Databases and Information Systems Commons^™

Using English Information In Non-English Web Search, Wei Gao, Wei Gao, Ming Zhou

Research Collection School Of Computing and Information Systems

The leading web search engines have spent a decade building highly specialized ranking functions for English web pages. One of the reasons these ranking functions are effective is that they are designed around features such as PageRank, automatic query and domain taxonomies, and click-through information, etc. Unfortunately, many of these features are absent or altered in other languages. In this work, we show how to exploit these English features for a subset of Chinese queries which we call linguistically non-local (LNL). LNL Chinese queries have a minimally ambiguous English translation which also functions as a good English query. We first …

Adapting Ranking Functions To User Preference, Keke Chen, Ya Zhang, Zhaohui Zheng, Hongyuan Zha, Gordon Sun

Kno.e.sis Publications

Learning to rank has become a popular method for web search ranking. Traditionally, expert-judged examples are the major training resource for machine learned web ranking, which is expensive to get for training a satisfactory ranking function. The demands for generating specific web search ranking functions tailored for different domains, such as ranking functions for different regions, have aggravated this problem. Recently, a few methods have been proposed to extract training examples from user clickthrough log. Due to the low cost of getting user preference data, it is attractive to combine these examples in training ranking functions. However, because of the …

Description Logic Reasoning With Decision Diagrams: Compiling Shiq To Disjunctive Datalog, Sebastian Rudolph

Kno.e.sis Publications

We propose a novel method for reasoning in the description logic SHIQ. After a satisfiability preserving transformation from SHIQ to the description logic ALCIb, the obtained ALCIb Tbox T is converted into an ordered binary decision diagram (OBDD) which represents a canonical model for T. This OBDD is turned into a disjunctive datalog program that can be used for Abox reasoning. The algorithm is worst-case optimal w.r.t. data complexity, and admits easy extensions with DL-safe rules and ground conjunctive queries.

Near-Duplicate Keyframe Retrieval By Nonrigid Image Matching, Jianke Zhu, Steven C. H. Hoi, Michael R. Lyu, Shuicheng Yan

Research Collection School Of Computing and Information Systems

Near-duplicate image retrieval plays an important role in many real-world multimedia applications. Most previous approaches have some limitations. For example, conventional appearance-based methods may suffer from the illumination variations and occlusion issue, and local feature correspondence-based methods often do not consider local deformations and the spatial coherence between two point sets. In this paper, we propose a novel and effective Nonrigid Image Matching (NIM) approach to tackle the task of near-duplicate keyframe retrieval from real-world video corpora. In contrast to previous approaches, the NIM technique can recover an explicit mapping between two near-duplicate images with a few deformation parameters and …

Bayesian Tensor Approach For 3-D Face Modeling, Dacheng Tao, Mingli Song, Xuelong Li, Jialie Shen, Jimeng Sun, Xindong Wu, Christos Faloutsos, Stephen J. Maybank

Research Collection School Of Computing and Information Systems

Effectively modeling a collection of three-dimensional (3-D) faces is an important task in various applications, especially facial expression-driven ones, e.g., expression generation, retargeting, and synthesis. These 3-D faces naturally form a set of second-order tensors-one modality for identity and the other for expression. The number of these second-order tensors is three times of that of the vertices for 3-D face modeling. As for algorithms, Bayesian data modeling, which is a natural data analysis tool, has been widely applied with great success; however, it works only for vector data. Therefore, there is a gap between tensor-based representation and vector-based data analysis …

Recursive Pattern Based Hybrid Supervised Training, Kiruthika Ramanathan, Sheng Uei Guan

Research Collection School Of Computing and Information Systems

We propose, theorize and implement the Recursive Pattern-based Hybrid Supervised (RPHS) learning algorithm. The algorithm makes use of the concept of pseudo global optimal solutions to evolve a set of neural networks, each of which can solve correctly a subset of patterns. The pattern-based algorithm uses the topology of training and validation data patterns to find a set of pseudo-optima, each learning a subset of patterns. It is therefore well adapted to the pattern set provided. We begin by showing that finding a set of local optimal solutions is theoretically equivalent, and more efficient, to finding a single global optimum …

Output Regularized Metric Learning With Side Information, Wei Liu, Steven C. H. Hoi, Jianzhuang Liu

Research Collection School Of Computing and Information Systems

Distance metric learning has been widely investigated in machine learning and information retrieval. In this paper, we study a particular content-based image retrieval application of learning distance metrics from historical relevance feedback log data, which leads to a novel scenario called collaborative image retrieval. The log data provide the side information expressed as relevance judgements between image pairs. Exploiting the side information as well as inherent neighborhood structures among examples, we design a convex regularizer upon which a novel distance metric learning approach, named output regularized metric learning, is presented to tackle collaborative image retrieval. Different from previous distance metric …

Event Detection With Common User Interests, Meishan Hu, Aixin Sun, Ee Peng Lim

Research Collection School Of Computing and Information Systems

In this paper, we aim at detecting events of common user interests from huge volume of user-generated content. The degree of interest from common users in an event is evidenced by a significant surge of event-related queries issued to search for documents (e.g., news articles, blog posts) relevant to the event. Taking the stream of queries from users and the stream of documents as input, our proposed framework seamlessly integrates the two streams into a single stream of query profiles. A query profile is a set of documents matching a query at a given time. With the single stream of …

Comparison Of Online Social Relations In Volume Vs Interaction: A Case Study Of Cyworld, Hyunwoo Chun, Haewoon Kwak, Young-Ho Eom, Yong-Yeol Ahn, Sue Moon, Hawoong. Jeong

Research Collection School Of Computing and Information Systems

Online social networking services are among the most popular Internet services according to Alexa.com and have become a key feature in many Internet services. Users interact through various features of online social networking services: making friend relationships, sharing their photos, and writing comments. These friend relationships are expected to become a key to many other features in web services, such as recommendation engines, security measures, online search, and personalization issues. However, we have very limited knowledge on how much interaction actually takes place over friend relationships declared online. A friend relationship only marks the beginning of online interaction.Does the interaction …

An Effective Approach To 3d Deformable Surface Tracking, Jianke Zhu, Steven C. H. Hoi, Zenglin Xu, Michael R. Lyu

Research Collection School Of Computing and Information Systems

The key challenge with 3D deformable surface tracking arises from the difficulty in estimating a large number of 3D shape parameters from noisy observations. A recent state-of-the-art approach attacks this problem by formulating it as a Second Order Cone Programming (SOCP) feasibility problem. The main drawback of this solution is the high computational cost. In this paper, we first reformulate the problem into an unconstrained quadratic optimization problem. Instead of handling a large set of complicated SOCP constraints, our new formulation can be solved very efficiently by resolving a set of sparse linear equations. Based on the new framework, a …

Representative Entry Selection For Profiling Blogs, Jinfeng Zhuang, Steven C. H. Hoi, Aixin Sun, Rong Jin

Research Collection School Of Computing and Information Systems

Many applications on blog search and mining often meet the challenge of handling huge volume of blog data, in which one single blog could contain hundreds or even thousands of entries. We investigate novel techniques for profiling blogs by selecting a subset of representative entries for each blog. We propose two principles for guiding the entry selection task: representativeness and diversity. Further, we formulate the entry selection task into a combinatorial optimization problem and propose a greedy yet effective algorithm for finding a good approximate solution by exploiting the theory of submodular functions. We suggest blog classification for judging the …

Leveraging Social Context For Searching Social Media, Marc Smith, Vladimir Barash, Lise Getoor, Hady W. Lauw

Research Collection School Of Computing and Information Systems

The ability to utilize and benefit from today's explosion of social media sites depends on providing tools that allow users to productively participate. In order to participate, users must be able to find resources (both people and information) that they find valuable. Here, we argue that in order to do this effectively, we should make use of a user's "social context". A user's social context includes both their personal social context (their friends and the communities to which they belong) and their community social context (their role and identity in different communities).

Ontology Enhanced Web Image Retrieval: Aided By Wikipedia And Spreading Activation Theory, Huan Wang, Xing Jiang, Liang-Tien Chia, Ah-Hwee Tan

Research Collection School Of Computing and Information Systems

Ontology, as an efective approach to bridge the semantic gap in various domains, has attracted a lot of interests from multimedia researchers. Among the numerous possibilities enabled by ontology, we are particularly interested in exploiting ontology for a better understanding of media task (particularly, images) on the World Wide Web. To achieve our goal, two open issues are inevitably involved: 1) How to avoid the tedious manual work for ontology construction? 2) What are the effective inference models when using an ontology? Recent works about ontology learned from Wikipedia has been reported in conferences targeting the areas of knowledge management …

Spatio-Temporal Efficiency In A Taxi Dispatch System, Darshan Santani, Rajesh Krishna Balan, C. Jason Woodard

Research Collection School Of Computing and Information Systems

In this paper, we present an empirical analysis of the GPS-enabled taxi dispatch system used by the world’s second largest land transportation company. We first summarize the collective dynamics of the more than 6,000 taxicabs in this fleet. Next, we propose a simple method for evaluating the efficiency of the system over a given period of time and geographic zone. Our method yields valuable insights into system performance—in particular, revealing significant inefficiencies that should command the attention of the fleet operator. For example, despite the state of the art dispatching system employed by the company, we find imbalances in supply …

Segmenting Brain Tumors Using Pseudo-Conditional Random Fields, Chi-Hoon Lee, Shaojun Wang, Albert Murtha, Matthew R.G. Brown, Russell Greiner

Kno.e.sis Publications

Locating Brain tumor segmentation within MR (magnetic resonance) images is integral to the treatment of brain cancer. This segmentation task requires classifying each voxel as either tumor or non-tumor, based on a description of that voxel. Unfortunately, standard classifiers, such as Logistic Regression (LR) and Support Vector Machines (SVM), typically have limited accuracy as they treat voxels as independent and identically distributed (iid). Approaches based on random fields, which are able to incorporate spatial constraints, have recently been applied to brain tumor segmentation with notable performance improvement over iid classifiers. However, previous random field systems involved computationally intractable …

A Faceted Classification Based Approach To Search And Rank Web Apis, Karthik Gomadam, Ajith Harshana Ranabahu, Meenakshi Nagarajan, Amit P. Sheth, Kunal Verma

Kno.e.sis Publications

Web application hybrids, popularly known as mashups, are created by integrating services on the Web using their APIs. Support for finding an API is currently provided by generic search engines or domain specific solutions such as Google and ProgrammableWeb. Shortcomings of both these solutions in terms of and reliance on user tags make the task of identifying an API challenging. Since these APIs are described in HTML documents, it is essential to look beyond the boundaries of current approaches to Web service discovery that rely on formal descriptions. In this work, we present a faceted approach to searching and ranking …

Semantics Enhanced Services: Meteor-S, Sawsdl And Sa-Rest, Amit P. Sheth, Karthik Gomadam, Ajith Harshana Ranabahu

Kno.e.sis Publications

Services Research Lab at the Knoesis center and the LSDIS lab at University of Georgia have played a significant role in advancing the state of research in the areas of workflow management, semantic Web services and service oriented computing. Starting with the METEOR workflow management system in the 90's, researchers have addressed key issues in the area of semantic Web services and more recently, in the domain of RESTful services and Web 2.0. In this article, we present a brief discussion on the various contributions of METEOR-S including SAWSDL, publication and discovery of semantic Web services, data mediation, dynamic configuration …

Special Issue Introduction: Hci Studies In Mis, Fiona Fui-Hoon Nah, Xiaowen Fang, Traci Hess, Weiyin Hong

Research Collection School Of Computing and Information Systems

We are grateful to the editors-in-chief for this opportunity and their strong support of the second AIS SIGHCI-sponsored special issue on HCI studies in MIS. We also thank the following reviewers who have played an important role in the development of the manuscripts included in this special issue: Steven Bellman, Damon Campbell, Jinwei Cao, Jane Carey, Andrea Everard, Mark Fuller, Matt Germonprez, Maggie Guo, Susanna Ho, De Liu, Hong Sheng, Chuan Hoo Tan, Horst Treiblmaier, June Wei, and Yunjie Calvin Xu.

Cascade Rsvm In Peer-To-Peer Network, Hock Hee Ang, Vivekanand Gopalkrishnan, Steven C. H. Hoi, Wee Keong Ng

Research Collection School Of Computing and Information Systems

The goal of distributed learning in P2P networks is to achieve results as close as possible to those from centralized approaches. Learning models of classification in a P2P network faces several challenges like scalability, peer dynamism, asynchronism and data privacy preservation. In this paper, we study the feasibility of building SVM classifiers in a P2P network. We show how cascading SVM can be mapped to a P2P network of data propagation. Our proposed P2P SVM provides a method for constructing classifiers in P2P networks with classification accuracy comparable to centralized classifiers and better than other distributed classifiers. The proposed algorithm …

Relative Importance, Specific Investment And Ownership In Interorganizational Systems., Kunsoo Han, Robert J. Kauffman, Barrie R. Nault

Research Collection School Of Computing and Information Systems

Implementation and maintenance of interorganizational systems (IOS) require investments by all the participating firms. Compared with intraorganizational systems, however, there are additional uncertainties and risks. This is because the benefits of IOS investment depend not only on a firm's own decisions, but also on those of its business partners. Without appropriate levels of investment by all the firms participating in an IOS, they cannot reap the full benefits. Drawing upon the literature in institutional economics, we examine IOS ownership as a means to induce value-maximizing noncontractible investments. We model the impact of two factors derived from the theory of incomplete …

Developing A Virtual City For Emergency Preparedness Planning And Training, Jon K. Morgan

Theses

Existing techniques for emergency preparedness planning and training fail or lack the ability to convey training on a broad scale and timely fashion. Skill sets that are required for planning, mitigation, response and recovery issues are lost through information overload or failure to identify other channels in which to convey the information. In order to resolve some of the issues with currently existing methods such as tabletop training exercises (TTX), instructional video learning and full-scale exercises we can turn to virtual environments.

In a virtual environment teams can interact with their surroundings from the comfort of the office without having …

Challenges Of Creating A Knowledge-Based Society: Education & Research For India & Gujarat, Amit P. Sheth

Kno.e.sis Publications

No abstract provided.

A Generic Approach And Framework For Managing Complex Information, Essam Mansour

Doctoral

Several application domains, such as healthcare, incorporate domain knowledge into their day-to-day activities to standardise and enhance their performance. Such incorporation produces complex information, which contains two main clusters (active and passive) of information that have internal connections between them. The active cluster determines the recommended procedure that should be taken as a reaction to specific situations. The passive cluster determines the information that describes these situations and other descriptive information plus the execution history of the complex information. In the healthcare domain, a medical patient plan is an example for complex information produced during the disease management activity from …

An Infrastructure For Performance Measurement And Comparison Of Information Retrieval Solutions, Gary Saunders

Theses and Dissertations

The amount of information available on both public and private networks continues to grow at a phenomenal rate. This information is contained within a wide variety of objects, including documents, e-mail archives, medical records, manuals, pictures and music. To be of any value, this data must be easily searchable and accessible. Information Retrieval (IR) is concerned with the ability to find and gain access to relevant information. As electronic data repositories continue to proliferate, so too, grows the variety of methods used to locate and access the information contained therein. Similarly, the introduction of innovative retrieval strategies—and the optimization of …

Tcruzikb: Enabling Complex Queries For Genomic Data Exploration, Pablo N. Mendes, Bobby Mcknight, Amit P. Sheth, Jessica C. Kissinger

Kno.e.sis Publications

We developed a novel analytical environment to aid in the examination of the extensive amount of interconnected data available for genome projects. Our focus is to enable flexibility and abstraction from implementation details, while retaining the expressivity required for post-genomic research. To achieve this goal, we associated genomics data to ontologies and implemented a query formulation and execution environment with added visualization capabilities. We use ontology schemas to guide the user through the process of building complex queries in a flexible Web interface. Queries are serialized in SPARQL and sent to servers via Ajax. A component for visualization of the …

Text Analytics For Semantic Computing - The Good, The Bad And The Ugly, Meenakshi Nagarajan, Cartic Ramakrishnan, Amit P. Sheth

Kno.e.sis Publications

This tutorial was give at the Second IEEE International Conference on Semantic Computing Santa Clara, CA, USA - August 4-7, 2008.

E-Transcript Web Services System Supporting Dynamic Conversion Between Xml And Edi, Myungjae Kwak '11, Woohyun Kang '14, Gondy Leroy, Samir Chatterjee

CGU Faculty Publications and Research

As XML becomes a standard for communications between distributed heterogeneous machines, many schools plan to implement Web Services systems using the XML e-transcript (electronic transcript) standard. We propose a framework that supports both XML e-transcript Web Services and existing EDI e-transcript systems. The framework uses the workflow engine to exploit the benefits of workflow management mechanisms. The workflow engine manages the e-transcript business process by enacting and completing the tasks and sub-processes within the main business process. We implemented the proposed framework by using various open source projects including Java, Eclipse, and Apache Software Foundation’s Web Services projects. Compared with …

Connectionist Model Generation: A First-Order Approach, Sebastian Bader, Pascal Hitzler, Steffen Holldobler

Computer Science and Engineering Faculty Publications

Knowledge-based artificial neural networks have been applied quite successfully to propositional knowledge representation and reasoning tasks. However, as soon as these tasks are extended to structured objects and structure-sensitive processes as expressed e.g., by means of first-order predicate logic, it is not obvious at all what neural-symbolic systems would look like such that they are truly connectionist, are able to learn, and allow for a declarative reading and logical reasoning at the same time. The core method aims at such an integration. It is a method for connectionist model generation using recurrent networks with feed-forward core. We show in this …

Mediatability: Estimating The Degree Of Human Involvement In Xml Schema Mediation, Karthik Gomadam, Ajith Harshana Ranabahu, Lakshmish Ramaswamy, Amit P. Sheth, Kunal Verma

Kno.e.sis Publications

Mediation and integration of data are significant challenges because the number of services on the Web, and heterogeneities in their data representation, continue to increase rapidly. To address these challenges we introduce a new measure, mediatability, which is a quantifiable and computable metric for the degree of human involvement in XML schema mediation. We present an efficient algorithm to compute mediatability and an experimental study to analyze how semantic annotations affect the ease of mediating between two schemas. We validate our approach by comparing mediatability scores generated by our system with user-perceived difficulty. We also evaluate the scalability of our …

An Xml-Based Approach To Handling Tables In Documents, Krishnaprasad Thirunarayan, Trivikram Immaneni

Kno.e.sis Publications

We explore application of XML technology for handling tables in legacy semi-structured documents. Specifically, we analyze annotating heterogeneous documents containing tables to obtain a formalized XML Master document that improves traceability (hence easing verification and update) and enables manipulation using XSLT stylesheets. This approach is useful when table instances far outnumber distinct table types because the effort required to annotate a table instance is relatively less compared to formalizing table processing that respects table’s semantics. This work is also relevant for authoring new documents with tables that should be accessible to both humans and machines.