Databases and Information Systems Commons^™

Salient Object Detection With Pyramid Attention And Salient Edges, Wenguan Wang, Shuyang Zhao, Jianbing Shen, Steven C. H. Hoi, Ali Borji

Research Collection Yong Pung How School Of Law

This paper presents a new method for detecting salient objects in images using convolutional neural networks (CNNs). The proposed network, named PAGE-Net, offers two key contributions. The first is the exploitation of an essential pyramid attention structure for salient object detection. This enables the network to concentrate more on salient regions while considering multi-scale saliency information. Such a stacked attention design provides a powerful tool to efficiently improve the representation ability of the corresponding network layer with an enlarged receptive field. The second contribution lies in the emphasis on the importance of salient edges. Salient edge information offers a strong …

Time-Based Payout Ratio For Coordinating Supply And Demand On An On-Demand Service Platform, Jiaru Bai, Kut C. So, Christopher S. Tang, Xiqun Chen, Hai Wang

Research Collection School Of Computing and Information Systems

Many on-demand service platforms use a fixed payout ratio (i.e., the percentage of the platform’s revenue that is paid to the providers) regardless of the customer demand and the number of participating providers that tend to vary over time. In this chapter, we examine the implications of time-based payout ratios. To do so, we first present a queueing model with endogenous supply (number of participating providers) and endogenous demand (customer request rate) to model this on-demand service platform. In our model, earnings-sensitive independent providers have heterogeneous reservation price (for work participation) to serve wait-time and price-sensitive customers with heterogeneous valuation …

Matching Passengers And Drivers With Multiple Objectives In Ride Sharing Markets, Guodong Lyu, Chung Piaw Teo, Wangchi Cheung, Hai Wang

Research Collection School Of Computing and Information Systems

In many cities in the world, ride sharing companies, such as Uber, Didi, Grab and Lyft, have been able to leverage on Internet-based platforms to conduct online decision making to connect passengers and drivers. These online platforms facilitate the integration of passengers and drivers’ mobility data on smart phones in real-time, which enables a convenient matching between demand and supply in real time. These clear operational advantages have motivated many similar shared service business models in the public transportation arena, and have been a disruptive force to the traditional taxi industry.

Online Multimodal Co-Indexing And Retrieval Of Social Media Data, Lei Meng, Ah-Hwee Tan, Donald C. Wunsch

Research Collection School Of Computing and Information Systems

Effective indexing of social media data is key to searching for information on the social Web. However, the characteristics of social media data make it a challenging task. The large-scale and streaming nature is the first challenge, which requires the indexing algorithm to be able to efficiently update the indexing structure when receiving data streams. The second challenge is utilizing the rich meta-information of social media data for a better evaluation of the similarity between data objects and for a more semantically meaningful indexing of the data, which may allow the users to search for them using the different types …

Multimodal Review Generation For Recommender Systems, Quoc Tuan Truong, Hady W. Lauw

Research Collection School Of Computing and Information Systems

Key to recommender systems is learning user preferences, which are expressed through various modalities. In online reviews, for instance, this manifests in numerical rating, textual content, as well as visual images. In this work, we hypothesize that modelling these modalities jointly would result in a more holistic representation of a review towards more accurate recommendations. Therefore, we propose Multimodal Review Generation (MRG), a neural approach that simultaneously models a rating prediction component and a review text generation component. We hypothesize that the shared user and item representations would augment the rating prediction with richer information from review text, while sensitizing …

Learning Two-Layer Neural Networks With Symmetric Inputs, Rong Ge, Rohith Kuditipudi, Zhize Li, Xiang Wang

Research Collection School Of Computing and Information Systems

We give a new algorithm for learning a two-layer neural network under a very general class of input distributions. Assuming there is a ground-truth two-layer network $y = A \sigma(Wx) + \xi$, where A, W are weight matrices, $\xi$ represents noise, and the number of neurons in the hidden layer is no larger than the input or output, our algorithm is guaranteed to recover the parameters A, W of the ground-truth network. The only requirement on the input x is that it is symmetric, which still allows highly complicated and structured input. Our algorithm is based on the method-of-moments framework …

Concluding Remarks, Lei Meng, Ah-Hwee Tan, Donald C. Wunsch

Research Collection School Of Computing and Information Systems

This chapter summarizes the major contributions in this book and discusses their possible positions and requirements in some future scenarios. Section 8.1 follows the book structure to revisit the key contributions of this book in both theories and applications. The developed algorithms, such as the VA-ARTs for hyperparameter adaptation and the GHF-ART for multimedia representation and fusion, and the four applications, such as clustering and retrieving socially enriched multimedia data, are concentrated using one paragraph and three paragraphs, respectively. In Sect. 8.2, the roles of the proposed ART-embodied algorithms in social media clustering tasks are highlighted, and their possible evolutions …

Clustering And Its Extensions In The Social Media Domain, Lei Meng, Ah-Hwee Tan, Donald C. Wunsch

Research Collection School Of Computing and Information Systems

This chapter summarizes existing clustering and related approaches for the identified challenges as described in Sect. 1.2 and presents the key branches of social media mining applications where clustering holds a potential. Specifically, several important types of clustering algorithms are first illustrated, including clustering, semi-supervised clustering, heterogeneous data co-clustering, and online clustering. Subsequently, Sect. 2.5 presents a review on existing techniques that help decide the value of the predefined number of clusters (required by most clustering algorithms) automatically and highlights the clustering algorithms that do not require such a parameter. It better illustrates the challenge of input parameter sensitivity of …

Robust Factorization Machine: A Doubly Capped Norms Minimization, Chenghao Liu, Teng Zhang, Jundong Li, Jianwen Yin, Peilin Zhao, Jianling Sun, Steven C. H. Hoi

Research Collection School Of Computing and Information Systems

Factorization Machine (FM) is a general supervised learning framework for many AI applications due to its powerful capability of feature engineering. Despite being extensively studied, existing FM methods have several limitations in common. First of all, most existing FM methods often adopt the squared loss in the modeling process, which can be very sensitive when the data for learning contains noises and outliers. Second, some recent FM variants often explore the low-rank structure of the feature interactions matrix by relaxing the low-rank minimization problem as a trace norm minimization, which cannot always achieve a tight approximation to the original one. …

How To Derive Causal Insights For Digital Commerce In China? A Research Commentary On Computational Social Science Methods, David C.W. Phang, Kanliang Wang, Qiu-Hong Wang, Robert John Kauffman, Maurizio Naldi

Research Collection School Of Computing and Information Systems

The transformation of empirical research due to the arrival of big data analytics and data science, as well as the new availability of methods that emphasize causal inference, are moving forward at full speed. In this Research Commentary, we examine the extent to which this has the potential to influence how e-commerce research is conducted. China offers the ultimate in data-at-scale settings, and the construction of real-world natural experiments. Chinese e-commerce includes some of the largest firms involved in e-commerce, mobile commerce, social media and social networks. This article was written to encourage young faculty and doctoral students to engage …

Unifying Knowledge Graph Learning And Recommendation: Towards A Better Understanding Of User Preferences, Yixin Cao, Xiang Wang, Xiangnan He, Zikun Hu, Tat-Seng Chua

Research Collection School Of Computing and Information Systems

Incorporating knowledge graph (KG) into recommender system is promising in improving the recommendation accuracy and explainability. However, existing methods largely assume that a KG is complete and simply transfer the ”knowledge” in KG at the shallow level of entity raw data or embeddings. This may lead to suboptimal performance, since a practical KG can hardly be complete, and it is common that a KG has missing facts, relations, and entities. Thus, we argue that it is crucial to consider the incomplete nature of KG when incorporating it into recommender system. In this paper, we jointly learn the model of recommendation …

Neural Multimodal Belief Tracker With Adaptive Attention For Dialogue Systems, Zheng Zhang, Lizi Liao, Minlie Huang, Xiaoyan Zhu, Tat-Seng Chua

Research Collection School Of Computing and Information Systems

Multimodal dialogue systems are attracting increasing attention with a more natural and informative way for human-computer interaction. As one of its core components, the belief tracker estimates the user's goal at each step of the dialogue and provides a direct way to validate the ability of dialogue understanding. However, existing studies on belief trackers are largely limited to textual modality, which cannot be easily extended to capture the rich semantics in multimodal systems such as those with product images. For example, in fashion domain, the visual appearance of clothes play a crucial role in understanding the user's intention. In this …

Project Sidewalk: A Web-Based Crowdsourcing Tool For Collecting Sidewalk Accessibility Data At Scale, Manaswi Saha, Michael Saugstad, Hanuma Maddali, Aileen Zeng, Ryan Holland, Steven Bower, Aditya Dash, Sage Chen, Anthony Li, Kotaro Hara, Jon Froehlich

Research Collection School Of Computing and Information Systems

We introduce Project Sidewalk, a new web-based tool that enables online crowdworkers to remotely label pedestrian-related accessibility problems by virtually walking through city streets in Google Street View. To train, engage, and sustain users, we apply basic game design principles such as interactive onboarding, mission-based tasks, and progress dashboards. In an 18-month deployment study, 797 online users contributed 205,385 labels and audited 2,941 miles of Washington DC streets. We compare behavioral and labeling quality differences between paid crowdworkers and volunteers, investigate the effects of label type, label severity, and majority vote on accuracy, and analyze common labeling errors. To complement …

Detect Rumors On Twitter By Promoting Information Campaigns With Generative Adversarial Learning, Jing Ma, Wei Gao, Kam-Fai Wong

Research Collection School Of Computing and Information Systems

Rumors can cause devastating consequences to individual and/or society. Analysis shows that widespread of rumors typically results from deliberately promoted information campaigns which aim to shape collective opinions on the concerned news events. In this paper, we attempt to fight such chaos with itself to make automatic rumor detection more robust and effective. Our idea is inspired by adversarial learning method originated from Generative Adversarial Networks (GAN). We propose a GAN-style approach, where a generator is designed to produce uncertain or conflicting voices, complicating the original conversational threads in order to pressurize the discriminator to learn stronger rumor indicative representations …

Socially-Enriched Multimedia Data Co-Clustering, Ah-Hwee Tan

Research Collection School Of Computing and Information Systems

Heterogeneous data co-clustering is a commonly used technique for tapping the rich meta-information of multimedia web documents, including category, annotation, and description, for associative discovery. However, most co-clustering methods proposed for heterogeneous data do not consider the representation problem of short and noisy text and their performance is limited by the empirical weighting of the multimodal features. This chapter explains how to use the Generalized Heterogeneous Fusion Adaptive Resonance Theory (GHF-ART) generalized heterogeneous fusion adaptive resonance theory for clustering large-scale web multimedia documents. Specifically, GHF-ART is designed to handle multimedia data with an arbitrarily rich level of meta-information. For handling …

Community Discovery In Heterogeneous Social Networks, Lei Meng, Ah-Hwee Tan, Donald C. Wunsch

Research Collection School Of Computing and Information Systems

Discovering social communities of web users through clustering analysis of heterogeneous link associations has drawn much attention. However, existing approaches typically require the number of clusters a priori, do not address the weighting problem for fusing heterogeneous types of links, and have a heavy computational cost. This chapter studies the commonly used social links of users and explores the feasibility of the proposed heterogeneous data co-clustering algorithm GHF-ART, as introduced in Sect. 3.6, for discovering user communities in social networks. Contrary to the existing algorithms proposed for this task, GHF-ART performs real-time matching of patterns and one-pass learning, which guarantees …

Beyond Autonomy: The Self And Life Of Social Agents, Budhitama Subagdja, Ah-Hwee Tan

Research Collection School Of Computing and Information Systems

Agents have gained popularity nowadays as virtual assistants and companions of their human users supporting daily activities in many aspects of personal life. Designed to be sociable, an agent engages its user(s) to communicate and even develop friendships. Rather than just as a lifeless toy, it is supposed to be perceived as an individual with its own personality, experiences, and social life. In this paper, we seek to highlight self-hood as another dimension that characterizes an agent. Besides levels of autonomy and reasoning, an agent can be defined based on its capacity to process and reflect on its own self …

Cure: Flexible Categorical Data Representation By Hierarchical Coupling Learning, Songlei Jian, Guansong Pang, Longbing Cao, Kai Lu, Hang Gao

Research Collection School Of Computing and Information Systems

The representation of categorical data with hierarchical value coupling relationships (i.e., various value-to-value cluster interactions) is very critical yet challenging for capturing complex data characteristics in learning tasks. This paper proposes a novel and flexible coupled unsupervised categorical data representation (CURE) framework, which not only captures the hierarchical couplings but is also flexible enough to be instantiated for contrastive learning tasks. CURE first learns the value clusters of different granularities based on multiple value coupling functions and then learns the value representation from the couplings between the obtained value clusters. With two complementary value coupling functions, CURE is instantiated into …

Neural Multimodal Belief Tracker With Adaptive Attention For Dialogue Systems, Zheng Zhang, Lizi Liao, Minlie Huang, Xiaoyan Zhu, Tat-Seng Chua

Research Collection School Of Computing and Information Systems

Multimodal dialogue systems are attracting increasing attention with a more natural and informative way for human-computer interaction. As one of its core components, the belief tracker estimates the user's goal at each step of the dialogue and provides a direct way to validate the ability of dialogue understanding. However, existing studies on belief trackers are largely limited to textual modality, which cannot be easily extended to capture the rich semantics in multimodal systems such as those with product images. For example, in fashion domain, the visual appearance of clothes play a crucial role in understanding the user's intention. In this …

Personalized Web Image Organization, Lei Meng, Ah-Hwee Tan, Donald C. Wunsch

Research Collection School Of Computing and Information Systems

Due to the problem of semantic gap, i.e. the visual content of an image may not represent its semantics well, existing efforts on web image organization usually transform this task to clustering the surrounding text. However, because the surrounding text is usually short and the words therein usually appear only once, existing text clustering algorithms can hardly use the statistical information for image representation and may achieve downgraded performance with higher computational cost caused by learning from noisy tags. This chapter presents using the Probabilistic ART with user preference architecture, as introduced in Sects. 3.5 and 3.4, for personalized web …

Adaptive Resonance Theory (Art) For Social Media Analytics, Lei Meng, Ah-Hwee Tan, Donald C. Ii Wunsch

Research Collection School Of Computing and Information Systems

The last decade has witnessed how social media in the era of Web 2.0 reshapes the way people communicate, interact, and entertain in daily life and incubates the prosperity of various user-centric platforms, such as social networking, question answering, massive open online courses (MOOC), and e-commerce platforms. The available rich user-generated multimedia data on the web has evolved traditional ways of understanding multimedia research and has led to numerous emerging topics on human-centric analytics and services, such as user profiling, social network mining, crowd behavior analysis, and personalized recommendation. Clustering, as an important tool for mining information groups and in-group …

Peerlens: Peer-Inspired Interactive Learning Path Planning In Online Question Pool, Meng Xia, Mingfei Sun, Huan Wei, Qing Chen, Yong Wang, Lei Shi, Huamin Qu, Xiaojuan Ma

Research Collection School Of Computing and Information Systems

Online question pools like LeetCode provide hands-on exercises of skills and knowledge. However, due to the large volume of questions and the intent of hiding the tested knowledge behind them, many users find it hard to decide where to start or how to proceed based on their goals and performance. To overcome these limitations, we present PeerLens, an interactive visual analysis system that enables peer-inspired learning path planning. PeerLens can recommend a customized, adaptable sequence of practice questions to individual learners, based on the exercise history of other users in a similar learning scenario. We propose a new way to …

Big Data And The Consumer, Seema Chokshi

MITB Thought Leadership Series

What is big data? The intuitive meaning of the phrase ‘big data’ might be “data that is huge in quantity”. But is that interpretation enough? Data of this type has existed for as long as humans have made records of their work. Some of the earliest writings, such as cuneiform, contain vast amounts of data covering areas as diverse as law, mapping and mathematical equations.

Question Answering With Textual Sequence Matching, Shuohang Wang

Dissertations and Theses Collection (Open Access)

Question answering (QA) is one of the most important applications in natural language processing. With the explosive text data from the Internet, intelligently getting answers of questions will help humans more efficiently collect useful information. My research in this thesis mainly focuses on solving question answering problem with textual sequence matching model which is to build vectorized representations for pairs of text sequences to enable better reasoning. And our thesis consists of three major parts.

In Part I, we propose two general models for building vectorized representations over a pair of sentences, which can be directly used to solve the …

Modeling Sequential And Basket-Oriented Associations For Top-K Recommendation, Duc-Trong Le Duc Trong

Dissertations and Theses Collection (Open Access)

Top-K recommendation is a typical task in Recommender Systems. In traditional approaches, it mainly relies on the modeling of user-item associations, which emphasizes the user-specific factor or personalization. Here, we investigate another direction that models item-item associations, especially with the notions of sequence-aware and basket-level adoptions . Sequences are created by sorting item adoptions chronologically. The associations between items along sequences, referred to as “sequential associations”, indicate the influence of the preceding adoptions on the following adoptions. Considering a basket of items consumed at the same time step (e.g., a session, a day), “basket-oriented associations” imply correlative dependencies among these …

Automatic Short Answer Grading Using Siamese Bidirectional Lstm Based Regression, Arya Prabhudesai, Nguyen Binh Duong Ta

Research Collection School Of Computing and Information Systems

Automatic student assessment plays an important role in education - it provides instant feedback to learners, and at the same time reduces tedious grading workload for instructors. In this paper, we investigate new machine learning techniques for automatic short answer grading (ASAG). The ASAG problem mainly involves assessing short, natural language responses to given questions automatically. While current research in the field has focused either on feature engineering or deep learning, we propose a new approach which combines the advantages of both. More specifically, we propose a Siamese Bidirectional LSTM Neural Network based Regressor in conjunction with handcrafted features for …

Efficient Algorithms For Solving Aggregate Keyword Routing Problems, Qize Jiang, Weiwei Sun, Baihua Zheng, Kunjie Chen

Research Collection School Of Computing and Information Systems

With the emergence of smart phones and the popularity of GPS, the number of point of interest (POIs) is growing rapidly and spatial keyword search based on POIs has attracted significant attention. In this paper, we study a more sophistic type of spatial keyword searches that considers multiple query points and multiple query keywords, namely Aggregate Keyword Routing (AKR). AKR looks for an aggregate point m together with routes from each query point to m. The aggregate point has to satisfy the aggregate keywords, the routes from query points to the aggregate point have to pass POIs in order to …

Online Collaborative Filtering With Implicit Feedback, Jianwen Yin, Chenghao Liu, Jundong Li, Bing Tian Dai, Yun-Chen Chen, Min Wu, Jianling Sun

Research Collection School Of Computing and Information Systems

Studying recommender systems with implicit feedback has become increasingly important. However, most existing works are designed in an offline setting while online recommendation is quite challenging due to the one-class nature of implicit feedback. In this paper, we propose an online collaborative filtering method for implicit feedback. We highlight three critical issues of existing works. First, when positive feedback arrives sequentially, if we treat all the other missing items for this given user as the negative samples, the mis-classified items will incur a large deviation since some items might appear as the positive feedback in the subsequent rounds. Second, the …

Maximizing Multifaceted Network Influence, Yuchen Li, Ju Fan, George V. Ovchinnikov, Panagiotis Karras

Research Collection School Of Computing and Information Systems

An information dissemination campaign is often multifaceted, involving several facets or pieces of information disseminating from different sources. The question then arises, how should we assign such pieces to eligible sources so as to achieve the best viral dissemination results? Past research has studied the problem of Influence Maximization (IM), which is to select a set of k promoters that maximizes the expected reach of a message over a network. However, in this classical IM problem, each promoter spreads out the same unitary piece of information. In this paper, we propose the Optimal Influential Pieces Assignment (OIPA) problem, which is …

Fine-Grained Geolocation Of Tweets In Temporal Proximity, Wen Haw Chong, Ee Peng Lim

Research Collection School Of Computing and Information Systems

In fine-grained tweet geolocation, tweets are linked to the specific venues (e.g., restaurants, shops) fromwhich they were posted. This explicitly recovers the venue context that is essential for applications such aslocation-based advertising or user profiling. For this geolocation task, we focus on geolocating tweets that arecontained in tweet sequences. In a tweet sequence, tweets are posted from some latent venue(s) by the sameuser and within a short time interval. This scenario arises from two observations: (1) It is quite common thatusers post multiple tweets in a short time and (2) most tweets are not geocoded. To more accurately geolocatea tweet, …