a reduction of peak traffic load and network congestion.

However, data storage units require additional energy, which offers a challenge to researchers that intend to reduce energy consumption up to 90% …

A Probabilistic Machine Learning Framework For Cloud Resource Selection On The Cloud, Syeduzzaman Khan

University of the Pacific Theses and Dissertations

The execution of the scientific applications on the Cloud comes with great flexibility, scalability, cost-effectiveness, and substantial computing power. Market-leading Cloud service providers such as Amazon Web service (AWS), Azure, Google Cloud Platform (GCP) offer various general purposes, memory-intensive, and compute-intensive Cloud instances for the execution of scientific applications. The scientific community, especially small research institutions and undergraduate universities, face many hurdles while conducting high-performance computing research in the absence of large dedicated clusters. The Cloud provides a lucrative alternative to dedicated clusters, however a wide range of Cloud computing choices makes the instance selection for the end-users. This thesis …

Network Sampling Through Crawling, Katchaguy Areekijseree

Dissertations - ALL

In recent years, researchers have increasingly used OSN data to study human behavior. Before such a study can begin, one must first obtain appropriate data. A platform, e.g, Facebook or Twitter, may provide an API for accessing data, but such APIs are often ratelimited, restricting the amount of data that an individual collects in a given amount of time. In order for the data collector to efficiently collect data, she needs to make intelligent use of her limited budget. Therefore, when collecting data, efficiency is extremely important. We consider the problem of network sampling through crawling, in which the data …

Sampling And Characterizing Evolving Communities In Social Networks, Humphrey Appiah Mensah

Dissertations - ALL

One of the most important structures in social networks is communities. Understanding communities is useful in many applications, such as suggesting a friend for a user in an online friendship network, recommending a product for a user in an e-commerce network, etc. However, before studying anything about communities, researchers first need to collect appropriate data. Getting complete access to the data for community studies is unrealistic in most cases. In this work, we address the problem of crawling networks to identify community structure. Firstly, we present a network sampling technique to crawl the community structure of dynamic networks when there …

Introducing Phonetic Information To Speaker Embedding For Speaker Verification, Yi Liu, Liang He, Michael T. Johnson

Electrical and Computer Engineering Faculty Publications

Phonetic information is one of the most essential components of a speech signal, playing an important role for many speech processing tasks. However, it is difficult to integrate phonetic information into speaker verification systems since it occurs primarily at the frame level while speaker characteristics typically reside at the segment level. In deep neural network-based speaker verification, existing methods only apply phonetic information to the frame-wise trained speaker embeddings. To improve this weakness, this paper proposes phonetic adaptation and hybrid multi-task learning and further combines these into c-vector and simplified c-vector architectures. Experiments on National Institute of Standards and Technology …

Stock Price Using Domain Specific Lexicons, Zane Turner

Computer Science and Computer Engineering Undergraduate Honors Theses

Sentiment analysis is a broad and expanding field that aims to extract and classifying opinions from textual data. Lexicon-based approaches are based on using a sentiment lexicon, a list of words each mapped to a sentiment score, to rate the sentiment of a text chunk. Our work focuses on predicting stock price change using a sentiment lexicon built from financial conference call logs. We introduce a method to generate a sentiment lexicon based upon an existing probabilistic approach. By using a domain-specific lexicon, we outperform traditional techniques and demonstrate that domain-specific sentiment lexicons provide higher accuracy than generic sentiment lexicons …

Counting The Number Of Possible Fuzzy Controllers, Ambalal Vinayak Patel

Manipal Journal of Science and Technology

Fuzzy logic is prominent among all types of logics due to the inherent ability to express human intelligence qualitatively. Thus, it is being used in a variety of applications to design embedded intelligent systems. Although it originated to model the complex and ill-defined systems, it eventually also resulted in utilization for designing the intelligent controllers. While designing the fuzzy logic controllers (FLC), we came across a major challenge of selecting the various elements thereof. The flexibility in selecting the combinations of elements leads to arrive at a number of FLCs. Therefore, the challenge came across in selecting the suitable set …

Extracting Social Network From Literary Prose, Tarana Tasmin Bipasha

Graduate Theses and Dissertations

This thesis develops an approach to extract social networks from literary prose, namely, Jane Austen’s published novels from eighteenth- and nineteenth- century. Dialogue interaction plays a key role while we derive the networks, thus our technique relies upon our ability to determine when two characters are in conversation. Our process involves encoding plain literary text into the Text Encoding Initiative’s (TEI) XML format, character name identification, conversation and co-occurrence detection, and social network construction. Previous work in social network construction for literature have focused on drama, specifically manually TEI-encoded Shakespearean plays in which character interactions are much easier to track …

Seer: An Explainable Deep Learning Midi-Based Hybrid Song Recommender System, Khalil Damak, Olfa Nasraoui

Faculty and Staff Scholarship

State of the art music recommender systems mainly rely on either matrix factorization-based collaborative filtering approaches or deep learning architectures. Deep learning models usually use metadata for content-based filtering or predict the next user interaction by learning from temporal sequences of user actions. Despite advances in deep learning for song recommendation, none has taken advantage of the sequential nature of songs by learning sequence models that are based on content. Aside from the importance of prediction accuracy, other significant aspects are important, such as explainability and solving the cold start problem. In this work, we propose a hybrid deep learning …

Visual Speech Recognition Using A 3d Convolutional Neural Network, Matthew Rochford

Master's Theses

Main stream automatic speech recognition (ASR) makes use of audio data to identify spoken words, however visual speech recognition (VSR) has recently been of increased interest to researchers. VSR is used when audio data is corrupted or missing entirely and also to further enhance the accuracy of audio-based ASR systems. In this research, we present both a framework for building 3D feature cubes of lip data from videos and a 3D convolutional neural network (CNN) architecture for performing classification on a dataset of 100 spoken words, recorded in an uncontrolled envi- ronment. Our 3D-CNN architecture achieves a testing accuracy of …

A Systematic Approach For Automatically Answering General-Purpose Objective And Subjective Questions, Lok Prasad Acharya

Theses and Dissertations

In this era of information explosion, people generally rely on the Internet, and more precisely, the search engines to get answers to their questions. However, what a search engine can do is just retrieve documents. Given some keywords, it only returns the relevant ranked documents that contain the keywords. Although users often want a precise answer to a question, they are left to extract answers from the documents themselves. This is where Automatic Question Answering (AQA) systems come into play. An AQA system takes questions in natural language as input and searches related answers in the set of documents and …

Aspect And Opinion Aware Abstractive Review Summarization With Reinforced Hard Typed Decoder, Yufei Tian, Jianfei Yu, Jing Jiang

Research Collection School Of Computing and Information Systems

In this paper, we study abstractive review summarization. Observing that review summaries often consist of aspect words, opinion words and context words, we propose a two-stage reinforcement learning approach, which first predicts the output word type from the three types, and then leverages the predicted word type to generate the final word distribution. Experimental results on two Amazon product review datasets demonstrate that our method can consistently outperform several strong baseline approaches based on ROUGE scores.

Personality Prediction Through Curriculam Vitae Analysis Involving Password Encryption And Prediction Analysis, Gagandeep Kaur, Shruti Maheshwari

Library Philosophy and Practice (e-journal)

A recruitment process requires an eligibility check, an aptitude evaluation and a psychometric analysis of prospective candidates. The work puts forward an application where the system allows employers to post new job offerings and registered candidates can apply. The application estimates applicant’s emotional aptitude through a psychometric analysis based on a test whereas the professional standard is verified via a technical aptitude test. OCEAN Model is used to assess emotional quotient and predict the personality traits. Machine learning techniques such as Logistic Regression are used for modelling the personality predictor. The details of the candidates are kept secure by using …

Geometric Formation Of Schemes Of Technical Objects In Automated Design Systems, M.A. Khadzhimuhametova

Journal of Tashkent Institute of Railway Engineers

This paper discusses the issues of mathematical support for the formation of various schemes of complex technical objects for a computer-aided design subsystem. Information is given on the description of the station diagram using finite element methods. On the one hand, it is required that the mathematical model be as universal as possible (cover all types of de-signed schemes), accurate, algorithmically reliable. On the other hand, it is required to ensure the minimum cost of the required memory of com-puter time. These contradictions are eliminated by creating a bank of models. Models are planned from separate modules of structural and …

Computing Without Computing: Dna Version, Vladik Kreinovich, Julio C. Urenda

Departmental Technical Reports (CS)

The traditional DNA computing schemes are based on using or simulating DNA-related activity. This is similar to how quantum computers use quantum activities to perform computations. Interestingly, in quantum computing, there is another phenomenon known as computing without computing, when, somewhat surprisingly, the result of the computation appears without invoking the actual quantum processes. In this chapter, we show that similar phenomenon is possible for DNA computing: in addition to the more traditional way of using or simulating DNA activity, we can also use DNA inactivity to solve complex problems. We also show that while DNA computing without …

Using Uncertainty To Interpret Supervised Machine Learning Predictions, Michael C. Darling

Electrical and Computer Engineering ETDs

Traditionally, machine learning models are assessed using methods that estimate an average performance against samples drawn from a particular distribution. Examples include the use of cross-validation or hold0out to estimate classification error, F-score, precision, and recall.

While these measures provide valuable information, they do not tell us a model's certainty relative to particular regions of the input space. Typically there are regions where the model can differentiate the classes with certainty, and regions where the model is much less certain about its predictions.

In this dissertation we explore numerous approaches for quantifying uncertainty in the individual predictions made by supervised …

Participatory Sensing-Based Geospatial Localization Of Distant Objects For Disaster Preparedness In Urban Built Environments, Hongjo Kim, Youngjib Ham

Civil and Environmental Engineering and Engineering Mechanics Faculty Publications

Although the benefit of participatory sensing for collecting local data over large areas has long been recognized, it has not been widely used for various applications such as disaster preparation due to a lack of geospatial localization capability with respect to a distant object. In such applications, objects of interest need to be ro- bustly localized and documented for supporting data-driven decision-making in site inspection and resource mobilization. However, participatory sensing is inappropriate to localize a distant object due to the absence of ranging sensors in citizens' mobile devices; thus, the localization accuracy varies to a large extent. To address …