Databases and Information Systems Commons^™

Causal Intervention For Weakly-Supervised Semantic Segmentation, Zhang Dong, Hanwang Zhang, Jinhui Tang, Xian-Sheng Hua, Qianru Sun

Research Collection School Of Computing and Information Systems

We present a causal inference framework to improve Weakly-Supervised Semantic Segmentation (WSSS). Specifically, we aim to generate better pixel-level pseudo-masks by using only image-level labels --- the most crucial step in WSSS. We attribute the cause of the ambiguous boundaries of pseudo-masks to the confounding context, e.g., the correct image-level classification of "horse'' and "person'' may be not only due to the recognition of each instance, but also their co-occurrence context, making the model inspection (e.g., CAM) hard to distinguish between the boundaries. Inspired by this, we propose a structural causal model to analyze the causalities among images, contexts, and …

Debunking Rumors On Twitter With Tree Transformer, Jing Ma, Wei Gao

Research Collection School Of Computing and Information Systems

Rumors are manufactured with no respect for accuracy, but can circulate quickly and widely by "word-of-post" through social media conversations. Conversation tree encodes important information indicative of the credibility of rumor. Existing conversation-based techniques for rumor detection either just strictly follow tree edges or treat all the posts fully-connected during feature learning. In this paper, we propose a novel detection model based on tree transformer to better utilize user interactions in the dialogue where post-level self-attention plays the key role for aggregating the intra-/inter-subtree stances. Experimental results on the TWITTER and PHEME datasets show that the proposed approach consistently improves …

A Bert-Based Dual Embedding Model For Chinese Idiom Prediction, Minghuan Tan, Jing Jiang

Research Collection School Of Computing and Information Systems

Chinese idioms are special fixed phrases usually derived from ancient stories, whose meanings are oftentimes highly idiomatic and non-compositional. The Chinese idiom prediction task is to select the correct idiom from a set of candidate idioms given a context with a blank. We propose a BERT-based dual embedding model to encode the contextual words as well as to learn dual embeddings of the idioms. Specifically, we first match the embedding of each candidate idiom with the hidden representation corresponding to the blank in the context. We then match the embedding of each candidate idiom with the hidden representations of all …

A Social Network Analysis Of Jobs And Skills, Derrick Ming Yang Lee, Dion Wei Xuan Ang, Grace Mei Ching Pua, Lee Ning Ng, Sharon Purbowo, Eugene Wen Jia Choy, Kyong Jin Shim

Research Collection School Of Computing and Information Systems

In this study, we analyzed job roles and skills across industries in Singapore. Using social network analysis, we identified job roles with similar required skills, and we also identified relationships between job skills. Our analysis visualizes such relationships in an intuitive way. Insights derived from our analyses are expected to assist job seekers, employers as well as recruitment agencies wanting to understand trending and required job roles and skills in today’s fast changing world.

Social Media Analytics: A Case Study Of Singapore General Election 2020, Sebastian Zhi Tao Khoo, Leong Hock Ho, Ee Hong Lee, Danston Kheng Boon Goh, Zehao Zhang, Swee Hong Ng, Haodi Qi, Kyong Jin Shim

Research Collection School Of Computing and Information Systems

The 2020 Singaporean General Election (GE2020) was a general election held in Singapore on July 10, 2020. In this study, we present an analysis on social conversations about GE2020 during the election period. We analyzed social conversations from popular platforms such as Twitter, HardwareZone, and TR Emeritus.

Co-Embedding Attributed Networks With External Knowledge, Pei-Chi Lo, Ee Peng Lim

Research Collection School Of Computing and Information Systems

Attributed network embedding aims to learn representations of nodes and their attributes in a low-dimensional space that preserves their semantics. The existing embedding models, however, consider node connectivity and node attributes only while ignoring external knowledge that can enhance node representations for downstream applications. In this paper, we propose a set of new VAE-based embedding models called External Knowledge-Aware Co-Embedding Attributed Network (ECAN) Embeddings to incorporate associations among attributes from relevant external knowledge. Such external knowledge can be extracted from text corpus and knowledge graphs. We use multi-VAE structures to model the attribute associations. To cope with joint encoding of …

A Near-Optimal Change-Detection Based Algorithm For Piecewise-Stationary Combinatorial Semi-Bandits, Huozhi Zhou, Lingda Wang, Lav N. Varshney, Ee-Peng Lim

Research Collection School Of Computing and Information Systems

We investigate the piecewise-stationary combinatorial semi-bandit problem. Compared to the original combinatorial semi-bandit problem, our setting assumes the reward distributions of base arms may change in a piecewise-stationary manner at unknown time steps. We propose an algorithm, GLR-CUCB, which incorporates an efficient combinatorial semi-bandit algorithm, CUCB, with an almost parameter-free change-point detector, the Generalized Likelihood Ratio Test (GLRT). Our analysis shows that the regret of GLR-CUCB is upper bounded by O(√NKT logT), where N is the number of piecewise-stationary segments, K is the number of base arms, and T is the number of time steps. As a complement, we also …

Learning To Dispatch For Job Shop Scheduling Via Deep Reinforcement Learning, Cong Zhang, Wen Song, Zhiguang Cao, Jie Zhang, Puay Siew Tan, Xu Chi

Research Collection School Of Computing and Information Systems

Priority dispatching rule (PDR) is widely used for solving real-world Job-shop scheduling problem (JSSP). However, the design of effective PDRs is a tedious task, requiring a myriad of specialized knowledge and often delivering limited performance. In this paper, we propose to automatically learn PDRs via an end-to-end deep reinforcement learning agent. We exploit the disjunctive graph representation of JSSP, and propose a Graph Neural Network based scheme to embed the states encountered during solving. The resulting policy network is size-agnostic, effectively enabling generalization on large-scale instances. Experiments show that the agent can learn high-quality PDRs from scratch with elementary raw …

Deep Multi-Task Learning For Depression Detection And Prediction In Longitudinal Data, Guansong Pang, Ngoc Thien Anh Pham, Emma Baker, Rebecca Bentley, Anton Van Den Hengel

Research Collection School Of Computing and Information Systems

Depression is among the most prevalent mental disorders, affecting millions of people of all ages globally. Machine learning techniques have shown effective in enabling automated detection and prediction of depression for early intervention and treatment. However, they are challenged by the relative scarcity of instances of depression in the data. In this work we introduce a novel deep multi-task recurrent neural network to tackle this challenge, in which depression classification is jointly optimized with two auxiliary tasks, namely one-class metric learning and anomaly ranking. The auxiliary tasks introduce an inductive bias that improves the classification model's generalizability on small depression …

Heterogeneous Univariate Outlier Ensembles In Multidimensional Data, Guansong Pang, Longbing Cao

Research Collection School Of Computing and Information Systems

In outlier detection, recent major research has shifted from developing univariate methods to multivariate methods due to the rapid growth of multidimensional data. However, one typical issue of this paradigm shift is that many multidimensional data often mainly contains univariate outliers, in which many features are actually irrelevant. In such cases, multivariate methods are ineffective in identifying such outliers due to the potential biases and the curse of dimensionality brought by irrelevant features. Those univariate outliers might be well detected by applying univariate outlier detectors in individually relevant features. However, it is very challenging to choose a right univariate detector …

Design Of A Two-Echelon Freight Distribution System In An Urban Area Considering Third-Party Logistics And Loading-Unloading Zones, Vincent F. Yu, Winarno, Shih-Wei Lin, Aldy Gunawan

Research Collection School Of Computing and Information Systems

This research examines the problem of designing a two-echelon freight distribution system in a dense urban area that considers third-party logistics (TPL) and loading–unloading zones (LUZs). The proposed system takes advantage of outsourcing the last mile deliveries to a TPL provider and utilizing LUZs as temporary intermediate facilities instead of using permanent intermediate facilities to consolidate freight. A mathematical model and a simulated annealing (SA) algorithm are developed to solve the problem. The efficiency and effectiveness of the proposed SA heuristic are verified by testing it on existing benchmark instances. Computational results show that the performance of the proposed SA …

Implementation Of Smartphone Navigation Features By Combined Forces In Determining The Hazards Of Terrorism In Poso, Mahturai Rian Fitra Mrf, Arthur Josias Simon Runturambi Ajsr

Journal of Terrorism Studies

The presence of armed terrorist groups in Poso can threaten security conditions in the country because their activities are considered quite dangerous for the surrounding community. This terrorist group did not hesitate to kill civilians who tried to deny its existence. Therefore, various joint military operations have been launched to crush this armed terrorist group, such as Camar Maleo and Tinombala. However, until now this terrorist group is difficult to destroy, due to the condition of the operating area in the form of dense tropical rainforest and steep slopes. This makes it difficult for troops to carry out chases and …

Making Sense Of Online Public Health Debates With Visual Analytics Systems, Anton Ninkov

Electronic Thesis and Dissertation Repository

Online debates occur frequently and on a wide variety of topics. Particularly, online debates about various public health topics (e.g., vaccines, statins, cannabis, dieting plans) are prevalent in today’s society. These debates are important because of the real-world implications they can have on public health. Therefore, it is important for public health stakeholders (i.e., those with a vested interest in public health) and the general public to have the ability to make sense of these debates quickly and effectively. This dissertation investigates ways of enabling sense-making of these debates with the use of visual analytics systems (VASes). VASes are computational …

Barriers And Drivers Influencing The Growth Of E-Commerce In Uzbekistan, Madinakhon Tursunboeva

Theses and Dissertations

Electronic commerce (e-commerce) has become a major retail channel for businesses in developed countries. However, it is still considered an innovation in developing countries. Specifically, e-commerce in Uzbekistan is in the early stages of emergence despite its advance in recent years in terms of Internet penetration, a strong retail sector, new national regulations, and a young population. The study aimed to identify barriers and drivers influencing e-commerce growth in Uzbekistan. A Delphi research design was utilized to answer the research questions of the study, which categorized and ranked factors that Uzbekistani entrepreneurs are facing when engaging in e-commerce processes. A …

An Analysis Of Technological Components In Relation To Privacy In A Smart City, Kayla Rutherford, Ben Lands, A. J. Stiles

James Madison Undergraduate Research Journal (JMURJ)

A smart city is an interconnection of technological components that store, process, and wirelessly transmit information to enhance the efficiency of applications and the individuals who use those applications. Over the course of the 21st century, it is expected that an overwhelming majority of the world’s population will live in urban areas and that the number of wireless devices will increase. The resulting increase in wireless data transmission means that the privacy of data will be increasingly at risk. This paper uses a holistic problem-solving approach to evaluate the security challenges posed by the technological components that make up a …

An Introduction To Seshat: Global History Databank, Peter Turchin, Harvey Whitehouse, Pieter François, Daniel Hoyer, Abel Alves, John Baines, David Baker, Marta Bartkowiak, Jennifer Bates, James Bennett, Julye Bidmead, Peter Bol, Alessandro Ceccarelli, Kostis Christakis, David Christian, Alan Covey, Franco De Angelis, Timothy K. Earle, Neil R. Edwards, Gary Feinman, Stephanie Grohmann, Philip B. Holden, Árni Júlíusson, Andrey Korotayev, Axel Kristinsson, Jennifer Larson, Oren Litwin, Victor Mair, Joseph G. Manning, Patrick Manning, Arkadiusz Marciniak, Gregory Mcmahon, John Miksic, Juan Carlos Moreno Garcia, Ian Morris, Ruth Mostern, Daniel Mullins, Oluwole Oyebamiji, Peter Peregrine, Cameron Petrie, Johannes Preiser-Kapeller, Peter Rudiak-Gould, Paula Sabloff, Patrick Savage, Charles Spencer, Miriam Stark, Barend Ter Haar, Stefan Thurner, Vesna Wallace, Nina Witoszek, Liye Xie

Religious Studies Faculty Articles and Research

This article introduces the Seshat: Global History Databank, its potential, and its methodology. Seshat is a databank containing vast amounts of quantitative data buttressed by qualitative nuance for a large sample of historical and archaeological polities. The sample is global in scope and covers the period from the Neolithic Revolution to the Industrial Revolution. Seshat allows scholars to capture dynamic processes and to test theories about the co-evolution (or not) of social scale and complexity, agriculture, warfare, religion, and any number of such Big Questions. Seshat is rapidly becoming a massive resource for innovative cross-cultural and cross-disciplinary research. Seshat is …

Digital Identity: A Human-Centered Risk Awareness Study, Toufic N. Chebib

USF Tampa Graduate Theses and Dissertations

Cybersecurity threats and compromises have been at the epicenter of media attention; their risk and effect on people’s digital identity is something not to be taken lightly. Though cyber threats have affected a great number of people in all age groups, this study focuses on 55 to 75-year-olds, as this age group is close to retirement or already retired. Therefore, a notable compromise impacting their digital identity can have a major impact on their life.

To help guide this study, the following research question was formulated, “What are the risk perceptions of individuals, between the ages of 55 and 75 …

A Framework For Identifying Host-Based Artifacts In Dark Web Investigations, Arica Kulm

Masters Theses & Doctoral Dissertations

The dark web is the hidden part of the internet that is not indexed by search engines and is only accessible with a specific browser like The Onion Router (Tor). Tor was originally developed as a means of secure communications and is still used worldwide for individuals seeking privacy or those wanting to circumvent restrictive regimes. The dark web has become synonymous with nefarious and illicit content which manifests itself in underground marketplaces containing illegal goods such as drugs, stolen credit cards, stolen user credentials, child pornography, and more (Kohen, 2017). Dark web marketplaces contribute both to illegal drug usage …

Exploring And Evaluating Attributes, Values, And Structures For Entity Alignment, Zhiyuan Liu, Yixin Cao, Liangming Pan, Juanzi Li, Zhiyuan Liu, Tat-Seng Chua

Research Collection School Of Computing and Information Systems

Entity alignment (EA) aims at building a unified Knowledge Graph (KG) of rich content by linking the equivalent entities from various KGs. GNN-based EA methods present promising performance by modeling the KG structure defined by relation triples. However, attribute triples can also provide crucial alignment signal but have not been well explored yet. In this paper, we propose to utilize an attributed value encoder and partition the KG into subgraphs to model the various types of attribute triples efficiently. Besides, the performances of current EA methods are overestimated because of the name-bias of existing EA datasets. To make an objective …

A Survey Of Typical Attributed Graph Queries, Yanhao Wang, Yuchen Li, Ju Fan, Chang Ye, Mingke Chai

Research Collection School Of Computing and Information Systems

Graphs are commonly used for representing complex structures such as social relationships, biological interactions, and knowledge bases. In many scenarios, graphs not only represent topological relationships but also store the attributes that denote the semantics associated with their vertices and edges, known as attributed graphs. Attributed graphs can meet demands for a wide range of applications, and thus a variety of queries on attributed graphs have been proposed. However, these diverse types of attributed graph queries have not been systematically investigated yet. In this paper, we provide an extensive survey of several typical types of attributed graph queries. We propose …

Empowering Singapore’S Smes: Fintech P2p Lending — A Lifeline For Smes’ Survival?, Grace Lee, Alan Megargel

Research Collection School Of Computing and Information Systems

The COVID-19 pandemic has sent shock waves throughout the world, pushed countries into lockdown, and wreaked havoc on the world’s people and the global economy. The damage to economies around the world caused by the COVID-19 pandemic has far exceeded that of the global financial crisis. While all businesses suffered hugely, it would be of grave consequence if the small and medium-sized enterprises (SMEs), an important segment of every country’s economy, are unable to withstand the shock wave and sustain themselves beyond this pandemic. The COVID-19 pandemic has highlighted the importance of cash flow or working capital for the viability …

Coupled Hierarchical Transformer For Stance-Aware Rumor Verification In Social Media Conversations, Jianfei Yu, Jing Jiang, Ling Min Serena Khoo, Hai Leong Chieu, Rui Xia

Research Collection School Of Computing and Information Systems

The prevalent use of social media enables rapid spread of rumors on a massive scale, which leads to the emerging need of automatic rumor verification (RV). A number of previous studies focus on leveraging stance classification to enhance RV with multi-task learning (MTL) methods. However, most of these methods failed to employ pre-trained contextualized embeddings such as BERT, and did not exploit inter-task dependencies by using predicted stance labels to improve the RV task. Therefore, in this paper, to extend BERT to obtain thread representations, we first propose a Hierarchical Transformer1 , which divides each long thread into shorter subthreads, …

How Should We Understand The Digital Economy In Asia? Critical Assessment And Research Agenda, Kai Li, Dan J. Kim, Karl R. Lang, Robert J. Kauffman, Maurizio Naldi

Research Collection School Of Computing and Information Systems

By Asian digital economy, we refer to high-tech developments, business and social transformations, and information-driven changes in the region's growth. We discuss its background and foundations, significance in Asia and contribution to removal of historical barriers in traditional business. We assess how new value chains are transforming country-level involvement in worldwide manufacturing and note "smiling curve theory" predictions about the global value chain in Asia for high-tech firms and their economies. The takeaway is that the digital economy in Asian nations involves revamping business processes through technology innovation, government policies for growth, and digital entrepreneurship. We analyze the "digital economy …

Cross-Thought For Sentence Encoder Pre-Training, Shuohang Wang, Yuwei Fang, Siqi Sun, Zhe Gan, Yu Cheng, Jingjing Liu, Jing Jiang

Research Collection School Of Computing and Information Systems

In this paper, we propose Cross-Thought, a novel approach to pre-training sequence encoder, which is instrumental in building reusable sequence embeddings for large-scale NLP tasks such as question answering. Instead of using the original signals of full sentences, we train a Transformer-based sequence encoder over a large set of short sequences, which allows the model to automatically select the most useful information for predicting masked words. Experiments on question answering and textual entailment tasks demonstrate that our pre-trained encoder can outperform state-of-the-art encoders trained with continuous sentence signals as well as traditional masked language modeling baselines. Our proposed approach also …

Leveraging Profanity For Insincere Content Detection: A Neural Network Approach, Swapna Gottipati, Annabel Tan, David Jing Shan Chow, Joel Wee Kiat Lim

Research Collection School Of Computing and Information Systems

Community driven social media sites are rich sources of knowledge and entertainment and at the same vulnerable to the flames or toxic content that can be dangerous to various users of these platforms as well as to the society. Therefore, it is crucial to identify and remove such content to have a better and safe online experience. Manually eliminating flames is tedious and hence many research works focus on machine learning or deep learning models for automated methods. In this paper, we primarily focus on detecting the insincere content using neural network-based learning methods. We also integrated the profanity features …

Learning Personal Conscientiousness From Footprints In E-Learning Systems, Lo Pang-Yun Ting, Shan Yun Teng, Kun Ta Chuang, Ee-Peng Lim

Research Collection School Of Computing and Information Systems

Personality inference has received widespread attention for its potential to infer psychological well being, job satisfaction, romantic relationship success, and professional performance. In this research, we focus on Conscientiousness, one of the well studied Big Five personality traits, which determines if a person is self-disciplined, organized, and hard-working. Research has shown that Conscientiousness is related to a person's academic and workplace success. For an expert to evaluate a person's Conscientiousness, long-term observation of the person's behavior at work place or at home is usually required. To reduce this evaluation effort as well as to cope with the increasing trend of …

Cost-Sensitive Deep Forest For Price Prediction, Chao Ma, Zhenbing Liu, Zhiguang Cao, Wen Song, Jie Zhang, Weiliang Zeng

Research Collection School Of Computing and Information Systems

For many real-world applications, predicting a price range is more practical and desirable than predicting a concrete value. In this case, price prediction can be regarded as a classification problem. Although deep forest is recognized as the best solution to many classification problems, a crucial issue limits its direct application to price prediction, i.e., it treated all the misclassifications equally no matter how far away they are from the real classes, since their impacts on the accuracy are the same. This is unreasonable to price prediction as the misclassification should be as close to the real price range as possible …

Exploring The Impact Of Covid-19 On Aviation Industry: A Text Mining Approach, Gottipati Swapna, Kyong Jin Shim, Weiling Angeline Jiang, Sheng Wei Andre Justin Lee

Research Collection School Of Computing and Information Systems

Our study presents a comprehensive analysis of news articles from FlightGlobal website during the first half of 2020. Our analyses reveal useful insights on themes and trends concerning the aviation industry during the COVID-19 period. We applied text mining and NLP techniques to analyse the articles for extracting the aviation themes and article sentiments (positive and negative). Our results show that there is a variation in the sentiment trends for themes aligned with the real-world developments of the pandemic. The article sentiment analysis can offer industry players a quick sense of the nature of developments in the industry. Our article …

Capitalising Product Associations In A Supermarket Retail Environment, Michelle L. F. Cheong, Yong Qing Chia

Research Collection School Of Computing and Information Systems

This paper explores methods to capitalise on retail companies’ transactional databases, to mine meaningful product associations, and to design product placement strategies as a means to drive sales. We implemented three in-store initiatives based on our hypotheses – placing products with high associations together will induce an increase in sales of consequent; introducing an antecedent that is new to store will bring about a similar impact on sales of consequent based on established product association rules uncovered from other stores. Sales tracking over twelve weeks revealed that there were improvements in sales of consequents across all three initiatives performed in-store.

The Role Of Information And Knowledge Of Weather Warnings In Marine Access Behavior : A Field Experiment In Coastal Area Of Bangladesh, Khan Mehedi Hasan

Lingnan Theses and Dissertations (MPhil & PhD)

The world’s largest mangrove forest named Sundarban is located in the Bay of Bengal. Due to richness of aqua and forest resources, the coastal community of Khulna district of Bangladesh immensely depends on the forest for income and livelihoods, all the year round. For extracting resources, thousands of people enter into the forest by crossing river, generally with small boats. The region faces various natural disasters repeatedly. Each year about 13-14 cyclones are formed in the Bay of Bengal, which are threats for coastal households. Those hazards bear more risk for marine entrants. Analyzing coastal households’ marine access for two …