Computer Sciences Commons^™

Waste Cooking Oil Classification Using Artificial Intelligence Technology, Kar Sin Lau

Student Works (2020-2029)

Palm oil – one of the most common edible oil consumed in Malaysia. It is because Malaysia is one of the countries which supply palm oil to the global market and it is cheap to obtain for the consumer in Malaysia. Most of the Malaysian consume it via food preparation such as deep-frying and cooking. However, due to widely available for Malaysians, consumers also lacking awareness in dealing after using the edible oil. Most of the household consumers discard excess waste cooking oil (WCO) into sewage and with courtesy, some of them stored them in containers and sell to NGOs. …

Emergency Department Systems Engineering: Modelling And Event Simulation, O Alharethi Salman Zayed

Student Works (2020-2029)

Emergency department systems are meant to ensure the timely delivery for essential healthcare services through complex systems. (ED) offers various services and have several components in its parts. Examined long previous research on system engineering and quality of healthcare resulting on these are influenced by operational management. Real time data are very helpful for decision making for any systems. Patient output feedback and satisfaction can be ensured with engineered management. EDs have become evident particularly in current scenario when the world is in the grip of the deadly COVID-19. All this implies that analysis is significant for improvement of services …

Off-Policy Reinforcement Learning For Efficient And Effective Gan Architecture Search, Tian Yuan, Wang Qin, Zhiwu Huang, Wen Li, Dengxin Dai, Minghao Yang, Jun Wang, Olga Fink

Research Collection School Of Computing and Information Systems

In this paper, we introduce a new reinforcement learning (RL) based neural architecture search (NAS) methodology for effective and efficient generative adversarial network (GAN) architecture search. The key idea is to formulate the GAN architecture search problem as a Markov decision process (MDP) for smoother architecture sampling, which enables a more effective RL-based search algorithm by targeting the potential global optimal architecture. To improve efficiency, we exploit an off-policy GAN architecture search algorithm that makes efficient use of the samples generated by previous policies. Evaluation on two standard benchmark datasets (i.e., CIFAR-10 and STL-10) demonstrates that the proposed method is …

An Ensemble Of Epoch-Wise Empirical Bayes For Few-Shot Learning, Yaoyao Liu, Bernt Schiele, Qianru Sun

Research Collection School Of Computing and Information Systems

Few-shot learning aims to train efficient predictive models with a few examples. The lack of training data leads to poor models that perform high-variance or low-confidence predictions. In this paper, we propose to meta-learn the ensemble of epoch-wise empirical Bayes models (E3BM) to achieve robust predictions. “Epoch-wise'' means that each training epoch has a Bayes model whose parameters are specifically learned and deployed. ”Empirical'' means that the hyperparameters, e.g., used for learning and ensembling the epoch-wise models, are generated by hyperprior learners conditional on task-specific data. We introduce four kinds of hyperprior learners by considering inductive vs. transductive, and epoch-dependent …

An Attention-Based Rumor Detection Model With Tree-Structured Recursive Neural Networks, Jing Ma, Wei Gao, Shafiq Joty, Kam-Fai Wong

Research Collection School Of Computing and Information Systems

Rumor spread in social media severely jeopardizes the credibility of online content. Thus, automatic debunking of rumors is of great importance to keep social media a healthy environment. While facing a dubious claim, people often dispute its truthfulness sporadically in their posts containing various cues, which can form useful evidence with long-distance dependencies. In this work, we propose to learn discriminative features from microblog posts by following their non-sequential propagation structure and generate more powerful representations for identifying rumors. For modeling non-sequential structure, we first represent the diffusion of microblog posts with propagation trees, which provide valuable clues on how …

Commanding And Re-Dictation: Developing Eyes-Free Voice-Based Interaction For Editing Dictated Text, Debjyoti Ghosh, Can Liu, Shengdong Zhao, Kotaro Hara

Research Collection School Of Computing and Information Systems

Existing voice-based interfaces have limited support for text editing, especially when seeing the text is difficult, e.g., while walking or cooking. This research develops voice interaction techniques for eyes-free text editing. First, with a Wizard-of-Oz study, we identified two primary user strategies: using commands, e.g., “replace go with goes” and re-dictating over an erroneous portion, e.g., correcting “he go there” by saying “he goes there.” To support these user strategies with an actual system implementation, we developed two eyes-free voice interaction techniques, Commanding and Re-dictation, and evaluated them with a controlled experiment. Results showed that while Re-dictation performs significantly better …

Learning Transferrable Parameters For Long-Tailed Sequential User Behavior Modeling, Jianwen Yin, Chenghao Liu, Weiqing Wang, Jianling Sun, Steven C. H. Hoi

Research Collection School Of Computing and Information Systems

Sequential user behavior modeling plays a crucial role in online user-oriented services, such as product purchasing, news feed consumption, and online advertising. The performance of sequential modeling heavily depends on the scale and quality of historical behaviors. However, the number of user behaviors inherently follows a long-tailed distribution, which has been seldom explored. In this work, we argue that focusing on tail users could bring more benefits and address the long tails issue by learning transferrable parameters from both optimization and feature perspectives. Specifically, we propose a gradient alignment optimizer and adopt an adversarial training scheme to facilitate knowledge transfer …

An Energy Efficient Routing Approach For Iot Enabled Underwater Wsns In Smart Cities, Nighat Usman, Omar Alfandi, Saeeda Usman, Asad Masood Khattak, Muhammad Awais, Bashir Hayat, Ahthasham Sajid

All Works

© 2020 by the authors. Licensee MDPI, Basel, Switzerland. Nowadays, there is a growing trend in smart cities. Therefore, Terrestrial and Internet of Things (IoT) enabled Underwater Wireless Sensor Networks (TWSNs and IoT-UWSNs) are mostly used for observing and communicating via smart technologies. For the sake of collecting the desired information from the underwater environment, multiple acoustic sensors are deployed with limited resources, such as memory, battery, processing power, transmission range, etc. The replacement of resources for a particular node is not feasible due to the harsh underwater environment. Thus, the resources held by the node needs to be used …

Hierarchical Syntactic Models For Human Activity Recognition Through Mobility Traces, Enrico Casella, Marco Ortolani, Simone Silvestri, Sajal K. Das

Computer Science Faculty Research & Creative Works

Recognizing users’ daily life activities without disrupting their lifestyle is a key functionality to enable a broad variety of advanced services for a Smart City, from energy-efficient management of urban spaces to mobility optimization. In this paper, we propose a novel method for human activity recognition from a collection of outdoor mobility traces acquired through wearable devices. Our method exploits the regularities naturally present in human mobility patterns to construct syntactic models in the form of finite state automata, thanks to an approach known as grammatical inference. We also introduce a measure of similarity that accounts for the intrinsic hierarchical …

Evaluating Active Traffc Management (Atm) Strategies Under Non-Recurring Congestion: Simulation-Based With Benefit Cost Analysis Case Study, Siham G. Farrag, Fatma Outay, Ansar Ul Haque Yasar, Moulay Youssef El-Hansali

All Works

© 2020 by the authors. Dynamic hard shoulder running and ramp closure are two active traffic management (ATM) strategies that are used to alleviate highway traffic congestion. This study aims to evaluate the effects of these two strategies on congested freeways under non-recurring congestion. The study's efforts can be considered in two parts. First, we performed a detailed microsimulation analysis to quantify the potential benefits of these two ATM strategies in terms of safety, traffic operation, and environmental impact. Second, we evaluated the implementation feasibility of these two strategies. The simulation results indicated that the implementation of the hard shoulder …

Predictive Insights For Improving The Resilience Of Global Food Security Using Artificial Intelligence, Meng Leong How, Yong Jiet Chan, Sin Mei Cheah

Research Collection Lee Kong Chian School Of Business

Unabated pressures on food systems affect food security on a global scale. A human-centric artificial intelligence-based probabilistic approach is used in this paper to perform a unified analysis of data from the Global Food Security Index (GFSI). The significance of this intuitive probabilistic reasoning approach for predictive forecasting lies in its simplicity and user-friendliness to people who may not be trained in classical computer science or in software programming. In this approach, predictive modeling using a counterfactual probabilistic reasoning analysis of the GFSI dataset can be utilized to reveal the interplay and tensions between the variables that underlie food affordability, …

Machine Learning Enhanced Free-Space And Underwater Oam Optical Communications, Patrick L. Neary

All Graduate Theses and Dissertations, Spring 1920 to Summer 2023

Communications, bandwidth, security, and hardware simplicity are principles of interest to society at large. Recent advances in optics and in understanding properties of light, such as orbital angular momentum (OAM), have provided new potential mediums for communication.

Machine learning has wound its way into a broad range of fascinating areas. An emerging field of research is the use of a unique property of lasers called orbital angular momentum (OAM). With the proper hardware, a laser can go from a Gaussian shaped distribution to a doughnut shaped pattern, where the radius can be changed. Multiple OAM patterns, or modes, can be …

Bootstrapping Web Archive Collections From Micro-Collections In Social Media, Alexander C. Nwala

Computer Science Theses & Dissertations

In a Web plagued by disappearing resources, Web archive collections provide a valuable means of preserving Web resources important to the study of past events. These archived collections start with seed URIs (Uniform Resource Identifiers) hand-selected by curators. Curators produce high quality seeds by removing non-relevant URIs and adding URIs from credible and authoritative sources, but this ability comes at a cost: it is time consuming to collect these seeds. The result of this is a shortage of curators, a lack of Web archive collections for various important news events, and a need for an automatic system for generating seeds. …

How Practitioners Perceive Automated Bug Report Management Techniques, Weiqin Zou, David Lo, Zhenyu Chen, Xin Xia, Yang Feng, Baowen Xu

Research Collection School Of Computing and Information Systems

Bug reports play an important role in the process of debugging and fixing bugs. To reduce the burden of bug report managers and facilitate the process of bug fixing, a great amount of software engineering research has been invested into automated bug report management techniques. However, the verdict is still open whether such techniques are actually required and applicable outside of the theoretical research domain. To fill this gap, in this paper, we conducted a survey among 327 practitioners to gain their insights into various categories of automated bug report management techniques. Specifically, in the survey, we asked them to …

Dual-Dropout Graph Convolutional Network For Predicting Synthetic Lethality In Human Cancers, Ruichu Cai, Xuexin Chen, Yuan Fang, Min Wu, Yuexing Hao

Research Collection School Of Computing and Information Systems

Motivation: Synthetic lethality (SL) is a promising form of gene interaction for cancer therapy, as it is able to identify specific genes to target at cancer cells without disrupting normal cells. As high-throughput wet-lab settings are often costly and face various challenges, computational approaches have become a practical complement. In particular, predicting SLs can be formulated as a link prediction task on a graph of interacting genes. Although matrix factorization techniques have been widely adopted in link prediction, they focus on mapping genes to latent representations in isolation, without aggregating information from neighboring genes. Graph convolutional networks (GCN) can capture …

A Longitudinal Study Of A Capstone Course, Benjamin Gan, Eng Lieh Ouh, Yin Yin Fiona Lee

Research Collection School Of Computing and Information Systems

This is a 7 years study on a capstone course completed by 1700+ students for 200+ organizations involving 300+ projects. Student teams deliver a system to solve real-world problems proposed by industry partners. We want to understand what independent variables influence student performance. We analyzed the deployment status of systems delivered, the type of organization/industry, the number of meetings and the technology used. Our results show some organization value proof of concept over fully deployed systems, student strengths are in Infocomm and Finance projects, the number of meetings is a weak correlation to performance and best performing projects are fully …

A Multicut Outer-Approximation Approach For Competitive Facility Location Under Random Utilities, Tien Mai, Andrea Lodi

Research Collection School Of Computing and Information Systems

This work concerns the maximum capture facility location problem with random utilities, i.e., the problem of seeking to locate new facilities in a competitive market such that the captured demand of users is maximized, assuming that each individual chooses among all available facilities according to a random utility maximization model. The main challenge lies in the nonlinearity of the objective function. Motivated by the convexity and separable structure of such an objective function, we propose an enhanced implementation of the outer approximation scheme. Our algorithm works in a cutting plane fashion and allows to separate the objective function into a …

Rethinking Pruning For Accelerating Deep Inference At The Edge, Dawei Gao, Xiaoxi He, Zimu Zhou, Yongxin Tong, Ke Xu, Lothar Thiele

Research Collection School Of Computing and Information Systems

There is a growing trend to deploy deep neural networks at the edge for high-accuracy, real-time data mining and user interaction. Applications such as speech recognition and language understanding often apply a deep neural network to encode an input sequence and then use a decoder to generate the output sequence. A promising technique to accelerate these applications on resource-constrained devices is network pruning, which compresses the size of the deep neural network without severe drop in inference accuracy. However, we observe that although existing network pruning algorithms prove effective to speed up the prior deep neural network, they lead to …

Adaptive Task Sampling For Meta-Learning, Chenghao Liu, Zhihao Wang, Doyen Sahoo, Yuan Fang, Kun Zhang, Steven C. H. Hoi

Research Collection School Of Computing and Information Systems

Meta-learning methods have been extensively studied and applied in computer vision, especially for few-shot classification tasks. The key idea of meta-learning for few-shot classification is to mimic the few-shot situations faced at test time by randomly sampling classes in meta-training data to construct fewshot tasks for episodic training. While a rich line of work focuses solely on how to extract meta-knowledge across tasks, we exploit the complementary problem on how to generate informative tasks. We argue that the randomly sampled tasks could be sub-optimal and uninformative (e.g., the task of classifying “dog” from “laptop” is often trivial) to the meta-learner. …

Feature Pyramid Transformer, Dong Zhang, Hanwang Zhang, Jinhui Tang, Meng Wang, Xian-Sheng Hua, Qianru Sun

Research Collection School Of Computing and Information Systems

Feature interactions across space and scales underpin modern visual recognition systems because they introduce beneficial visual contexts. Conventionally, spatial contexts are passively hidden in the CNN’s increasing receptive fields or actively encoded by non-local convolution. Yet, the non-local spatial interactions are not across scales, and thus they fail to capture the non-local contexts of objects (or parts) residing in different scales. To this end, we propose a fully active feature interaction across both space and scales, called Feature Pyramid Transformer (FPT). It transforms any feature pyramid into another feature pyramid of the same size but with richer contexts, by using …

Prevalence, Contents And Automatic Detection Of Kl-Satd, Leevi Rantala, Mika Mantyla, David Lo

Research Collection School Of Computing and Information Systems

When developers use different keywords such as TODO and FIXME in source code comments to describe self-admitted technical debt (SATD), we refer it as Keyword-Labeled SATD (KL-SATD). We study KL-SATD from 33 software repositories with 13,588 KL-SATD comments. We find that the median percentage of KL-SATD comments among all comments is only 1,52%. We find that KL-SATD comment contents include words expressing code changes and uncertainty, such as remove, fix, maybe and probably. This makes them different compared to other comments. KL-SATD comment contents are similar to manually labeled SATD comments of prior work. Our machine learning classifier using logistic …

Intersentiment: Combining Deep Neural Models On Interaction And Sentiment For Review Rating Prediction, Shi Feng, Kaisong Song, Daling Wang, Wei Gao, Yifei Zhang

Research Collection School Of Computing and Information Systems

Review rating prediction is commonly approached from the perspective of either Collaborative Filtering (CF) or Sentiment Classification (SC). CF-based approach usually resorts to matrix factorization based on user–item interaction, and does not fully utilize the valuable review text features. In contrast, SC-based approach is focused on mining review content, but can just incorporate some user- and product-level features, and fails to capture sufficient interactions between them represented typically in a sparse matrix as CF can do. In this paper, we propose a novel, extensible review rating prediction model called InterSentiment by bridging the user-product interaction model and the sentiment model …

A Unified Framework For Sparse Online Learning, Peilin Zhao, Dayong Wong, Pengcheng Wu, Steven C. H. Hoi

Research Collection School Of Computing and Information Systems

The amount of data in our society has been exploding in the era of big data. This article aims to address several open challenges in big data stream classification. Many existing studies in data mining literature follow the batch learning setting, which suffers from low efficiency and poor scalability. To tackle these challenges, we investigate a unified online learning framework for the big data stream classification task. Different from the existing online data stream classification techniques, we propose a unified Sparse Online Classification (SOC) framework. Based on SOC, we derive a second-order online learning algorithm and a cost-sensitive sparse online …

Accelerating Exact Constrained Shortest Paths On Gpus, Shengliang Lu, Bingsheng He, Yuchen Li, Hao Fu

Research Collection School Of Computing and Information Systems

The recently emerging applications such as software-defined networks and autonomous vehicles require efficient and exact solutions for constrained shortest paths (CSP), which finds the shortest path in a graph while satisfying some user-defined constraints. Compared with the common shortest path problems without constraints, CSP queries have a significantly larger number of subproblems. The most widely used labeling algorithm becomes prohibitively slow and impractical. Other existing approaches tend to find approximate solutions and build costly indices on graphs for fast query processing, which are not suitable for emerging applications with the requirement of exact solutions. A natural question is whether and …

An Analysis Of Sketched Irls For Accelerated Sparse Residual Regression, Daichi Iwata, Michael Waechter, Wen-Yan Lin, Yasuyuki Matsushita

Research Collection School Of Computing and Information Systems

This paper studies the problem of sparse residual regression, i.e., learning a linear model using a norm that favors solutions in which the residuals are sparsely distributed. This is a common problem in a wide range of computer vision applications where a linear system has a lot more equations than unknowns and we wish to find the maximum feasible set of equations by discarding unreliable ones. We show that one of the most popular solution methods, iteratively reweighted least squares (IRLS), can be significantly accelerated by the use of matrix sketching. We analyze the convergence behavior of the proposed method …

A Generalised Bound For The Wiener Attack On Rsa, Willy Susilo, Joseph Tonien, Guomin Yang

Research Collection School Of Computing and Information Systems

Since Wiener pointed out that the RSA can be broken if the private exponent d is relatively small compared to the modulus N, it has been a general belief that the Wiener attack works for d

Aim 2020 Challenge On Video Extreme Super-Resolution: Methods And Results, D. Fuoli, Zhiwu Huang, S. Gu, R. Timofte, A. Raventos, A. Esfandiari, S. Karout, X. Xu, X. Li, X. Xiong, J. Wang, Michelini P. Navarrete, W. Zhang, D. Zhang, H. Zhu, D. Xia, H. Chen, J. Gu, Z. Zhang, T. Zhao

Research Collection School Of Computing and Information Systems

This paper reviews the video extreme super-resolution challenge associated with the AIM 2020 workshop at ECCV 2020. Common scaling factors for learned video super-resolution (VSR) do not go beyond factor 4. Missing information can be restored well in this region, especially in HR videos, where the high-frequency content mostly consists of texture details. The task in this challenge is to upscale videos with an extreme factor of 16, which results in more serious degradations that also affect the structural integrity of the videos. A single pixel in the lowresolution (LR) domain corresponds to 256 pixels in the high-resolution (HR) domain. …

Measuring Privacy Concerns With Government Surveillance And Right-To-Be-Forgotten In Nomological Net Of Trust And Willingness-To-Share, Gaurav Bansal, Fiona Fui-Hoon Nah

Research Collection School Of Computing and Information Systems

In the post Snowden revelations era, concerns related to government surveillance and oversight have come to the forefront. The ability of the Internet to remember “everything” (or forget anything) also raises a privacy concern associated with the “right to be forgotten”. Hence, in this paper, we propose and examine privacy concerns by extending the Hong and Thong’s (2013) model with the addition of two dimensions: right to be forgotten as well as government surveillance and oversight. We tested two different measurement models using privacy concerns as a second-order and a third-order construct within a nomological net that includes trusting beliefs …

Through-The-Wall Radar Detection Using Machine Learning, Aihua W. Wood, Ryan Wood, Matthew Charnley

Faculty Publications

This paper explores the through-the-wall inverse scattering problem via machine learning. The reconstruction method seeks to discover the shape, location, and type of hidden objects behind walls, as well as identifying certain material properties of the targets. We simulate RF sources and receivers placed outside the room to generate observation data with objects randomly placed inside the room. We experiment with two types of neural networks and use an 80-20 train-test split for reconstruction and classification.

Towards Dynamic Vehicular Clouds, Aida Ghazizadeh

Computer Science Theses & Dissertations

Motivated by the success of the conventional cloud computing, Vehicular Clouds were introduced as a group of vehicles whose corporate computing, sensing, communication, and physical resources can be coordinated and dynamically allocated to authorized users. One of the attributes that set Vehicular Clouds apart from conventional clouds is resource volatility. As vehicles enter and leave the cloud, new computing resources become available while others depart, creating a volatile environment where the task of reasoning about fundamental performance metrics becomes very challenging. The goal of this thesis is to design an architecture and model for a dynamic Vehicular Cloud built on …