iDEA Collection: Research Day Posters (IST)

EIPW: A Knowledge-based Database Modeling Tool

Thu, 07 Apr 2011 22:58:59 GMT

Title: EIPW: A Knowledge-based Database Modeling Tool

Authors: Thonggoom, Ornsiri; Song, Il-Yeol; An, Yuan

Abstract: Reuse of already existing resources and solutions has become a strategy for cost reduction and efficient improvement in the information system development process. Currently, building a repository of reusable artifacts involves explication of human developer’s knowledge, which is a major obstacle in facilitating reuse of knowledge. In this research, we explore knowledge-based and pattern-based approaches that help database designers develop quality conceptual data models based on reusable patterns. Our methodology includes database reverse engineering concepts. We propose new types of reusable instance patterns containing knowledge about an application domain, called entity instance repository (EIR) and relationship instance repository (RIR), which can be automatically generated from prior designs. The knowledge-based system (KBS) with EIR, RIR, and WordNet called EIPW (Entity Instance Pattern WordNet) is developed. The empirical study was conducted to test the efficiency of the KBS and indicated the significant improvement in novice designer performance.

HMM-based artificial designer for search interface segmentation

Wed, 22 Apr 2009 22:58:59 GMT

Title: HMM-based artificial designer for search interface segmentation

Authors: Khare, Ritu; An, Yuan; Song, Il-Yeol

Abstract: Search interfaces are the primary "doors" to the data-rich and continually growing deep Web. Hence, their understanding is a prerequisite to the design of applications that are dedicated to make deep Web contents more useful. The problem of automatic search interface understanding has recently been addressed by several works using techniques such as rules, heuristics, and even machine learning. However, a very challenging portion (segmentation) of this problem has not received appropriate attention. In this poster, we present an approach to segment a search interface into logical attributes and assign semantic labels to the interface components. The key feature in our approach is a 2-layered hidden Markov model (HMM), which encodes the learned knowledge about designing a search interface for querying underlying databases. We tested our approach on the less-explored family of search interfaces belonging to scientific databases and found promising results. On comparison with an existing approach, our approach improved the segmentation accuracy by approximately 10%.

Social network data sharing and integration

Wed, 22 Apr 2009 22:58:59 GMT

Title: Social network data sharing and integration

Authors: Tang, Xuning

Abstract: Social networks are valuable resources for intelligence and law enforcement force in their investigations when they want to identify suspects, terrorist or criminal subgroups and their communication patterns. However, missing information in a terrorist or criminal social network always diminish the effectiveness of investigation. It is unfortunate that individual agency usually has only part of a terrorist or criminal social network due to their limited information sources. Sharing and integration of social networks from different agencies help increasing its effectiveness; however, information sharing is usually forbidden due to the concern of privacy preservation. Our goal is to provide a framework of integrating social networks and doing graph mining while preserving information privacy. We will compare different sub-graph generalization methods and design a mechanism to integrate the generalized information to conduct social network analysis and mining.

Use of Microblogging in an Online Course

Thu, 07 Apr 2011 22:58:59 GMT

Title: Use of Microblogging in an Online Course

Authors: Purcell, Michelle

Abstract: Microblogging has been proposed as having potential for aiding in learning. Articles suggest benefits such as increasing social presence and facilitating reflection and informal learning. However, little research has been conducted investigating its use by students and teachers in actual educational settings. As a first step to understand the use of microblogging for educational purposes, a qualitative analysis of tweets, Twitter messages of 140 characters or less, over a 4-week period for a graduate online class was performed. Five categories of use were identified: information sharing, information seeking, giving assistance, giving feedback, and coordinating group activity. Frequency of use by student and by usage category is discussed.

System Design for Health Information Search

Thu, 07 Apr 2011 22:58:59 GMT

Title: System Design for Health Information Search

Authors: Zarro, Michael

Abstract: Health information consumers today take a leading role in managing wellness, treatment, and care for themselves and others. Recent studies report that 80% of US adults online have looked for health-related information on the Web, and a majority access user-generated content. Traditional information gatekeepers, like doctors and hospitals, have been partially replaced with social media websites and other Internet resources. As technology advances and consumer information behaviors change, new systems should emerge to assist health information seekers. This poster describes a prototype health information guided search system enhanced with expert-created and user-generated meta-data. The study is intended to further the understanding of guided search for health information seekers, and compare meta-data contributions of expert indexers to those of the untrained public. Evaluation of the interface with 10 laypersons and 7 medically-trained participants found the guided search features useful and desirable for both groups. Early results suggest expert-created meta-data can guide users of the system to relevant results retrieved from a general search engine.

An Associative Browser for the ACM Digital Library

Thu, 07 Apr 2011 22:58:59 GMT

Title: An Associative Browser for the ACM Digital Library

Authors: Zhang, Mi; Buzydlowski, Jan; Lin, Xia

Abstract: Classification systems have been a powerful browsing tool in traditional libraries. However, in digital libraries, the browsing capability of classification systems has yet to be fully explored. For example, in the recent release of ACM Digital Library interface, the classification can only be used as an option for advanced search. Users cannot view the hierarchical structure of the ACM Classification system while conducting searches or browsing search results. Nor can they take advantage of relationships of subject terms defined in the classification system for document browsing or search result filtering. In this poster, we describe a new associative browser for the ACM Digital Library. The new browser provides multiple ways for the user to engage with the classification system and search results. The major components and functions of the associative browser include: A fully interactive tree implementation of the classification system that allows the user to go up and down along the hierarchy of the classification system to select subject terms for searching or filtering search results. An interactive associative map for any subject term in the classification system that shows associative relationships of the terms. We define the term association by their co-occurrences in the whole ACM Digital Library. To get the co-occurrence counts, an extensive pre-calculation was run and the association of every two terms of the classification system was stored in a database. The associative map was generated by the pathfinder algorithm, which selects only the most relevant links to the centered terms. Integration of searching and browsing. The hierarchical browser and associative map could help users identify the structure of ACM classification system from different aspects. As each of user’s browsing action is linked to the search engine of the digital library, users will be able to use the structures of the classification system to guide their searches.

The Detent Mouse: A Practical Implementation of a Force Feedback Mouse

Thu, 07 Apr 2011 22:58:59 GMT

Title: The Detent Mouse: A Practical Implementation of a Force Feedback Mouse

Authors: Black, Alan

Abstract: The graphical user interface (GUI) has become ubiquitous in the computing world. Interfaces of this type generally rely upon pointing devices in order to interact with the system in an efficient manner. However, the task of operating a pointing device may present a challenge to users with motion control issues (e.g. patients with Parkinson’s disease). We have developed a GUI pointing device and associated software that is intended to help users who may experience motion control difficulties. This hardware/software system implements a computer mouse that provides increased resistance to movement, a detent, which is synchronized with GUI elements. The effect is to help the user maintain positioning of the mouse pointer over targets like hyperlinks and form controls (e.g. buttons and check boxes) within the context of a Web browser. The mouse itself is comprised of a commercially available optical mouse that has been fitted with an embedded electromagnet. The system provides a software programmable normal force that simulates a mechanical detent by varying the attraction between the mouse’s electromagnet and a ferrous mouse pad. The software component of the system is implemented as a JavaScript/XPConnect-based hardware driver and supporting code. These elements are injected into Web pages as the Web browser loads them. The page’s content is accessed via the document object model (DOM) whereby driver scripts and event handlers are added. This approach makes it possible to provide mouse positioning assistance for existing Web pages without modification by content authors, thereby providing a practical solution.

Information Sharing on Genealogy Message Boards

Thu, 07 Apr 2011 22:58:59 GMT

Title: Information Sharing on Genealogy Message Boards

Authors: Willever-Farr, Heather; Zach, Lisl

Abstract: According to the PEW Internet and American Life Project, 54 million Americans belong to a family in which a family member has used the Internet to research their genealogy. Researchers have further found that amateur genealogists prize informal information sharing with other genealogists to advance their research interests, and that the Internet has played a key role in facilitating information sharing. Given the importance of the Internet for the sharing of genealogical information, this poster will examine the information seeking and sharing activities of genealogists on the heavily used message boards on Ancestry.com, a popular family history website. In this study, the researchers extracted text from 951 posts from Ancestry.com’s “United States - General” message board. Employing a grounded theory approach, content analysis was utilized to categorize the primary content theme of each message board post. In addition, quantitative data, such as the number of participants in the message board exchanges and the answer rate to posted inquiries, was also collected. The exchange of messages was also mapped to determine whether there were reciprocal information exchanges occurring. Analysis reveals that posts offering specific genealogical data in response to posted queries were the most common type of post, and family history queries were the second most common type of post. Of the 187 queries posted to the message board, 167 received responses. Of those 167 queries, an average of 2.81 responses per query were posted. Thus, it appears that the message board is serving as a communal research service for genealogists by genealogists. Further analysis of the messages reveals that information exchange was unequal in nature: the majority of the answers posted in response to family history queries were provided by a small number of frequent contributors. This suggests the presence of “super sharers,” who disproportionately provide information to other genealogists.

Learning Web Development: Challenges at an Earlier Stage of Computing Education

Mon, 29 Mar 2010 22:58:59 GMT

Title: Learning Web Development: Challenges at an Earlier Stage of Computing Education

Authors: Park, Tom; Wiedenbeck, Susan

Abstract: Web development can serve as a rich and motivating context for introducing computation to new audiences. However, little is known of the difficulties that non-technical students face when learning web development. We report a study of forty-nine library and information science students taking an introductory web development course. By applying content analysis to their online help forums, we found that over ninety percent of help-seeking instances related to development, instruction, and technology. We applied a second level of content analysis to the development challenges, discovering that basic topics like images and links presented significant barriers to further learning. We present these findings and discuss their implications for computing education curricula.

What Do You Call It? A Comparison of Library-Created and User-Created Metadata

Thu, 07 Apr 2011 22:58:59 GMT

Title: What Do You Call It? A Comparison of Library-Created and User-Created Metadata

Authors: Hall, Catherine; Zarro, Michael

Abstract: In this poster, we describe an exploratory study comparing the abstracting and indexing practices of a semi-expert LIS community (metadata creators for the digital library, ipl2) and the social tags generated by Delicious users for the same corpus of materials. We find over 88% of the resources in the ipl2 History collection were tagged at least once in Delicious. Overlap between the tags applied to ipl2 resources and indexing shows terms that the two groups use are similar enough to be useful, yet dissimilar enough to provide new access points and descriptors.