Databases and Information Systems Commons^™

Brief Announcement: Space Adaptation: Privacy-Preserving Multiparty Collaborative Mining With Geometric Perturbation, Keke Chen, Ling Liu

Kno.e.sis Publications

No abstract provided.

Evolution And Maintenance Of Frequent Pattern Space When Transactions Are Removed, Mengling Feng, Guozhu Dong, Jinyan Li, Yap-Peng Tan, Limsoon Wong

Kno.e.sis Publications

This paper addresses the maintenance of discovered frequent patterns when a batch of transactions are removed from the original dataset. We conduct an in-depth investigation on how the frequent pattern space evolves under transaction removal updates using the concept of equivalence classes. Inspired by the evolution analysis, an effective and exact algorithm TRUM is proposed to maintain frequent patterns. Experimental results demonstrate that our algorithm outperforms representative state-of-the-art algorithms.

Best Practices For Implementing Agile Methods: A Guide For Department Of Defense Software Developers, Ann L. Fruhling, Alvin E. Tarrell

Information Systems and Quantitative Analysis Faculty Publications

Traditional plan-driven software development has been widely used in the government because it's considered to be less risky, more consistent, and structured. But there has been a shift from this approach to Agile methods which are more flexible, resulting in fast releases by working in an incremental fashion to adapt to the reality of the changing or unclear requirements.

This report describes the Agile software development philosophy, methods, and best practices in launching software design projects using the Agile approach. It is targeted to Defense Department software developers because they face broad challenges in creating enterprise-wide information systems, where Agile …

Brief Description And Analysis Of The Census Bureau's 2006 Population Estimates For Incorporated Places For Cleveland And Other Ohio Cities, Mark Salling

All Maxine Goodman Levin School of Urban Affairs Publications

No abstract provided.

Factors Influencing Speed Of Cancer Diagnosis In Rural Wa, Moyez Jiwa, Georgia Halkett, Samar Aoun, Hayley Arnet, Marthe Smith, Megan Pilkington, Cheryl Mcmullen

Research outputs pre 2011

Introduction: The speed of diagnosis impacts on prognosis and survival in all types of cancer. In most cases survival and prognosis are significantly worse in rural and remote Australian populations who have less access to diagnostic and therapeutic services than metropolitan communities in this country. Research suggests that in general delays in diagnosis were a factor of misdiagnosis, the confounding effect of existing conditions and delayed or misleading investigation of symptoms. The aim of this study is to further explore the factors that impact on the speed of diagnosis in rural Western Australia with direct reference to General Practitioners (GPs) …

Mining Multiple Visual Appearances Of Semantics For Image Annotation, Hung-Khoon Tan, Chong-Wah Ngo

Research Collection School Of Computing and Information Systems

This paper investigates the problem of learning the visual semantics of keyword categories for automatic image annotation. Supervised learning algorithms which learn only a single concept point of a category are limited in their effectiveness for image annotation. We propose to use data mining techniques to mine multiple concepts, where each concept may consist of one or more visual parts, to capture the diverse visual appearances of a single keyword category. For training, we use the Apriori principle to efficiently mine a set of frequent blobsets to capture the semantics of a rich and diverse visual category. Each concept is …

Direct Code Access In Self-Organizing Neural Networks For Reinforcement Learning, Ah-Hwee Tan

Research Collection School Of Computing and Information Systems

TD-FALCON is a self-organizing neural network that incorporates Temporal Difference (TD) methods for reinforcement learning. Despite the advantages of fast and stable learning, TD-FALCON still relies on an iterative process to evaluate each available action in a decision cycle. To remove this deficiency, this paper presents a direct code access procedure whereby TD-FALCON conducts instantaneous searches for cognitive nodes that match with the current states and at the same time provide maximal reward values. Our comparative experiments show that TD-FALCON with direct code access produces comparable performance with the original TD-FALCON while improving significantly in computation efficiency and network complexity.

Integrating Semantic Templates With Decision Tree For Image Semantic Learning, Ying Liu, Dengsheng Zhang, Guojun Lu, Ah-Hwee Tan

Research Collection School Of Computing and Information Systems

Decision tree (DT) has great potential in image semantic learning due to its simplicity in implementation and its robustness to incomplete and noisy data. Decision tree learning naturally requires the input attributes to be nominal (discrete). However, proper discretization of continuous-valued image features is a difficult task. In this paper, we present a decision tree based image semantic learning method, which avoids the difficult image feature discretization problem by making use of semantic template (ST) defined for each concept in our database. A ST is the representative feature of a concept, generated from the low-level features of a collection of …

Searching And Tagging: Two Sides Of The Same Coin?, Qiaozhu Mei, Jing Jiang, Hang Su, Chengxiang Zhai

Research Collection School Of Computing and Information Systems

This paper presents the duality hypothesis of search and tagging, two important behaviors of web users. The hypothesis states that if a user views a document D in the search results for query Q, the user would tend to assign document $D$ a tag identical to or similar to Q; similarly, if a user tags a document D with a tag T, the user would tend to view document D if it is in the search results obtained using T as a query. We formalize this hypothesis with a unified probabilistic model for search and tagging, and show that empirical …

Anticipatory Event Detection Via Classification, He Qi, Kuiyu Chang, Ee Peng Lim

Research Collection School Of Computing and Information Systems

The idea of event detection is to identify interesting patterns from a constant stream of incoming news documents. Previous research in event detection has largely focused on identifying the first event or tracking subsequent events belonging to a set of pre-assigned topics such as earthquakes, airline disasters, etc. In this paper, we describe a new problem, called anticipatory event detection (AED), which aims to detect if a user-specified event has transpired. AED can be viewed as a personalized combination of event tracking and new event detection. We propose using sentence-level and document-level classification approaches to solve the AED problem for …

National Culture And Its Effects On Knowledge Communication In Online Virtual Communities, Keng Siau, Fiona Fui-Hoon Nah, Min Ling

Research Collection School Of Computing and Information Systems

Online virtual communities provide a powerful means of knowledge sharing. Despite the prevalence of online virtual communities, there is a paucity of research to investigate the effect of national culture differences on knowledge sharing in online virtual communities. Are there differences between online virtual communities from different national cultures? This research studies the differences in knowledge-sharing activities between US-based and China-based online virtual communities. Hofstede's dimensions of national culture serve as the theoretical foundation for this research.

Data Management Plans: Stages, Components, And Activities, Abbas S. Tavakoli, Kirby Jackson, Linda Moneyham, Kenneth D. Phillips, Carolyn Murdaugh, Gene Meding

Applications and Applied Mathematics: An International Journal (AAM)

Data management strategies have become increasingly important as new computer technologies allow for larger and more complex data sets to be analyzed easily. As a consequence, data management has become a specialty requiring specific skills and knowledge. Many new investigators have no formal training in management of data sets. This paper describes common basic strategies critical to the management of data as applied to a data set from a longitudinal study. The stages of data management are identified. Moreover, key components and strategies, at each stage are described.

Regression Cubes With Lossless Compression And Aggregation, Yixin Chen, Guozhu Dong, Jiawei Han, Jian Pei, Benjamin W. Wah, Jianyong Wang

Kno.e.sis Publications

As OLAP engines are widely used to support multidimensional data analysis, it is desirable to support in data cubes advanced statistical measures, such as regression and filtering, in addition to the traditional simple measures such as count and average. Such new measures will allow users to model, smooth, and predict the trends and patterns of data. Existing algorithms for simple distributive and algebraic measures are inadequate for efficient computation of statistical measures in a multidimensional space. In this paper, we propose a fundamentally new class of measures, compressible measures, in order to support efficient computation of the statistical models. For …

Implicit Online Learning With Kernels, Li Cheng, S. V. N. Vishwanathan, Dale Schuurmans, Shaojun Wang, Terry Caelli

Kno.e.sis Publications

We present two new algorithms for online learning in reproducing kernel Hilbert spaces. Our first algorithm, ILK (implicit online learning with kernels), employs a new, implicit update technique that can be applied to a wide variety of convex loss functions. We then introduce a bounded memory version, SILK (sparse ILK), that maintains a compact representation of the predictor without compromising solution quality, even in non-stationary environments. We prove loss bounds and analyze the convergence rate of both. Experimental evidence shows that our proposed algorithms outperform current methods on synthetic and real data.

Clique Percolation For Finding Naturally Cohesive And Overlapping Document Clusters, Wei Gao, Kam-Fai Wong, Yunqing Xia, Ruifeng Xu

Research Collection School Of Computing and Information Systems

Techniques for find document clusters mostly depend on models that impose strong explicit and/or implicit priori assumptions. As a consequence, the clustering effects tend to be unnatural and stray away from the intrinsic grouping natures of a document collection. We apply a novel graph-theoretic technique called Clique Percolation Method (CPM) for document clustering. In this method, a process of enumerating highly cohesive maximal document cliques is performed in a random graph, where those strongly adjacent cliques are mingled to form naturally overlapping clusters. Our clustering results can unveil the inherent structural connections of the underlying data. Experiments show that CPM …

Designing Web Sites For Customer Loyalty Across Business Domains: A Multilevel Analysis, S. Mithas, Narayanasamy Ramasubbu, M. S. Krishnan, C. Fornell

Research Collection School Of Computing and Information Systems

Web Sites are important components of Internet strategy for organizations. This paper develops a theoretical model for understanding the effect of Web site design elements on customer loyalty to a Web site. We show the relevance of the business domain of a Web site to gain a contextual understanding of relative importance of Web site design elements. We use a hierarchical linear modeling approach to model multilevel and cross-level interactions that have not been explicitly considered in previous research. By analyzing, data on more than 12,000 online customer surveys for 43 Web sites in several business domains, we find that …

Towards Effective Content-Based Music Retrieval With Multiple Acoustic Feature Combination, Jialie Shen, John Shepherd, Ann H. H. Ngu

Research Collection School Of Computing and Information Systems

In this paper, we present a new approach to constructing music descriptors to support efficient content-based music retrieval and classification. The system applies multiple musical properties combined with a hybrid architecture based on principal component analysis (PCA) and a multilayer perceptron neural network. This architecture enables straightforward incorporation of multiple musical feature vectors, based on properties such as timbral texture, pitch, and rhythm structure, into a single low-dimensioned vector that is more effective for classification than the larger individual feature vectors. The use of supervised training enables incorporation of human musical perception that further enhances the classification process. We compare …

Measuring Qualities Of Articles Contributed By Online Communities, Ee Peng Lim, Ba-Quy Vuong, Hady W. Lauw, Aixin Sun

Research Collection School Of Computing and Information Systems

Using open source Web editing software (e.g., wiki), online community users can now easily edit, review and publish articles collaboratively. While much useful knowledge can be derived from these articles, content users and critics are often concerned about their qualities. In this paper, we develop two models, namely basic model and peer review model, for measuring the qualities of these articles and the authorities of their contributors. We represent collaboratively edited articles and their contributors in a bipartite graph. While the basic model measures an article's quality using both the authorities of contributors and the amount of contribution from each …

On The Lower Bound Of Local Optimums In K-Means Algorithms, Zhenjie Zhang, Bing Tian Dai, Anthony K.H. Tung

Research Collection School Of Computing and Information Systems

No abstract provided.

Rapid Identification Of Column Heterogeneity, Bing Tian Dai, Nick Koudas, Beng Chin Ooi, Divesh Srivastava, Suresh Venkatasubramanian

Research Collection School Of Computing and Information Systems

No abstract provided.

Query-Based Watermarking For Xml Data, Xuan Zhou, Hwee Hwa Pang, Kian-Lee Tan

Research Collection School Of Computing and Information Systems

As increasing amount of XML data is exchanged over the internet, copyright protection of this type of data is becoming an important requirement for many applications. In this paper, we introduce a rights protection scheme for XML data based on digital watermarking. One of the main challenges for watermarking XML data is that the data could be easily reorganized by an adversary in an attempt to destroy any embedded watermark. To overcome it, we propose a query-based watermarking scheme, which creates queries to identify available watermarking capacity, such that watermarks could be recovered from reorganized data through query rewriting. The …

An Experimental Study On U-Commerce Adoption: Impact Of Personalization And Privacy Concerns, H. Sheng, Fiona Fui-Hoon Nah, Keng Siau

Research Collection School Of Computing and Information Systems

Ubiquitous commerce (u-commerce) represents "anytime, anywhere" commerce. U-commerce can provide a high level of personalization, which can bring significant benefits to customers. However, privacy is a major concern to customers and an obstacle to the adoption of u-commerce. This research examines how personalization and context can impact customers' privacy concerns as well as intention to adopt u-commerce applications. As u-commerce is new and emerging, we used the scenario-based approach to operationalize personalization and context in an experimental study. The experimental results show that the effects of personalization on customers' privacy concerns and adoption intention are situation dependent.

Identifying Difficulties In Learning Uml, Keng Siau, Poi-Peng Loo

Research Collection School Of Computing and Information Systems

Despite its recognition as a standard object-oriented modeling language, Unified Modeling Language (UML) has been criticized for such deficiencies as semantic inconsistencies, vagueness, and conflicting notations. the relationship between these deficiencies and the difficulties in the learning process is the focus of this study. A concept mapping technique is used to unveil the learning difficulties and suggestions for alleviating them are provided.

Towards Effective Content-Based Music Retrieval With Multiple Acoustic Feature Composition, Jialie Shen, John Shepherd, Ngu Ahh

Research Collection School Of Computing and Information Systems

In this paper, we present a new approach to constructing music descriptors to support efficient content-based music retrieval and classification. The system applies multiple musical properties combined with a hybrid architecture based on principal component analysis (PCA) and a multilayer perceptron neural network. This architecture enables straightforward incorporation of multiple musical feature vectors, based on properties such as timbral texture, pitch, and rhythm structure, into a single low-dimensioned vector that is more effective for classification than the larger individual feature vectors. The use of supervised training enables incorporation of human musical perception that further enhances the classification process. We compare …

Continuous Monitoring Of Knn Queries In Wireless Sensor Networks, Yuxia Yao, Xueyan Tang, Ee Peng Lim

Research Collection School Of Computing and Information Systems

Wireless sensor networks have been widely used for civilian and military applications, such as environmental monitoring and vehicle tracking. In these applications, continuous query processing is often required and their efficient evaluation is a critical requirement to be met. Due to the limited power supply for sensor nodes, energy efficiency is a major performance measure in such query evaluation. In this paper, we focus on continuous kNN query processing. We observe that the centralized data storage and monitoring schemes do not favor energy efficiency. We therefore propose a localized scheme to monitor long running nearest neighbor queries in sensor networks. …

Modeling Heterogeneous User Churn And Local Resilience Of Unstructured P2p Networks, Zhongmei Yao, Derek Leonard, Dmitri Loguinov, Xiaoming Wang

Computer Science Faculty Publications

Previous analytical results on the resilience of unstructured P2P systems have not explicitly modeled heterogeneity of user churn (i.e., difference in online behavior) or the impact of in-degree on system resilience. To overcome these limitations, we introduce a generic model of heterogeneous user churn, derive the distribution of the various metrics observed in prior experimental studies (e.g., lifetime distribution of joining users, joint distribution of session time of alive peers, and residual lifetime of a randomly selected user), derive several closed-form results on the transient behavior of in-degree, and eventually obtain the joint in/out degree isolation probability as a simple …

{Ontology: Resource} X {Matching : Mapping} X {Schema : Instance} :: Components Of The Same Challenge, Amit P. Sheth

Kno.e.sis Publications

Ontologies enable us to elevate syntactic and structural processing in an information system/Web to an information system/Web powered with semantic processing. Experience has shown that monolithic and tightly coupled approaches seldom succeed, and majority of information systems and applications will need to deal with plurality of ontologies in a loosely coupled environment (i.e., independently evolving ontologies and inter-ontology relationships, existence of different contexts for different users/applications etc.) Development of such loosely-coupled multi-ontology environments entails development of techniques for ontology mapping/alignment, multi-ontology query processing, and much more.

Active Semantic Electronic Medical Record, Amit P. Sheth, Sangeeta Agrawal, Jonathan Lathem, Nicole Oldham, H. Wingate, K. Gallagher

Kno.e.sis Publications

The healthcare industry is rapidly advancing towards the widespread use of electronic medical records systems to manage the increasingly large amount of patient data and reduce medical errors. In addition to patient data there is a large amount of data describing procedures, treatments, diagnoses, drugs, insurance plans, coverage, formularies and the relationships between these data sets. While practices have benefited from the use of EMRs, infusing these essential programs with rich domain knowledge and rules can greatly enhance their performance and ability to support clinical decisions. Active Semantic Electronic Medical Record (ASEMR) application discussed here uses Semantic Web technologies to …

On The Complexity Of Horn Description Logics, Markus Krotzsch, Sebastian Rudolph, Pascal Hitzler

Computer Science and Engineering Faculty Publications

Horn-SHIQ has been identified as a fragment of the description logic SHIQ for which inferencing is in PT_IME with respect to the size of the ABox. This enables reasoning with larger ABoxes in situations where the TBox is static, and represents one approach towards tractable description logic reasoning. In this paper, we show that reasoning in Horn-SHIQ, in spite of its low datacomplexity, is E_xpT_IME-hard with respect to the overall size of the knowledge base. While this result is not unexpected, the proof is not a mere modification of existing reductions since …

A Framework For Schema-Driven Relationship Discovery From Unstructured Text, Cartic Ramakrishnan, Krzysztof Kochut, Amit P. Sheth

Kno.e.sis Publications

We address the issue of extracting implicit and explicit relationships between entities in biomedical text. We argue that entities seldom occur in text in their simple form and that relationships in text relate the modified, complex forms of entities with each other. We present a rule-based method for (1) extraction of such complex entities and (2) relationships between them and (3) the conversion of such relationships into RDF. Furthermore, we present results that clearly demonstrate the utility of the generated RDF in discovering knowledge from text corpora by means of locating paths composed of the extracted relationships.