Numerical Analysis and Scientific Computing Commons^™

Ar-Miner: Mining Informative Reviews For Developers From Mobile App Marketplace, Ning Chen, Jialiu Lin, Steven C. H. Hoi, Xiaokui Xiao, Boshen Zhang

Research Collection School Of Computing and Information Systems

With the popularity of smartphones and mobile devices, mobile application (a.k.a. “app”) markets have been growing exponentially in terms of number of users and downloads. App developers spend considerable effort on collecting and exploiting user feedback to improve user satisfaction, but suffer from the absence of effective user review analytics tools. To facilitate mobile app developers discover the most “informative” user reviews from a large and rapidly increasing pool of user reviews, we present “AR-Miner” — a novel computational framework for App Review Mining, which performs comprehensive analytics from raw user reviews by (i) first extracting informative user reviews by …

Air Indexing For On-Demand Xml Data Broadcast, Weiwei Sun, Rongrui Qin, Jinjin Wu, Baihua Zheng

Research Collection School Of Computing and Information Systems

XML data broadcast is an efficient way to disseminate semi-structured information in wireless mobile environments. In this paper, we propose a novel two-tier index structure to facilitate the access of XML document in an on-demand broadcast system. It provides the clients with an overall image of all the XML documents available at the server side and hence enables the clients to locate complete result sets accordingly. A pruning strategy is developed to cut down the index size and a two-tier structure is proposed to further remove any redundant information. In addition, two index distribution strategies, namely naive distribution and partial …

On Efficient Reverse Skyline Query Processing, Yunjun Gao, Qing Liu, Baihua Zheng, Gang Chen

Research Collection School Of Computing and Information Systems

Given a D-dimensional data set P and a query point q, a reverse skyline query (RSQ) returns all the data objects in P whose dynamic skyline contains q. It is important for many real life applications such as business planning and environmental monitoring. Currently, the state-of-the-art algorithm for answering the RSQ is the reverse skyline using skyline approximations (RSSA) algorithm, which is based on the precomputed approximations of the skylines. Although RSSA has some desirable features, e.g., applicability to arbitrary data distributions and dimensions, it needs for multiple accesses of the same nodes, incurring redundant I/O and CPU costs. In …

Graph-Based Semi-Supervised Learning: Realizing Pointwise Smoothness Probabilistically, Yuan Fang, Kevin Chen-Chuan Chang, Hady W. Lauw

Research Collection School Of Computing and Information Systems

As the central notion in semi-supervised learning, smoothness is often realized on a graph representation of the data. In this paper, we study two complementary dimensions of smoothness: its pointwise nature and probabilistic modeling. While no existing graph-based work exploits them in conjunction, we encompass both in a novel framework of Probabilistic Graph-based Pointwise Smoothness (PGP), building upon two foundational models of data closeness and label coupling. This new form of smoothness axiomatizes a set of probability constraints, which ultimately enables class prediction. Theoretically, we provide an error and robustness analysis of PGP. Empirically, we conduct extensive experiments to show …

Hydrographic Surface Modeling Through A Raster Based Spline Creation Method, Julie G. Alexander

LSU New Orleans Theses and Dissertations

The United States Army Corp of Engineers relies on accurate and detailed surface models for various construction projects and preventative measures. To aid in these efforts, it is necessary to work for advancements in surface model creation. Current methods for model creation include Delaunay triangulation, raster grid interpolation, and Hydraulic Spline grid generation. While these methods produce adequate surface models, attempts for improved methods can still be made.

A method for raster based spline creation is presented as a variation of the Hydraulic Spline algorithm. By implementing Hydraulic Splines in raster data instead of vector data, the model creation process …

Automatic Objects Removal For Scene Completion, Jianjun Yang, Yin Wang, Honggang Wang, Kun Hua, Wei Wang, Ju Shen

Computer Science Faculty Publications

With the explosive growth of Web-based cameras and mobile devices, billions of photographs are uploaded to the Internet. We can trivially collect a huge number of photo streams for various goals, such as 3D scene reconstruction and other big data applications. However, this is not an easy task due to the fact the retrieved photos are neither aligned nor calibrated. Furthermore, with the occlusion of unexpected foreground objects like people, vehicles, it is even more challenging to find feature correspondences and reconstruct realistic scenes. In this paper, we propose a structure-based image completion algorithm for object removal that produces visually …

On Modeling Community Behaviors And Sentiments In Microblogging, Tuan Anh Hoang, William Cohen, Ee Peng Lim

Research Collection School Of Computing and Information Systems

In this paper, we propose the CBS topic model, a probabilistic graphical model, to derive the user communities in microblogging networks based on the sentiments they express on their generated content and behaviors they adopt. As a topic model, CBS can uncover hidden topics and derive user topic distribution. In addition, our model associates topic-specific sentiments and behaviors with each user community. Notably, CBS has a general framework that accommodates multiple types of behaviors simultaneously. Our experiments on two Twitter datasets show that the CBS model can effectively mine the representative behaviors and emotional topics for each community. We also …

On Finding The Point Where There Is No Return: Turning Point Mining On Game Data, Wei Gong, Ee Peng Lim, Feida Zhu, Achananuparp Palakorn, David Lo

Research Collection School Of Computing and Information Systems

Gaming expertise is usually accumulated through playing or watching many game instances, and identifying critical moments in these game instances called turning points. Turning point rules (shorten as TPRs) are game patterns that almost always lead to some irreversible outcomes. In this paper, we formulate the notion of irreversible outcome property which can be combined with pattern mining so as to automatically extract TPRs from any given game datasets. We specifically extend the well-known PrefixSpan sequence mining algorithm by incorporating the irreversible outcome property. To show the usefulness of TPRs, we apply them to Tetris, a popular game. We mine …

Modeling Contextual Agreement In Preferences, Ha Loc Do, Hady Wirawan Lauw

Research Collection School Of Computing and Information Systems

Personalization, or customizing the experience of each individual user, is seen as a useful way to navigate the huge variety of choices on the Web today. A key tenet of personalization is the capacity to model user preferences. The paradigm has shifted from that of individual preferences, whereby we look at a user's past activities alone, to that of shared preferences, whereby we model the similarities in preferences between pairs of users (e.g., friends, people with similar interests). However, shared preferences are still too granular, because it assumes that a pair of users would share preferences across all items. We …

Retrieval-Based Face Annotation By Weak Label Regularized Local Coordinate Coding, Dayong Wang, Steven C. H. Hoi, Ying He, Jianke Zhu, Mei Tao, Jiebo Luo

Research Collection School Of Computing and Information Systems

Auto face annotation, which aims to detect human faces from a facial image and assign them proper human names, is a fundamental research problem and beneficial to many real-world applications. In this work, we address this problem by investigating a retrieval-based annotation scheme of mining massive web facial images that are freely available over the Internet. In particular, given a facial image, we first retrieve the top n similar instances from a large-scale web facial image database using content-based image retrieval techniques, and then use their labels for auto annotation. Such a scheme has two major challenges: 1) how to …

On Predicting User Affiliations Using Social Features In Online Social Networks, Minh Thap Nguyen

Dissertations and Theses Collection (Open Access)

User profiling such as user affiliation prediction in online social network is a challenging task, with many important applications in targeted marketing and personalized recommendation. The research task here is to predict some user affiliation attributes that suggest user participation in different social groups.

Online Feature Selection And Its Applications, Jialei Wang, Peilin Zhao, Steven C. H. Hoi, Rong Jin

Research Collection School Of Computing and Information Systems

Feature selection is an important technique for data mining. Despite its importance, most studies of feature selection are restricted to batch learning. Unlike traditional batch learning methods, online learning represents a promising family of efficient and scalable machine learning algorithms for large-scale applications. Most existing studies of online learning require accessing all the attributes/features of training instances. Such a classical setting is not always appropriate for real-world applications when data instances are of high dimensionality or it is expensive to acquire the full set of attributes/features. To address this limitation, we investigate the problem of online feature selection (OFS) in …

Social Correlation In Latent Spaces For Complex Networks, Freddy Chong Tat Chua

Dissertations and Theses Collection (Open Access)

This dissertation addresses the subject of measuring social correlation among users within a complex social network. Social correlation is closely related to the measurement of social influence in social sciences. While social influence focuses on the existence of causal influence among users, we take a computational approach to measure correlation strength among users based on their shared interactions. We call this social correlation. To formally model social correlation, we propose a framework which contains two major parts. The first part is that of representing users behavior in a computationally efficient and accurate manner. For example, social media users perform many …

Predicting Response In Mobile Advertising With Hierarchical Importance-Aware Factorization Machine, Richard Jayadi Oentaryo, Ee Peng Lim, Jia Wei Low, David Lo, Michael Finegold

Research Collection School Of Computing and Information Systems

Mobile advertising has recently seen dramatic growth, fueled by the global proliferation of mobile phones and devices. The task of predicting ad response is thus crucial for maximizing business revenue. However, ad response data change dynamically over time, and are subject to cold-start situations in which limited history hinders reliable prediction. There is also a need for a robust regression estimation for high prediction accuracy, and good ranking to distinguish the impacts of different ads. To this end, we develop a Hierarchical Importance-aware Factorization Machine (HIFM), which provides an effective generic latent factor framework that incorporates importance weights and hierarchical …

Cms Use Of A Data Federation, Kenneth A. Bloom, C.

Kenneth Bloom Publications

CMS is in the process of deploying an Xrootd based infrastructure to facilitate a global data federation. The services of the federation are available to export data from half the physical capacity and the majority of sites are configured to read data over the federation as a back-up. CMS began with a relatively modest set of use-cases for recovery of failed local file opens, debugging and visualization. CMS is finding that the data federation can be used to support small scale analysis and load balancing. Looking forward we see potential in using the federation to provide more flexibility in the …

Detecting Click Fraud In Online Advertising: A Data Mining Approach, Richard Oentaryo, Ee Peng Lim, Michael Finegold, David Lo, Feida Zhu, Clifton Phua, Eng-Yeow Cheu, Ghim-Eng Yap, Kelvin Sim, Kasun Perera, Bijay Neupane, Mustafa Faisal, Zeyar Aung, Wei Lee Woon, Wei Chen, Dhaval Patel, Daniel Berrar

Research Collection School Of Computing and Information Systems

Click fraud - the deliberate clicking on advertisements with no real interest on the product or service offered - is one of the most daunting problems in online advertising. Building an elective fraud detection method is thus pivotal for online advertising businesses. We organized a Fraud Detection in Mobile Advertising (FDMA) 2012 Competition, opening the opportunity for participants to work on real-world fraud data from BuzzCity Pte. Ltd., a global mobile advertising company based in Singapore. In particular, the task is to identify fraudulent publishers who generate illegitimate clicks, and distinguish them from normal publishers. The competition was held from …

Online Portfolio Selection: A Survey, Bin Li, Steven C. H. Hoi

Research Collection School Of Computing and Information Systems

Online portfolio selection is a fundamental problem in computational finance, which has been extensively studied across several research communities, including finance, statistics, artificial intelligence, machine learning, and data mining. This article aims to provide a comprehensive survey and a structural understanding of online portfolio selection techniques published in the literature. From an online machine learning perspective, we first formulate online portfolio selection as a sequential decision problem, and then we survey a variety of state-of-the-art approaches, which are grouped into several major categories, including benchmarks, Follow-the-Winner approaches, Follow-the-Loser approaches, Pattern-Matching--based approaches, and Meta-Learning Algorithms. In addition to the problem formulation …

Short-Term Inflation Forecasting Models For Nigeria, Sani I. Doguwa, Sarah O. Alade

CBN Journal of Applied Statistics (JAS)

Short-term inflation forecasting is an essential component of the monetary policy projections at the Central Bank of Nigeria. This paper proposes four short-term headline inflation forecasting models using the SARIMA and SARIMAX processes and compares their performance using the pseudo-out-of-sample forecasting procedure over July 2011 to September 2013. According to the results the best forecasting performance is demonstrated by the model based on the all items CPI estimated using the SARIMAX model. This model is, therefore, recommended for use in short-term forecasting of headline inflation in Nigeria. The forecasting performance up to eight months ahead, of the models based on …

An Efficient Two Sample Capture-Recapture Model With High Recaptures, Danjuma Jibasen, Yusuf J. Adams

CBN Journal of Applied Statistics (JAS)

This paper proposed an efficient two sample capture-recapture model (Ma) with high recaptures and compared it with the existing models like the model of no factor effect (Mo), behavioral response model (Mb) and the Petersen model (Ms), using simulated data. We found that the Petersen model provides a better estimate of the population size when the observations follow a hypergeometric distribution and the population is overestimated when recapture is high. It was also found that the proposed model provides a better estimator of the population size than the existing ones when the recapture is high. This model is particularly useful …

Causal Relationship Between Stock Market Index And Exchange Rate: Evidence From Nigeria, Abdulrasheed Zubair

CBN Journal of Applied Statistics (JAS)

This paper uses Johansen’s cointegration to test for the possibility of cointegration and Granger-causality to estimate the causal relationship between stock market index and monetary indicators (exchange rate and M2) before and during the global financial crisis for Nigeria, using monthly data for the period 2001–2011. Results suggest absence of long-run relationship before and during the crisis. The Granger-causality tests show a uni-directional causality running from M2 to ASI before the crisis while during the period of the crisis there is absence of causality between the variables. This suggests that ASI show responsiveness to M2. Thus, absence of the direct …

Investigating Chaos In The Nigerian Asset And Resource Management (Arm) Discovery Fund, Ibiyinka A. Fuwape, Samuel T. Ogunjo

CBN Journal of Applied Statistics (JAS)

This paper investigates chaos in a Nigerian mutual fund, Asset and Resource Management Company Limited (ARM) for a period of eleven years. The existence of chaotic signals in the data was identified by the reconstruction of the phase space of the daily closing price of the fund and the delay time was quantified using mutual information function and the embedding dimension by the false nearest neighbours, where the values were identified to be 15 and 20 respectively. The presence of chaotic signals in the ARM data was further confirmed by the correlation dimension method which yielded a dimension of 2.2 …

Modeling The Nigerian Inflation Rates Using Periodogram And Fourier Series Analysis, Chukwuemeka O. Omekara,, Emmanuel J. Ekpenyong, Micheal P. Ekerete

CBN Journal of Applied Statistics (JAS)

This work considers the application of Periodogram and Fourier Series Analysis to model all-items monthly inflation rates in Nigeria from 2003 to 2011. The main objectives are to identify inflation cycles, fit a suitable model to the data and make forecasts of future values. To achieve these objectives, monthly all-items inflation rates for the period were obtained from the Central Bank of Nigeria (CBN) website. Periodogram and Fourier series methods of analysis are used to analyze the data. Based on the analysis, it was found that inflation cycle within the period was fifty one (51) months, which coincides with the …

Nigerian Stock Index: A Search For Optimal Garch Model Using High Frequency Data, Olaoluwa Simon Yaya

CBN Journal of Applied Statistics (JAS)

This paper attempts to fit the best Generalized Autoregressive Conditional Heteroscedastic (GARCH) model for All Share Index (ASI) of Nigerian Stock Exchange (NSE) returns. A search is made on various GARCH variants specified on the assumptions of stationarity and asymmetry. Fractionally integrated types are also considered to capture the possibility of return series having property of long range dependency. The parameter estimations are carried out on the assumptions of normality and non-normality of GARCH innovations, with models and forecasts evaluated using information criteria and loss functions respectively. Under normality assumption, Hyperbolic GARCH (HYGARCH(1,d,1)) model is selected and Integrated GARCH (IGARCH(1,1)) …

Time Series Modeling Of Nigeria External Reserves, Iheanyichukwu S. Iwueze, Eleazar C. Nwogu, Valentine U. Nlebedim

CBN Journal of Applied Statistics (JAS)

This paper discusses the levels and trend of external reserves in Nigeria. The relevance of this lies in the fact that it could help to monitor the reserves and throw early warning signal about any economic crisis. Monthly data on Nigeria external reserves for the period January 1999 to December, 2008 derived from the 2008 CBN Statistical Bulletin was analyzed using ARIMA model. Results of the analyses show that (i) the data requires logarithmic transformation to stabilize the variance and make the distribution normal (ii) the appropriate model that best describes the pattern in the transformed data is the Autoregressive- …

A Markov Decision Process Approach To Optimal Control Of A Multi-Level Hierarchical Manpower System, Akaninyene U. Udom

CBN Journal of Applied Statistics (JAS)

A recurrent problem in manpower control is how to attain the desired structural configuration in an optimal way, since it is possible to reach a desired structural configuration using different control inputs. The major aim of this paper is to develop a Markov Decision Process for optimal control of a Multi-level Hierarchical Manpower System (MHMS) by promotion and interdepartmental transfers. This is examined under control by intervention and contraction cost Markov Decision Process.

Modeling Preferences With Availability Constraints, Bingtian Dai, Hady W. Lauw

Research Collection School Of Computing and Information Systems

User preferences are commonly learned from historical data whereby users express preferences for items, e.g., through consumption of products or services. Most work assumes that a user is not constrained in their selection of items. This assumption does not take into account the availability constraint, whereby users could only access some items, but not others. For example, in subscription-based systems, we can observe only those historical preferences on subscribed (available) items. However, the objective is to predict preferences on unsubscribed (unavailable) items, which do not appear in the historical observations due to their (lack of) availability. To model preferences in …

Topicsketch: Real-Time Bursty Topic Detection From Twitter, Wei Xie, Feida Zhu, Jing Jiang, Ee Peng Lim, Ke Wang

Research Collection School Of Computing and Information Systems

Twitter has become one of the largest platforms for users around the world to share anything happening around them with friends and beyond. A bursty topic in Twitter is one that triggers a surge of relevant tweets within a short time, which often reflects important events of mass interest. How to leverage Twitter for early detection of bursty topics has therefore become an important research problem with immense practical value. Despite the wealth of research work on topic modeling and analysis in Twitter, it remains a huge challenge to detect bursty topics in real-time. As existing methods can hardly scale …

Modeling Temporal Adoptions Using Dynamic Matrix Factorization, Freddy Chong-Tat Chua, Richard Jayadi Oentaryo, Ee Peng Lim

Research Collection School Of Computing and Information Systems

The problem of recommending items to users is relevant to many applications and the problem has often been solved using methods developed from Collaborative Filtering (CF). Collaborative Filtering model-based methods such as Matrix Factorization have been shown to produce good results for static rating-type data, but have not been applied to time-stamped item adoption data. In this paper, we adopted a Dynamic Matrix Factorization (DMF) technique to derive different temporal factorization models that can predict missing adoptions at different time steps in the users' adoption history. This DMF technique is an extension of the Non-negative Matrix Factorization (NMF) based on …

A Social Network-Empowered Research Analytics Framework For Project Selection, Thushari Silva, Zhiling Guo, Jian Ma, Hongbing Jiang, Huaping Chen

Research Collection School Of Computing and Information Systems

Traditional approaches for research project selection by government funding agencies mainly focus on the matching of research relevance by keywords or disciplines. Other research relevant information such as social connections (e.g., collaboration and co-authorship) and productivity (e.g., quality, quantity, and citations of published journal articles) of researchers is largely ignored. To overcome these limitations, this paper proposes a social network-empowered research analytics framework (RAF) for research project selections. Scholarmate.com, a professional research social network with easy access to research relevant information, serves as a platform to build researcher profiles from three dimensions, i.e., relevance, productivity and connectivity. Building upon profiles …

Social Sensing For Urban Crisis Management: The Case Of Singapore Haze, Philips Kokoh Prasetyo, Ming Gao, Ee Peng Lim, Christie N. Scollon

Research Collection School Of Computing and Information Systems

Sensing social media for trends and events has become possible as increasing number of users rely on social media to share information. In the event of a major disaster or social event, one can therefore study the event quickly by gathering and analyzing social media data. One can also design appropriate responses such as allocating resources to the affected areas, sharing event related information, and managing public anxiety. Past research on social event studies using social media often focused on one type of data analysis (e.g., hashtag clusters, diffusion of events, influential users, etc.) on a single social media data …