Ryan Richardson

Enhancing usability in CITIDEL: multimodal, multilingual, and interactive visualization interfaces

We describe four usability-enhancing interfaces to CITIDEL aimed at improving the user experience and supporting personalized information access by targeted communities. These comprise: a multimodal interaction facility with capability for out-of-turn input, interactive visualizations for exploratory analysis, a translation center exposing multilingual interfaces, as well as traditional usability enhancements. Pilot studies demonstrate the resulting improvements in quality, as measured across a number of metrics.

Java MARIAN: From an OPAC to a Modern Digital Library System

This paper describes the Java MARIAN system, its implementation, and its evolution from a C++ Online Public Access Catalog (OPAC) to a modern and complete digital library system. We focus on design, architectural, and implementation issues including: support for storage, retrieval, and automatic generation of collections of semi-structured digital objects; uniform and powerful representations based on semantic networks, digital library specific datatypes, and weighted sets; rich communities of searchers and fusion modules; and support for distributed computation, multi-lingual retrieval, and personalization. We present applications and some use statistics.

Support Vector Machines to Weight Voters in a Voting System of Entity Extractors

Support vector machines are used to combine the outputs of multiple entity extractors, thus creating a composite entity extraction system. The composite system has a significantly higher f-measure than any of the component systems. Compared to a standard voting technique for combining the results of multiple entity extractors, the SVM approach produces comparable precision and recall statistics but tends to utilize fewer of the component entity extractors, thus providing superior computational efficiency, which is critical in practical applications. In this paper, we present our experimental results of comparing a standard voting technique with SVM that each aggregate four entity extractors. We also describe our future plans of integrating agent-based technology into our experimental testbed where we examine the evolution of composite techniques as part of the analysis stream. Given that much of the improvement comes from tuning the algorithms to the data stream with a human-in-the-loop, we are considering the merits of employing cognitive agents that are strategically embedded in the workflow for processing data. As we tune the algorithms for better performance on the data streams, we envision agents learning the patterns of data streams and apply the appropriate tuning to ensure optimality.

Using concept maps in digital libraries as a cross-language resource discovery tool

The concept map, first suggested by Joseph Novak, has been extensively studied as a way for learners to increase understanding. We are automatically generating and translating concept maps from electronic theses and dissertations, for both English and Spanish, as a DL aid to discovery and summarization

Using concept maps as a cross-language resource discovery tool for large documents in digital libraries

Project Gutenburg, the Million Book Project, the Networked Digital Library of Theses and Dissertations, Amazons book search service, and the recently announced collaboration of Google and leading libraries, all aim to make available large numbers of book-length objects, in a variety of languages. Traditional approaches to discovering a suitable book for a particular purpose have generally relied on catalog records, sometimes enhanced with abstracts. Full-text searching - popular, e.g., with legal and government documents

Harvesting: Broadening the Field of Distributed Information Retrieval

and passage retrieval techniques, suitable for encyclopedias and reference works, have not been adequately tested with large collections of large objects

Using bilingual ETD collections to mine phrase translations

This chapter argues that in addition to federated search and gathering (as by Web crawlers), harvesting is an important approach to address the needs for distributed IR. We highlight the use of the Open Archives Initiative Protocol for Metadata Harvesting, illustrating its use in three projects: OAD, NDLTD, and CITIDEL. We explain how traditional services can be extended in a user-centered fashion, providing details of our new: ESSEX search engine, multischeming browsing, and quality-oriented filtering (using rules and SVMs). We conclude with an overview of work in progress on logging and component architectures, as well as a summary of our findings.

Mirroring an OAI archive on the I2-DSI channel

Phrase translation lists can enhance cross-language information retrieval. However, finding translations for technical phrases is difficult. Bilingual dictionaries have limited coverage for specialized fields, and even more limited coverage of technical phrases. Since phrases can have very specific meanings in technical fields, this limits the quality of translations produced by generic machine translation systems. We hypothesize that digital libraries of electronic theses and dissertations (ETDs) are a good source of technical phrase translations. We have acquired a collection of 3,086 Spanish ETDs about computer science from Scirus, the Universidad Nacional Autónoma de México (Mexico City), and Universidad de las Américas (Puebla). By using English ETDs from NDLTD, we have a comparable corpus of computing-related documents from which to mine phrase translations. We describe our method and its formative evaluation.

Using digital library components for biodiversity systems

The Open Archives Initiative (OAI) promotes interoperability among digital libraries and has created a protocol for data providers to easily export their metadata. One problem with this approach is that some of the more popular servers quickly become heavily loaded. The obvious solution is replication. Fortunately, the Internet-2 Distributed Storage Infrastructure (I2-DSI) has begun to develop technology for highly distributed transparent replication of servers. This paper presents our solution for transparent mirroring of OAI repositories within the I2-DSI.

Evaluating Concept Maps As A Cross-Language Knowledge Discovery Tool for NDLTD

Biodiversity information systems (BISs) involve all kinds of heterogeneous data, which include ecological and geographical features. However, available information systems offer very limited support for managing such data in an integrated fashion, and such integration is often based on geographic coordinates alone. Furthermore, such systems do not fully support image content management (e.g., photos of landscapes or living organisms), a requirement of many BIS end-users. In order to meet their needs, these users e.g., biologists, environmental experts - often have to alternate between distinct biodiversity and image information systems to combine information extracted from them. This cumbersome operational procedure is forced on users by lack of interoperability among these systems. This hampers the addition of new data sources, as well as cooperation among scientists. The approach provided in this project to meet these issues is based on taking advantage of advances in digital library (DL) innovations to integrate networked collections of heterogeneous data. It focuses on creating the basis for a biodiversity information system under the digital library perspective, combining new techniques of content-based image retrieval and database query processing mechanisms. This approach solves the problem of system switching, and provides users with a flexible platform from which to tailor a BIS to their needs.