Numerical Analysis and Scientific Computing Commons^™

Visualized Algorithm Engineering On Two Graph Partitioning Problems, Zizhen Chen

Computer Science and Engineering Theses and Dissertations

Concepts of graph theory are frequently used by computer scientists as abstractions when modeling a problem. Partitioning a graph (or a network) into smaller parts is one of the fundamental algorithmic operations that plays a key role in classifying and clustering. Since the early 1970s, graph partitioning rapidly expanded for applications in wide areas. It applies in both engineering applications, as well as research. Current technology generates massive data (“Big Data”) from business interactions and social exchanges, so high-performance algorithms of partitioning graphs are a critical need.

This dissertation presents engineering models for two graph partitioning problems arising from completely …

Understanding The Impacts Of Topobathymetric Data On Storm Surge Model Predictions, Sydni Crain

Master's Theses

The topobathymetric characteristics of a region are regularly altered by natural and anthropogenic causes. This directly impacts the resulting storm surge during a hurricane. The primary goal of this research was to gain a better understanding of the impact that topography and bathymetry have on storm surge models, particularly the Advanced Circulation (ADCIRC) Model. Hurricane Zeta (2020) and Hurricane Ida (2021) were chosen as case studies; therefore, the Gulf of Mexico (GOM) was chosen as the study site. This research was completed by comparing ADCIRC storm surge results which were based on older, lower-resolution data with results derived from more …

Monolithic Multiphysics Simulation Of Hypersonic Aerothermoelasticity Using A Hybridized Discontinuous Galerkin Method, William Paul England

Theses and Dissertations

This work presents implementation of a hybridized discontinuous Galerkin (DG) method for robust simulation of the hypersonic aerothermoelastic multiphysics system. Simulation of hypersonic vehicles requires accurate resolution of complex multiphysics interactions including the effects of high-speed turbulent flow, extreme heating, and vehicle deformation due to considerable pressure loads and thermal stresses. However, the state-of-the-art procedures for hypersonic aerothermoelasticity are comprised of low-fidelity approaches and partitioned coupling schemes. These approaches preclude robust design and analysis of hypersonic vehicles for a number of reasons. First, low-fidelity approaches limit their application to simple geometries and lack the ability to capture small scale flow …

Geochemical Analysis And Numerical Modeling Of Central And East Tennessee Mississippi Valley-Type Ore Districts: Constraints On Ore Genesis, Jackson Price Copeland

Geosciences Undergraduate Honors Theses

A simple two-way stochastic mixing model is presented for analysis of the lead (Pb) isotope compositions of the North American Mississippi Valley-Type (MVT) districts of East Tennessee, Central Tennessee, and Central Kentucky. Four distinct mixing scenarios were run to critically evaluate the stochastic model and examine different hypotheses regarding the genesis of Central Tennessee and Central Kentucky MVT deposits. Additionally, Pb isotope analysis was conducted on sphalerite samples from the Central and East Tennessee MVT districts. Model and sampling results suggest that Central Tennessee and Central Kentucky ores likely formed by mixing of three fluids. In contrast to conclusions from …

Bluetooth Low Energy Indoor Positioning System, Jackson T. Diamond, Jordan Hanson

Whittier Scholars Program

Robust indoor positioning systems based on low energy bluetooth signals will service a wide range of applications. We present an example of a low energy bluetooth positioning system. First, the steps taken to locate the target with the bluetooth data will be reviewed. Next, we describe the algorithms of the set of android apps developed to utilize the bluetooth data for positioning. Similar to GPS, the algorithms use trilateration to approximate the target location by utilizing the corner devices running one of the apps. Due to the fluctuating nature of the bluetooth signal strength indicator (RSSI), we used an averaging …

Brif: A Novel And Efficient Implementation Of Random Forests Based On Bit Packing And Parallel Computing, Yanchao Liu

Industrial and Systems Engineering Faculty Research Publications

Random forests are powerful and popular machine learning methods. While general principles of tree induction are straightforward and well-understood, the numerous algorithmic treatments implemented in software tools, as well as their impacts on performance, are less familiar to most users. This paper introduces a new random forest toolkit (the ‘brif’ package in R and Python) along with its key algorithmic design features, and demonstrates the effects of the forest’s hyper-parameters such as the split search method, tree depth and the voting mechanism, on the classification performance. Summaries of benchmarking experiments are also presented. Results show that ‘brif’ stands out among …

Machine Learning-Based Data And Model Driven Bayesian Uncertanity Quantification Of Inverse Problems For Suspended Non-Structural System, Zhiyuan Qin

All Dissertations

Inverse problems involve extracting the internal structure of a physical system from noisy measurement data. In many fields, the Bayesian inference is used to address the ill-conditioned nature of the inverse problem by incorporating prior information through an initial distribution. In the nonparametric Bayesian framework, surrogate models such as Gaussian Processes or Deep Neural Networks are used as flexible and effective probabilistic modeling tools to overcome the high-dimensional curse and reduce computational costs. In practical systems and computer models, uncertainties can be addressed through parameter calibration, sensitivity analysis, and uncertainty quantification, leading to improved reliability and robustness of decision and …

Generative Stresnet For Crime Prediction, Ba Phong Tran, Hoong Chuin Lau

Research Collection School Of Computing and Information Systems

In this work, we combine STResnet (Zhang et al., 2017) with VAE Kingma & Welling (2013) to generate crime distribution. The outputs can be used for downstream tasks such as patrol deployment planning Chase et al. (2021).

Msrl-Net: A Multi-Level Semantic Relation-Enhanced Learning Network For Aspect-Based Sentiment Analysis, Zhenda Hu, Zhaoxia Wang, Yinglin Wang, Ah-Hwee Tan

Research Collection School Of Computing and Information Systems

Aspect-based sentiment analysis (ABSA) aims to analyze the sentiment polarity of a given text towards several specific aspects. For implementing the ABSA, one way is to convert the original problem into a sentence semantic matching task, using pre-trained language models, such as BERT. However, for such a task, the intra- and inter-semantic relations among input sentence pairs are often not considered. Specifically, the semantic information and guidance of relations revealed in the labels, such as positive, negative and neutral, have not been completely exploited. To address this issue, we introduce a self-supervised sentence pair relation classification task and propose a …

Exploring A Gradient-Based Explainable Ai Technique For Time-Series Data: A Case Study Of Assessing Stroke Rehabilitation Exercises, Min Hun Lee, Yi Jing Choy

Research Collection School Of Computing and Information Systems

Explainable artificial intelligence (AI) techniques are increasingly being explored to provide insights into why AI and machine learning (ML) models provide a certain outcome in various applications. However, there has been limited exploration of explainable AI techniques on time-series data, especially in the healthcare context. In this paper, we describe a threshold-based method that utilizes a weakly supervised model and a gradient-based explainable AI technique (i.e. saliency map) and explore its feasibility to identify salient frames of time-series data. Using the dataset from 15 post-stroke survivors performing three upper-limb exercises and labels on whether a compensatory motion is observed or …

Learning-Based Stock Trending Prediction By Incorporating Technical Indicators And Social Media Sentiment, Zhaoxia Wang, Zhenda Hu, Fang Li, Seng-Beng Ho, Erik Cambria

Research Collection School Of Computing and Information Systems

Stock trending prediction is a challenging task due to its dynamic and nonlinear characteristics. With the development of social platform and artificial intelligence (AI), incorporating timely news and social media information into stock trending models becomes possible. However, most of the existing works focus on classification or regression problems when predicting stock market trending without fully considering the effects of different influence factors in different phases. To address this gap, this research solves stock trending prediction problem utilizing both technical indicators and sentiments of the social media text as influence factors in different situations. A 3-phase hybrid model is proposed …

Wearing Masks Implies Refuting Trump?: Towards Target-Specific User Stance Prediction Across Events In Covid-19 And Us Election 2020, Hong Zhang, Haewoon Kwak, Wei Gao, Jisun An

Research Collection School Of Computing and Information Systems

People who share similar opinions towards controversial topics could form an echo chamber and may share similar political views toward other topics as well. The existence of such connections, which we call connected behavior, gives researchers a unique opportunity to predict how one would behave for a future event given their past behaviors. In this work, we propose a framework to conduct connected behavior analysis. Neural stance detection models are trained on Twitter data collected on three seemingly independent topics, i.e., wearing a mask, racial equality, and Trump, to detect people’s stance, which we consider as their online behavior in …

Quantification Of Various Types Of Biases In Large Language Models, Sudhashree Sayenju

Doctor of Data Science and Analytics Dissertations

Natural Language Processing (NLP) systems are included everywhere on the internet from search engines, language translations to more advanced systems like voice assistant and customer service. Since humans are always on the receiving end of NLP technologies, it is very important to analyze whether or not the Large Language Models (LLMs) in use have bias and are therefore unfair. The majority of the research in NLP bias has focused on societal stereotype biases embedded in LLMs. However, our research focuses on all types of biases, namely model class level bias, stereotype bias and domain bias present in LLMs. Model class …

Head And Neck Tumor Histopathological Image Representation With Pre- Trained Convolutional Neural Network And Vision Transformer, Ranny Rahaningrum Herdiantoputri, Daisuke Komura, Tohru Ikeda, Shumpei Ishikawa

Journal of Dentistry Indonesia

Image representation via machine learning is an approach to quantitatively represent histopathological images of head and neck tumors for future applications of artificial intelligence-assisted pathological diagnosis systems. Objective: This study compares image representations produced by a pre-trained convolutional neural network (VGG16) to those produced by a vision transformer (ViT-L/14) in terms of the classification performance of head and neck tumors. Methods: W hole-slide images of five oral t umor categories (n = 319 cases) were analyzed. Image patches were created from manually annotated regions at 4096, 2048, and 1024 pixels and rescaled to 256 pixels. Image representations were …

Dynamic Target Assignment Of Multiple Unmanned Aerial Vehicles Based On Clustering Of Network Nodes, Tuo Zhao, Hanqiang Deng, Jialong Gao, Jian Huang

Journal of System Simulation

Abstract: In order to solve the problem that the distributed multi-UAV target assignment algorithm is prone to communication redundancy, which leads to the large communication scale of formation, a multi-UAV dynamic target assignment algorithm (CU-CBBA) based on node clustering in communication network is proposed.The algorithm introduces the communication network node grouping clustering strategy. According to the node's degree centrality, feature vector centrality, intermediate centrality and other attributes, the network node importance ranking model is established. A group of key nodes in the network topology structure are selected and the network topology node clustering is completed according to the shortest …

Trajectory Control Of Crawler Robot Based On Lstm And Smc, Dongyang Liu, Wenwen Zha, Liang Tao, Cheng Zhu, Lichuan Gu, Jun Jiao

Journal of System Simulation

Abstract: Trajectory tracking is an important part of mobile robot control technology and possesses prospect. Highly nonlinear dynamic characteristics are the main obstacles of controller design. A SMC method based on LSTM and quasi-sliding mode is proposed. The kinematics model and dynamics model of the tracked vehicle are given, and the sliding mode control system is established based on the dynamics model. LSTM network based on deep learning method is designed to control and compensate the unknown interference items, reduce the influence of external interference, and reduce the tremor phenomenon by combining the advantages of LSTM network and quasi-sliding …

Research On Modeling And Scheduling Of Virtual Power Plant With Dual Demand Response, Qiang Chen, Yi Wang, Kangshun Li

Journal of System Simulation

Abstract: Virtual power plant technology provides an effective means to aggregate distributed power and user side resources to participate in power scheduling. Most of the existing research focus on the scheduling optimization of distributed energy instead of the demand response of user side. The user side resources are divided into contracted reliable response load and non-contracted random response load, and the load response is regulated through price adjustment mechanism to adapt to the change of distributed. A virtual power plant optimal scheduling model with dual demands response is constructed, in which the maximizing overall profit of the power grid is …

Voltage And Reactive Power Combinational Evaluation Of Regional Power Grid Based On Ewm-Ahp-Bp Neural Network, Yuqi Ji, Huan Xie, Shaoyu Shi, Ping He, Nan Jin, Huili Wang

Journal of System Simulation

Abstract: In order to quantitatively evaluate the influence of renewable energy access on voltage and reactive power operation, a combinational evaluation method of voltage and reactive power based on EWM-AHP-BP neural network is proposed to carry out the multi-objective evaluation weight calculation. Considering voltage qualified rate, voltage fluctuation, power factor qualified rate and reactive power reserve, the comprehensive evaluation model is established. The operation data of renewable energy and power load are clustered to divide the typical scenarios and the evaluation model under multiple scenarios is scored by the combination method of entropy weight method and analytic hierarchy process. The …

Research On Modeling And Solution Method Of Operational Tasks Assignment, Yue Ma, Lin Wu, Shengming Guo

Journal of System Simulation

Abstract: Aiming at the prewar operational tasks assignment in operation task planning, a multi constraint model of operational tasks assignment is constructed to describe the dynamic mapping relationship between operational tasks and operational units. The solution strategy of decision space pruning and constraint condition judgment is proposed, and the methods of decision variable coding, assignment scheme decoding and phased fitness calculation are described. Differential evolution algorithm is used to work out the solution. The experimental results show that the multi constraint assignment model and solution algorithm can effectively reduce the scale of decision space, and can improve the rationality …

Knowledge Graph-Based Process Knowledge Reasoning Method For Intelligent Production System, Weikai Yang, Yan Wang, Zhicheng Ji

Journal of System Simulation

Abstract: Aiming at the disadvantages of high redundancy and weakness between knowledge and data in intelligent production system, and the difficulty to perform knowledge reasoning, a process knowledge reasoning method for knowledge maps is proposed. The input information is semantically labeled and classified, the characteristics of the information match are extracted, the extracted local feature and global feature are associated through graph convolution method, and the feature of the difference value information is integrated and mapped with the constructed knowledge graph. Different reasoning rules are used according to different reasoning types, and the association and topology information between instances are …