Evaluating The English-Turkish Parallel Treebank For Machine Translation,
2022
TÜBİTAK
Evaluating The English-Turkish Parallel Treebank For Machine Translation, Onur Görgün, Olcay Taner Yildiz
Turkish Journal of Electrical Engineering and Computer Sciences
This study extends our initial efforts in building an English-Turkish parallel treebank corpus for statistical machine translation tasks. We manually generated parallel trees for about 17K sentences selected from the Penn Treebank corpus. English sentences vary in length: 15 to 50 tokens including punctuation. We constrained the translation of trees by (i) reordering of leaf nodes based on suffixation rules in Turkish, and (ii) gloss replacement. We aim to mimic human annotator?s behavior in real translation task. In order to fill the morphological and syntactic gap between languages, we do morphological annotation and disambiguation. We also apply our heuristics by …
An Active Contour Model Using Matched Filter And Hessian Matrix For Retinalvessels Segmentation,
2022
TÜBİTAK
An Active Contour Model Using Matched Filter And Hessian Matrix For Retinalvessels Segmentation, Mahtab Shabani, Hossein Pourghassem
Turkish Journal of Electrical Engineering and Computer Sciences
Medical image analysis, especially of the retina, plays an important role in diagnostic decision support tools. The properties of retinal blood vessels are used for disease diagnoses such as diabetes, glaucoma, and hypertension. There are some challenges in the utilization of retinal blood vessel patterns such as low contrast and intensity inhomogeneities. Thus, an automatic algorithm for vessel extraction is required. Active contour is a strong method for edge extraction. However, it cannot extract thin vessels and ridges very well. In this research, we propose an improved active contour method that uses discrete wavelet transform for energy minimization to solve …
Stressed Or Just Running? Differentiation Of Mental Stress And Physical Activityby Using Machine Learning,
2022
TÜBİTAK
Stressed Or Just Running? Differentiation Of Mental Stress And Physical Activityby Using Machine Learning, Yekta Sai̇d Can
Turkish Journal of Electrical Engineering and Computer Sciences
Recently, modern people have excessive stress in their daily lives. With the advances in physiological sensors and wearable technology, people?s physiological status can be tracked, and stress levels can be recognized for providing beneficial services. Smartwatches and smartbands constitute the majority of wearable devices. Although they have an excellent potential for physiological stress recognition, some crucial issues need to be addressed, such as the resemblance of physiological reaction to stress and physical activity, artifacts caused by movements and low data quality. This paper focused on examining and differentiating physiological responses to both stressors and physical activity. Physiological data are collected …
An Effective Prediction Method For Network State Information In Sd-Wan,
2022
TÜBİTAK
An Effective Prediction Method For Network State Information In Sd-Wan, Erdal Akin, Ferdi̇ Saraç, Ömer Aslan
Turkish Journal of Electrical Engineering and Computer Sciences
In a software-defined wide area network (SD-WAN), a logically centralized controller is responsible for computing and installing paths in order to transfer packets among geographically distributed locations and remote users. Accordingly, this would necessitate obtaining the global view and dynamic network state information (NSI) of the network. Therefore, the centralized controller periodically collects link-state information from each port of each switch at fixed time periods. While collecting NSI in short periods causes protocol overhead on the controller, collecting in longer periods leads to obtaining inaccurate NSI. In both cases, packet losses are inevitable, which is not preferred for quality of …
Synthesizing Realistic Substitute Data For A Law Enforcement Database Using A Python Library,
2022
West Virginia University
Synthesizing Realistic Substitute Data For A Law Enforcement Database Using A Python Library, Anthony Carrola
Graduate Theses, Dissertations, and Problem Reports (ETD)
In many databases, there is private or sensitive data that should not be accessible to any but a few individuals, such as HIPAA (Health Insurance Portability and Accountability Act) protected or LE (law enforcement) data. However, there is often a need to work with the data or change it for proper and thorough testing, especially for the developers . In some cases, the developers may be authorized to access and view the data, but it is rarely allowable for that data to be changed. Further, it is unlikely, especially on a large project, that all of the developers will have …
A Cloud Computing-Based Dashboard For The Visualization Of Motivational Interviewing Metrics,
2022
Wright State University
A Cloud Computing-Based Dashboard For The Visualization Of Motivational Interviewing Metrics, E Jinq Heng
Browse all Theses and Dissertations
Motivational Interviewing (MI) is an evidence-based brief interventional technique that has been demonstrated to be effective in triggering behavior change in patients. To facilitate behavior change, healthcare practitioners adopt a nonconfrontational, empathetic dialogic style, a core component of MI. Despite its advantages, MI has been severely underutilized mainly due to the cognitive overload on the part of the MI dialogue evaluator, who has to assess MI dialogue in real-time and calculate MI characteristic metrics (number of open-ended questions, close-ended questions, reflection, and scale-based sentences) for immediate post-session evaluation both in MI training and clinical settings. To automate dialogue assessment and …
Can We Make Our Robot Play Soccer? Influence Of Collaborating With Preservice Teachers And Fifth Graders On Undergraduate Engineering Students' Learning During A Robotic Design Process (Work In Progress),
2022
Old Dominion University
Can We Make Our Robot Play Soccer? Influence Of Collaborating With Preservice Teachers And Fifth Graders On Undergraduate Engineering Students' Learning During A Robotic Design Process (Work In Progress), Krishnanand Kaipa, Jennifer Kidd, Julia Noginova, Francisco Cima, Stacie Ringleb, Orlando Ayala, Pilar Pazos, Kristie Gutierrez, Min Jung Lee
Mechanical & Aerospace Engineering Faculty Publications
This work-in-progress paper describes engineering students’ experiences in an NSF-funded project that partnered undergraduate engineering students with pre-service teachers to plan and deliver robotics lessons to fifth graders at a local school. This project aims to address an apparent gap between what is taught in academia and industry’s expectations of engineers to integrate perspectives from outside their field to solve modern societal problems requiring a multidisciplinary approach. Working in small teams over Zoom, participating engineering, education, and fifth grade students designed, built, and coded bio-inspired COVID companion robots. The goal for the engineering students was to build new interprofessional skills, …
Modeling Document Classification To Automate Mental Health Diagnosis,
2022
Eastern Washington University
Modeling Document Classification To Automate Mental Health Diagnosis, William M. Tadlock
EWU Masters Thesis Collection
The objective of this study is to determine if diagnosis documents can be used with document classification to automatically diagnose mental health conditions. Document classification allows text documents to be analyzed and organized into their appropriate classes based on the features and words presented in the text. One application of this is within the medical field to automatically classify different patient diagnosis based on medical or patient notes. This research applied mental health diagnosis documents to automatically diagnose a group of patients with a mental health condition based on text-based survey data. This classification was approached through several feature engineering …
Towards Cloud-Based Cost-Effective Serverless Information System,
2022
Eastern Washington University
Towards Cloud-Based Cost-Effective Serverless Information System, Isaac C. Angle
EWU Masters Thesis Collection
E-commerce information systems are becoming increasingly popular for businesses to adopt. In this work, we propose a serverless information system that will reduce costs for small businesses trying to create an e-commerce website. The proposed serverless system is built entirely in Amazon Web Services (AWS). The proposed serverless system allows businesses to pay for the use of cloud resources on a per-order granularity. This model reduces the cost of the information system when compared to a traditional cloud-based system. As e-commerce websites become more vital for small businesses, a cost effective serverless approach is promising.
Detecting User Emotions From Audio Conversations With The Smart Assistants,
2022
Missouri State University
Detecting User Emotions From Audio Conversations With The Smart Assistants, Sunanda Guha
Graduate Theses/Dissertations
With the proliferation of smart home devices like Google Home or Amazon Alexa, significant research endeavors are being carried out to improve the user experience while interacting with these smart assistants. One such dimension in this endeavor is ongoing research on successful emotion detection from short voice commands used in smart home environment. Besides facial expression and body language, etc., speech plays a pivotal role in the classification of emotions when it comes to smart home application. Upon successful implementation of accurate emotion recognition, the smart devices will be able to intelligently and empathetically suggest appropriate actions based on the …
Service Placement Optimization For Location-Based Service,
2022
Faculty of Engineering
Service Placement Optimization For Location-Based Service, Karnkitti Kittikamron
Chulalongkorn University Theses and Dissertations (Chula ETD)
Location-based service (LBS) is necessary and useful for several applications including navigation and games. These real-time applications require high accuracy and low delay. In general, the complexity of indoor localization algorithms used in LBS depends on the size of fingerprint data. This can lead to long delays when operating in large-scale areas. In this paper, we propose a novel optimization framework for edge service placement, aiming at minimizing the overall cost of edge computing deployment and service response time. Our placement strategy is used to solve the formulated edge node placement problems. The simulated annealing approach is then used in …
Machine Learning For Wavelet-Based Valve Stiction Detection,
2022
Faculty of Engineering
Machine Learning For Wavelet-Based Valve Stiction Detection, Kris Prasopsanti
Chulalongkorn University Theses and Dissertations (Chula ETD)
Valve stiction presents challenges in industrial process control, leading to oscillations and hindering the regulation of fluid flow. This thesis addresses the detection of valve stiction by exploring the limitations of current methods and proposes a novel approach that combines wavelet reconstruction and convolutional neural networks (CNN) to enhance stiction detection performance. The proposed method utilizes preprocessed process variable versus controller output (PV(OP)) plots as input to the CNN model, capitalizing on the distinctive characteristics of stiction. Training and evaluation employ both simulated and real-world data from the International Stiction Data Base (ISDB), with the F1 score serving as the …
กระบวนการและแบบจำลองสำหรับการคัดกรองเพื่อการจัดการคุณภาพข้อมูลในคราวด์ซอร์สซิงแพลตฟอร์ม,
2022
คณะวิศวกรรมศาสตร์
กระบวนการและแบบจำลองสำหรับการคัดกรองเพื่อการจัดการคุณภาพข้อมูลในคราวด์ซอร์สซิงแพลตฟอร์ม, กฤตย์ กังวาลพงศ์พันธุ์
Chulalongkorn University Theses and Dissertations (Chula ETD)
การเก็บรวบรวมข้อมูลด้วยคราวด์ซอร์สซิงเป็นวิธีที่โดยทั่วไปมีความเร็วมากกว่า มีต้นทุนต่ำกว่า และมีความหลากหลายมากกว่าวิธีการเก็บรวบรวมข้อมูลแบบอื่น ๆ อย่างไรก็ตาม คราวด์ซอร์สซิงอาจเผชิญกับปัญหาคุณภาพ เช่น การติดป้ายกำกับผิดหรือการนำมาใช้ในทางที่ไม่เหมาะสม ดังนั้น กระบวนการควบคุมคุณภาพเป็นสิ่งที่จำเป็นสำหรับแพลตฟอร์มคราวด์ซอร์สซิง วิทยานิพนธ์นี้ศึกษาค้นคว้าอุปสรรคและวิธีการแก้ไขที่เป็นไปได้ในการจัดการคุณภาพของผู้ใช้งานแพลตฟอร์มคราวด์ซอร์สซิง ส่วนแรกเน้นวิธีการเพิ่มกระบวนการในคราวด์ซอร์สซิง โดยศึกษา 3 วิธี ได้แก่ 1. งานที่จำเป็นต้องทำก่อน 2. คำถามมาตรฐานแบบทองคำ และ 3. การทำซ้ำของข้อมูล พบว่างานที่จำเป็นต้องทำก่อนเป็นสิ่งจำเป็นเพื่อคัดกรองให้ได้ผู้ปฏิบัติงานที่มีคุณภาพสูง โดยควรเน้นไปที่ลักษณะเฉพาะและรายละเอียดของงาน คำถามที่ตรวจสอบความสอดคล้องระหว่างงานดีกว่าคำถามแบบชัดเจนในการตรวจสอบด้วยคำถามมาตรฐานทองคำ ผู้ตรวจสอบข้อมูลคนเดียวอาจนำไปสู่การปรับปรุงคุณภาพข้อมูลได้มากที่สุด ส่วนที่สองคือ การใช้แบบจำลองการเรียนรู้ของเครื่องที่ใช้ข้อมูลพฤติกรรมในการทำนายคุณภาพของข้อมูล ซึ่งวิธีนี้ยังช่วยคัดกรองข้อมูลคุณภาพต่ำออกไปได้โดยไม่เสียทรัพยากรเพิ่มเติม
แบบจำลองคุณภาพซอฟต์แวร์โอเพนซอร์ซเพื่อการวัดคุณภาพอย่างอัตโนมัติ,
2022
คณะวิศวกรรมศาสตร์
แบบจำลองคุณภาพซอฟต์แวร์โอเพนซอร์ซเพื่อการวัดคุณภาพอย่างอัตโนมัติ, อัรกอม มะแดเฮาะ
Chulalongkorn University Theses and Dissertations (Chula ETD)
ปัจจุบันได้มีการนำเสนอแบบจำลองคุณภาพซอฟต์แวร์โอเพนซอร์ซหลายแบบจำลองเพื่อใช้ในการประเมินคุณภาพของซอฟต์แวร์โอเพนซอร์ซ แต่แบบจำลองเหล่านั้นมีข้อจำกัดเนื่องจากเป็นการประเมินตามความคิดส่วนตัวซึ่งต้องอาศัยผู้ใช้ในการประเมิน และแบบจำลองดังกล่าวต้องการข้อมูลจากหลากหลายแหล่ง เพื่อเสริมการประเมินคุณภาพซอฟต์แวร์โอเพนซอร์ซที่เป็นอยู่ให้ทำได้สมบูรณ์มากยิ่งขึ้น วิทยานิพนธ์นี้จึงได้นำเสนอแบบจำลองคุณภาพซอฟต์แวร์โอเพนซอร์ซใหม่ที่ชื่อว่าโอเอสเอส-เอคิวเอ็ม โดยมีเป้าหมายเพื่อการวัดคุณภาพซอฟต์แวร์โอเพนซอร์ซอย่างอัตโนมัติ แบบจำลองโอเอสเอส-เอคิวเอ็มได้นำเสนอตัววัดคุณภาพและเครื่องมืออัตโนมัติที่สามารถดึงข้อมูลเกี่ยวกับซอฟต์แวร์โอเพนซอร์ซจากกิตฮับ ซอร์ซโค้ด โซนาร์คิวบ์ และสแต็กเอกซ์เชนจ์ ทำให้สามารถกำหนดคะแนนคุณภาพของซอฟต์แวร์โอเพนซอร์ซได้ โอเอสเอส-เอคิวเอ็มได้รับการตรวจสอบจากวิศวกรซอฟต์แวร์ที่มีประสบการณ์ในการเลือกใช้ซอฟต์แวร์โอเพนซอร์ซ นอกจากนี้การจัดลำดับซอฟต์แวร์โอเพนซอร์ซโดยเครื่องมือโอเอสเอส-เอคิวเอ็มยังถูกนำไปเปรียบเทียบกับการจัดลำดับด้วยวิธีอื่น และพบว่าการจัดลำดับของโอเอสเอส-เอคิวเอ็มมีสหสัมพันธ์ระดับต่ำมากถึงปานกลางในทิศทางตรงกันข้ามกับวิธีจัดลำดับอื่น ๆ ตามความคิดเห็นและความนิยมของผู้ใช้ และมีสหสัมพันธ์ระดับปานกลางในทิศทางเดียวกันกับวิธีการจัดลำดับอื่นที่เน้นการตรวจสอบความมั่นคงที่ซอฟต์แวร์โดยตรง ทั้งนี้เนื่องจากโอเอสเอส-เอคิวเอ็มครอบคลุมปัจจัยคุณภาพหลายอย่างที่ไม่ได้ถูกพิจารณาโดยวิธีการจัดลำดับอื่น ๆ ดังกล่าว โอเอสเอส-เอคิวเอ็มจึงให้ข้อมูลด้านคุณภาพของซอฟต์แวร์โอเพนซอร์ซในเชิงลึกที่ดีกว่า
Developing Reactive Distributed Aerial Robotics Platforms For Real-Time Contaminant Mapping,
2022
University of Kentucky
Developing Reactive Distributed Aerial Robotics Platforms For Real-Time Contaminant Mapping, Joshua Ashley
Theses and Dissertations--Electrical and Computer Engineering
The focus of this research is to design a sensor data aggregation system and centralized sensor-driven trajectory planning algorithm for fixed-wing aircraft to optimally assist atmospheric simulators in mapping the local environment in real-time. The proposed application of this work is to be used in the event of a hazardous contaminant leak into the atmosphere as a fleet of sensing unmanned aerial vehicles (UAVs) could provide valuable information for evacuation measures. The data aggregation system was designed using a state-of-the-art networking protocol and radio with DigiMesh and a process/data management system in the ROS2 DDS. This system was tested to …
Core Point Pixel-Level Localization By Fingerprint Features In Spatial Domain,
2022
Old Dominion University
Core Point Pixel-Level Localization By Fingerprint Features In Spatial Domain, Xueyi Ye, Yuzhong Shen, Maosheng Zeng, Yirui Liu, Huahua Chen, Zhijing Zhao
Computational Modeling & Simulation Engineering Faculty Publications
Singular point detection is a primary step in fingerprint recognition, especially for fingerprint alignment and classification. But in present there are still some problems and challenges such as more false-positive singular points or inaccurate reference point localization. This paper proposes an accurate core point localization method based on spatial domain features of fingerprint images from a completely different viewpoint to improve the fingerprint core point displacement problem of singular point detection. The method first defines new fingerprint features, called furcation and confluence, to represent specific ridge/valley distribution in a core point area, and uses them to extract the innermost Curve …
Biomimetic Design, Modeling, And Adaptive Control Of Robotic Gripper For Optimal Grasping,
2022
University of Central Florida
Biomimetic Design, Modeling, And Adaptive Control Of Robotic Gripper For Optimal Grasping, Mushtaq Al-Mohammed
Electronic Theses and Dissertations, 2020-2023
Grasping is an essential skill for almost every assistive robot. Variations in shape and/or weight of different objects involved in Activities of Daily Living (ADL) lead to complications, especially, when the robot is trying to grip novel objects for which it has no prior information –too much force will deform or crush the object while too little force will lead to slipping and possibly dropped objects. Thus, successful grasping requires the gripper to immobilize an object with the minimal force. In Chapter 2, we present the design, analysis, and experimental implementation of an adaptive control to facilitate 1-click grasping of …
Energy And Area Efficient Machine Learning Architectures Using Spin-Based Neurons,
2022
University of Central Florida
Energy And Area Efficient Machine Learning Architectures Using Spin-Based Neurons, Hossein Pourmeidani
Electronic Theses and Dissertations, 2020-2023
Recently, spintronic devices with low energy barrier nanomagnets such as spin orbit torque-Magnetic Tunnel Junctions (SOT-MTJs) and embedded magnetoresistive random access memory (MRAM) devices are being leveraged as a natural building block to provide probabilistic sigmoidal activation functions for RBMs. In this dissertation research, we use the Probabilistic Inference Network Simulator (PIN-Sim) to realize a circuit-level implementation of deep belief networks (DBNs) using memristive crossbars as weighted connections and embedded MRAM-based neurons as activation functions. Herein, a probabilistic interpolation recoder (PIR) circuit is developed for DBNs with probabilistic spin logic (p-bit)-based neurons to interpolate the probabilistic output of the neurons …
Using Long Short-Term Memory Networks To Make And Train Neural Network Based Pseudo Random Number Generator,
2022
South Dakota State University
Using Long Short-Term Memory Networks To Make And Train Neural Network Based Pseudo Random Number Generator, Aditya Harshvardhan
Electronic Theses and Dissertations
Neural Networks have been used in many decision-making models and been employed in computer vision, and natural language processing. Several works have also used Neural Networks for developing Pseudo-Random Number Generators [2, 4, 5, 7, 8]. However, despite great performance in the National Institute of Standards and Technology (NIST) statistical test suite for randomness, they fail to discuss how the complexity of a neural network affects such statistical results. This work introduces: 1) a series of new Long Short- Term Memory Network (LSTM) based and Fully Connected Neural Network (FCNN – baseline [2] + variations) Pseudo Random Number Generators (PRNG) …
Formation Control With Bounded Controls And Collision Avoidance: Theory And Application To Quadrotor Unmanned Air Vehicles,
2022
University of Kentucky
Formation Control With Bounded Controls And Collision Avoidance: Theory And Application To Quadrotor Unmanned Air Vehicles, Zachary S. Lippay
Theses and Dissertations--Mechanical and Aerospace Engineering
This dissertation presents new results on multi-agent formation control and applies the new control algorithms to quadrotor unmanned air vehicles. First, this dissertation presents a formation control algorithm for double-integrator agents, where the formation is time varying and the agents’ controls satisfy a priori bounds (e.g., the controls accommodate actuator saturation). The main analytic results provide sufficient conditions such that all agents converge to the desired time-varying relative positions with one another and the leader, and have a priori bounded controls (if applicable). We also present results from rotorcraft experiments that demonstrate the algorithm with time-varying formations and bounded controls. …
