Databases and Information Systems Commons^™

Learning Natural Language Inference With Lstm, Shuohang Wang, Jing Jiang

Research Collection School Of Computing and Information Systems

Natural language inference (NLI) is a fundamentally important task in natural language processing that has many applications. The recently released Stanford Natural Language Inference (SNLI) corpus has made it possible to develop and evaluate learning-centered methods such as deep neural networks for natural language inference (NLI). In this paper, we propose a special long short-term memory (LSTM) architecture for NLI. Our model builds on top of a recently proposed neural attention model for NLI but is based on a significantly different idea. Instead of deriving sentence embeddings for the premise and the hypothesis to be used for classification, our solution …

Smart Living For Elderly: Design And Human-Computer Interaction Considerations, Ranjana Sharma, Fiona Fui-Hoon Nah, Kavya Sharma, Teja S. Katta, Natalie Pang, Alvin Yong

Research Collection School Of Computing and Information Systems

To address aging challenges, we examine the concept of smart living and its applications for the elderly. Smart living refers to improving quality of life by transforming environments to become more intelligent and adaptable to users. In this paper, we discuss how smart living applications can help to address the needs of the elderly, as well as the design and human-computer interaction considerations for such applications.

Poster: Improving Communication And Communicability With Smarter Use Of Text-Based Messages On Mobile And Wearable Devices, Kenny T. W. Choo

Research Collection School Of Computing and Information Systems

While smartphones have undoubtedly afforded many modern conveniences such as emails, instant messaging or web search, the notifications from smartphones conversely impact our lives through a deluge of information, or stress arising from expectations that we should turn our immediate attention to them (e.g., work emails). In my latest research, we find that the glanceability of smartwatches may provide an opportunity to reduce the perceived disruption from mobile notifications. Text is a common medium for communication in smart devices, the application of natural language processing on text, together with the physical affordances of smartwatches, present exciting opportunities for research to …

Collective Rumor Correction On The Death Hoax Of A Political Figure In Social Media, Alton Y. K. Chua, Sin-Mei Cheah, Dion Hoe-Lian Goh, Ee-Peng Lim

Research Collection School Of Computing and Information Systems

Conversations on social media networks that discuss a crisis incident as it unfolds have become a norm in recent years. Left to its own devices, such conversations could quickly degenerate into rumor mills. Little research has thus far examined the correction of rumors on social media. Using the third person effect as a theoretical underpinning, we developed a model of collective rumor correction on social media based on an incident surrounding the death hoax of a political figure. Tweets from Twitter were collected and analyzed for the period when a spike of circulating rumors speculating the demise of Singapore's first …

Qcri At Semeval-2016 Task 4: Probabilistic Methods For Binary And Ordinal Quantification, Giovanni Da San Martino, Wei Gao, Fabrizio Sebastiani

Research Collection School Of Computing and Information Systems

. (2016). n. In , pages 58—63, San Diego, California, USA. Association for Computational Linguistics. (1st place in sub-task E of Sentiment Analysis in Twitter)

Geometric Aspects And Auxiliary Features To Top-K Processing [Advanced Seminar], Kyriakos Mouratidis

Research Collection School Of Computing and Information Systems

Top-k processing is a well-studied problem with numerous applications that is becoming increasingly relevant with the growing availability of recommendation systems and decision making software on PCs, PDAs and smart-phones. The objective of this seminar is twofold. First, we will delve into the geometric aspects of top-k processing. Second, we will cover complementary features to top-k queries that have a strong geometric nature. The seminar will close with insights in the effect of dimensionality on the meaningfulness of top-k queries, and interesting similarities to nearest neighbor search.

An Experimental Investigation Of Product Competition And Marketing In Social Networks, Cen Chen, Zhiling Guo, Shih-Fen Cheng, Hoong Chuin Lau

Research Collection School Of Computing and Information Systems

We conduct computational experiment using Facebook data to evaluate competing firms’ initial market seeding and subsequent targeted marketing strategies that influence consumers’ new product adoption decisions. We find that firms generally overspend their advertising budget in the market seeding phase. In the subsequent market advertising phase, a coupon strategy (equivalent to price discount) generally yields higher market share than the strategy of distributing free product samples. The effect is more significant when both price and product quality are low. We offer managerial insights into firms’ effective competition strategies for new product introduction in the presence of consumers’ word of mouth …

An Economic Analysis Of Consumer Learning For Online Entertainment Shopping, Jin Li, Zhiling Guo, Geoffrey K.F. Tso

Research Collection School Of Computing and Information Systems

Entertainment shopping supported by pay-to-bid auction is an emerging online business model in recent years. Consumers expect both entertainment value and monetary return from their participation in entertainment shopping. We propose a dynamic structural model to study consumers’ online shopping behavior. We analyze the learning process of consumers from two perspectives based on the Bayesian updating framework: (1) consumers update their beliefs about the entertainment value through their repeated personal participation experiences, and (2) consumers infer the expected monetary payoffs on the website by observing the publically available auction ending price information. We estimate the model using a large dataset …

Efficient Multi-Class Selective Sampling On Graphs, Peng Yang, Peilin Zhao, Zhen Hai, Wei Liu, Hoi, Steven C. H., Xiao-Li Li

Research Collection School Of Computing and Information Systems

A graph-based multi-class classification problem is typically converted into a collection of binary classification tasks via the one-vs.-all strategy, and then tackled by applying proper binary classification algorithms. Unlike the one-vs.-all strategy, we suggest a unified framework which operates directly on the multi-class problem without reducing it to a collection of binary tasks. Moreover, this framework makes active learning practically feasible for multi-class problems, while the one-vs.-all strategy cannot. Specifically, we employ a novel randomized query technique to prioritize the informative instances. This query technique based on the hybrid criterion of "margin" and "uncertainty" can achieve a comparable mistake bound …

Protecting The Nectar Of The Ganga River Through Game-Theoretic Factory Inspections, Benjamin Ford, Matthew Brown, Amulya Yadav, Amandeep Singh, Arunesh Sinha, Biplav Srivastava, Christopher Kiekintveld, Tambe Millind

Research Collection School Of Computing and Information Systems

Leather is an integral part of the world economy and a substantial income source for developing countries. Despite government regulations on leather tannery waste emissions, inspection agencies lack adequate enforcement resources, and tanneries’ toxic wastewaters wreak havoc on surrounding ecosystems and communities. Previous works in this domain stop short of generating executable solutions for inspection agencies. We introduce NECTAR - the first security game application to generate environmental compliance inspection schedules. NECTAR’s game model addresses many important real-world constraints: a lack of defender resources is alleviated via a secondary inspection type; imperfect inspections are modeled via a heterogeneous failure rate; …

Video Modeling And Learning On Riemannian Manifold For Emotion Recognition In The Wild, Mengyi Liu, Ruiping Wang, Shaoxin Li, Zhiwu Huang, Shiguang Shan, Xilin Chen

Research Collection School Of Computing and Information Systems

In this paper, we present the method for our submission to the emotion recognition in the wild challenge (EmotiW). The challenge is to automatically classify the emotions acted by human subjects in video clips under real-world environment. In our method, each video clip can be represented by three types of image set models (i.e. linear subspace, covariance matrix, and Gaussian distribution) respectively, which can all be viewed as points residing on some Riemannian manifolds. Then different Riemannian kernels are employed on these set models correspondingly for similarity/ distance measurement. For classification, three types of classifiers, i.e. kernel SVM, logistic regression, …

Patient Engagement In The Medical Facility Waiting Room Using Gamified Healthcare Information Delivery, Raheel Hassan, Nathan W. Twyman, Fiona Fui-Hoon Nah, Keng Siau

Research Collection School Of Computing and Information Systems

This study explores the proposition that medical facility waiting rooms are an opportune setting to engage with and educate patients while they are waiting for care. In collaboration with emergency department (ED) personnel, we developed ER Hero, a tablet-based application for waiting rooms that introduces patients to ED professionals and operations through mini-games and story-like interaction. We evaluated this prototype with human participants to determine how well it performed when compared to paper-based information disclosure presenting the same information. Participants using the application exhibited increased ED knowledge, decreased nervousness, and increased interest. The gamified application outperformed a paper-based approach on …

Skewer: Sentiment Knowledge Extraction With Entity Recognition, Christopher James Wu

Master's Theses

The California state legislature introduces approximately 5,000 new bills each legislative session. While the legislative hearings are recorded on video, the recordings are not easily accessible to the public. The lack of official transcripts or summaries also increases the effort required to gain meaningful insight from those recordings. Therefore, the news media and the general population are largely oblivious to what transpires during legislative sessions.

Digital Democracy, a project started by the Cal Poly Institute for Advanced Technology and Public Policy, is an online platform created to bring transparency to the California legislature. It features a searchable database of state …

Collaborative Development Of A Small Business Emergency Planning Model, Arthur Henry Hendela

Dissertations

Small businesses, which are defined by the US Small Business Administration as entities with less than 500 employees, suffer interruptions from diverse risks such as financial events, legal situations, or severe storms exemplified by Hurricane Sandy. Proper preparations can help lessen the length of the interruption and put employees and owners back to work. Large corporations generally have large budgets available for planning, business continuity, and disaster recovery. Small businesses must decide which risks are the most important and how best to mitigate those risks using minimal resources.

This research uses a series of surveys followed by mathematical modeling to …

Mediating Chance Encounters Through Opportunistic Social Matching, Julia M. Mayer

Dissertations

Chance encounters, the unintended meeting between people unfamiliar with each other, serve as an important social lubricant helping people to create new social ties, such as making new friends or finding an activity, study or collaboration partner. Unfortunately, social barriers often prevent chance encounters in environments where people do not know each other and people have to rely on serendipity to meet or be introduced to interesting people around them. Little is known about the underlying dynamics of chance encounters and how systems could utilize contextual data to mediate chance encounters. This dissertation addresses this gap in research literature by …

Hybrid Similarity Function For Big Data Entity Matching With R-Swoosh, Vimal Chandra Gorijala

Master's Projects

Entity Matching (EM) is the problem of determining if two entities in a data set refer to the same real-world object. For example, it decides if two given mentions in the data, such as “Helen Hunt” and “H. M. Hunt”, refer to the same real-world entity by using different similarity functions. This problem plays a key role in information integration, natural language understanding, information processing on the World-Wide Web, and on the emerging Semantic Web. This project deals with the similarity functions and thresholds utilized in them to determine the similarity of the entities. The work contains two major parts: …

Efficient Pair-Wise Similarity Computation Using Apache Spark, Parineetha Gandhi Tirumali

Master's Projects

Entity matching is the process of identifying different manifestations of the same real world entity. These entities can be referred to as objects(string) or data instances. These entities are in turn split over several databases or clusters based on the signatures of the entities. When entity matching algorithms are performed on these databases or clusters, there is a high possibility that a particular entity pair is compared more than once. The number of comparison for any two entities depend on the number of common signatures or keys they possess. This effects the performance of any entity matching algorithm. This paper …

Library Writers Reward Project, Saravana Kumar Gajendran

Master's Projects

Open-source library development exploits the distributed intelligence of participants in Internet communities. Nowadays, contribution to the open-source community is fading [16] (Stackalytics, 2016) as there is not much recognition for library writers. They can start exploring ways to generate revenue as they actively contribute to the open-source community.

This project helps library writers to generate revenue in the form of bitcoins for their contribution. Our solution to generate revenue for library writers is to integrate bitcoin mining with existing JavaScript libraries, such as jQuery. More use of the library leads to more revenue for the library writers. It uses the …

Processing Posting Lists Using Opencl, Radha Kotipalli

Master's Projects

One of the main requirements of internet search engines is the ability to retrieve relevant results with faster response times. Yioop is an open source search engine designed and developed in PHP by Dr. Chris Pollett. The goal of this project is to explore the possibilities of enhancing the performance of Yioop by substituting resource-intensive existing PHP functions with C based native PHP extensions and the parallel data processing technology OpenCL. OpenCL leverages the Graphical Processing Unit (GPU) of a computer system for performance improvements.

Some of the critical functions in search engines are resource-intensive in terms of processing power, …

The Mexican Water Forest: Benefits Of Using Remote Sensing Techniques To Assess Changes In Land Use And Land Cover, Maria F. Lopez Ornelas

Master's Projects and Capstones

In the past 30 years, anthropogenic activities like urbanization, agriculture, road fragmentation and deforestation have resulted in changes in the land use and land cover (LULC) in the Mexican Water Forest. Due to the important ecosystem services, and the natural resources this forest provides, in Mexico, it has become increasingly necessary to use new technologies and tools to support the planning, implementation and integration of forest management and conservation plans, as well as ecological and socioeconomic analysis of this ecosystem. Remote Sensing techniques and Geographic Information Systems (GIS) have been a true technological and methodological revolution in the acquisition, management …