Information Retrieval Tools Research Articles

Text mining for Swiss-Prot curation: A story of success and failure

A text mining group has been set up at the Swiss Institute of Bioinformatics, with objective to develop and adapt information retrieval and extraction tools to help Swiss-Prot curators in their daily annotation work. After over 7 year activities, this group has gathered a significant amount of experience about the need in text mining for biocuration.The first observation we made is that there is no “in-a-box” solution which can satisfy every needs. Each curator has his/her own strategy to find information from the literature and none of the existing information retrieval systems is able to compete with it, more for reason of habits than for reason of performance. Second observation: to be completely operative, an information retrieval system should be embedded in the annotation platform. For instance, it should be possible to copy/paste information, such as the article reference or some interesting sentences, directly in the database format. Most of the existing online programs are hardly adaptable for this task and their use usually results in additional editing efforts for the curators. From this observation, we can derive the fact that integrating text mining services is usually more costly than expected since wrappers and user interfaces need significant developments sometimes fairly user-specific.After noticing these problems in the design and use of a generic information retrieval system for the Swiss-Prot curators, we focused our effort on text mining applications for database update. The follow-up of the literature is essential in the process of database maintenance and there are needs for automatic information extraction tools on a large panel of topics. We developed several IE applications in the field of:- PTM information (phosphorylation, glycosylation, disulfide bridge)- Subcellular localization- Variant/mutation detection and characterization- New sequence with enzymatic activities- New characterization of enzymes.These tools are integrated into pipelines which follow PubMed daily outcomes and generate list of selected abstracts with highlights on the relevant sentences. These procedures are done independently of the usual annotation workflow, so that curators can mine these preselected data whenever they work on database entry updates.To conclude, we have identified big challenges in text mining services after discussion with the curators. One of them is the detection of novel information, especially those related to a new function or a new characterization of a protein or one of its close homologues. We are currently working on this task in the framework of the collaborative project “EAGL”. Another challenge is definitely the large-scale screening of newly published full-text papers to complement the often incomplete information in abstracts. This becomes more and more indispensable, not really for the annotation of widely studied “hot” proteins, but to find new data on uncharacterized ones. For instance, when no gene name has been attributed to a sequence, the only way to retrieve information is to use the orf names, which are never provided in abstracts.Finally, one should definitely stress that many of these information retrieval and extraction tasks could be greatly simplified with the requirement of metadata at the article submission time, such as an official HGNC gene name or a UniProt reference.

The Weaknesses of Full-Text Searching

This paper provides a theoretical critique of the deficiencies of full-text searching in academic library databases. Because full-text searching relies on matching words in a search query with words in online resources, it is an inefficient method of finding information in a database. This matching fails to retrieve synonyms, and it also retrieves unwanted homonyms. Numerous other problems also make full-text searching an ineffective information retrieval tool. Academic libraries purchase and subscribe to numerous proprietary databases, many of which rely on full-text searching for access and discovery. An understanding of the weaknesses of full-text searching is needed to evaluate the search and discovery capabilities of academic library databases.

Identifying clusters of user behavior in intranet search engine log files

AbstractWhen studying how ordinary Web users interact with Web search engines, researchers tend to either treat the users as a homogeneous group or group them according to search experience. Neither approach is sufficient, we argue, to capture the variety in behavior that is known to exist among searchers. By applying automatic clustering technique based on self‐organizing maps to search engine log files from a corporate intranet, we show that users can be usefully separated into distinguishable segments based on their actual search behavior. Based on these segments, future tools for information seeking and retrieval can be targeted to specific segments rather than just made to fit the “the average user.” The exact number of clusters, and to some extent their characteristics, can be expected to vary between intranets, but our results indicate that some more generic groups may exist. In our study, a large group of users appeared to be “fact seekers” who would benefit from higher precision, a smaller group of users were more holistically oriented and would likely benefit from higher recall, and a third category of users seemed to constitute the knowledgeable users. These three groups may raise different design implications for search‐tool developers.

Classification models for the prediction of clinicians’ information needs

Objective: Clinicians face numerous information needs during patient care activities and most of these needs are not met. Infobuttons are information retrieval tools that help clinicians to fulfill their information needs by providing links to on-line health information resources from within an electronic medical record (EMR) system. The aim of this study was to produce classification models based on medication infobutton usage data to predict the medication-related content topics (e.g., dose, adverse effects, drug interactions, patient education) that a clinician is most likely to choose while entering medication orders in a particular clinical context. Design: We prepared a dataset with 3078 infobutton sessions and 26 attributes describing characteristics of the user, the medication, and the patient. In these sessions, users selected one out of eight content topics. Automatic attribute selection methods were then applied to the dataset to eliminate redundant and useless attributes. The reduced dataset was used to produce nine classification models from a set of state-of-the-art machine learning algorithms. Finally, the performance of the models was measured and compared. Measurements: Area under the ROC curve (AUC) and agreement (kappa) between the content topics predicted by the models and those chosen by clinicians in each infobutton session. Results: The performance of the models ranged from 0.49 to 0.56 (kappa). The AUC of the best model ranged from 0.73 to 0.99. The best performance was achieved when predicting choice of the adult dose, pediatric dose, patient education, and pregnancy category content topics. Conclusion: The results suggest that classification models based on infobutton usage data are a promising method for the prediction of content topics that a clinician would choose to answer patient care questions while using an EMR system.

Not Just Another Pretty Interface: Expert Information Searching and Retrieval

ABSTRACT Hospital librarians' practice includes the art and science of preparation for the search, use of complex tools for information retrieval, the search process, and the value-added delivery of results that completes information retrieval. They do focused searching for specific published information for themselves (for instance for evidence-based librarianship or collection development) or as intermediary searchers for their clients. Library schools' full semester courses in this subject usually are electives, so this article is an introduction for some and a review for others.

Topological multi-contour decomposition for image analysis and image retrieval

Successful classification, information retrieval and image analysis tools are intimately related with the quality of the features employed in the process. Pixel intensities, color, texture and shape are, generally, the basis from which most of the features are computed and used in such fields. This papers presents a novel shape-based feature extraction approach where an image is decomposed into multiple contours, and further characterized by Fourier descriptors. Unlike traditional approaches we make use of topological knowledge to generate well-defined closed contours, which are efficient signatures for image retrieval. The method has been evaluated in the CBIR context and image analysis. The results have shown that the multi-contour decomposition, as opposed to a single shape information, introduced a significant improvement in the discrimination power.

Bridging the Communication Gap at NASA - A Case Study in Communities of Practice

Following the loss of NASA's Space Shuttle Columbia in 2003, it was determined that problems in the agency's organization created an environment that led to the accident. One component of the proposed solution resulted in the formation of the NASA Engineering Network (NEN), a suite of information retrieval and knowledge-sharing tools. This paper describes the implementation of communities of practice, which are formed along engineering disciplines. Communities of practice enable engineers to leverage their knowledge and best practices to collaborate and take information learning back to their jobs and embed it into the procedures of the agency. This case study offers insight into using traditional engineering disciplines for virtual collaboration, including lessons learned during the creation and establishment of NASA's communities.

Controlled user evaluations of information visualization interfaces for text retrieval: Literature review and meta‐analysis

AbstractThis review describes experimental designs (users, search tasks, measures, etc.) used by 31 controlled user studies of information visualization (IV) tools for textual information retrieval (IR) and a meta‐analysis of the reported statistical effects. Comparable experimental designs allow research designers to compare their results with other reports, and support the development of experimentally verified design guidelines concerning which IV techniques are better suited to which types of IR tasks. The studies generally use a within‐subject design with 15 or more undergraduate students performing browsing to known‐item tasks on sets of at least 1,000 full‐text articles or Web pages on topics of general interest/news. Results of the meta‐analysis (N = 8) showed no significant effects of the IV tool as compared with a text‐only equivalent, but the set shows great variability suggesting an inadequate basis of comparison. Experimental design recommendations are provided which would support comparison of existing IV tools for IR usability testing.

The role of communication in collaborative information searching

AbstractCollaboration plays an important role in the information seeking and retrieval activities within a team setting. In this research, we examined the impact of collaborative design features in two information retrieval tools that explicitly support collaboration. We designed the two collaborative information searching prototype, MUSE (Multi‐User Search Engine) and MUST (Multi‐User Search and Talk) and evaluated both systems. Results indicate that the communication (i.e., chat) function played an important role in enhancing the information seeking process by establishing common ground among group members. We also identified unexpected challenges that arose as the prototypes were used during these activities. These challenges were both technical and social in nature. We discuss implications for system design and directions for future research.

A conceptual framework for improving information retrieval in folksonomy using Library of Congress subject headings

AbstractSocial tagging or collaborative tagging is a new trend in the organization, management, and discovery of digital information. Folksonomy, a collection of tags assigned to online information by all participants of social tagging systems, has many advantages as an information organization tool. As an information retrieval tool, however, folksonomy lacks the capability of discerning the relevancy of information. We propose a conceptual framework of IR for folksonomy on the basis of Library of Congress Subject Headings. Algorithms for constructing the conceptual framework are described, and values of the framework are also discussed.

I-FGM as a Real Time Information Retrieval Tool for E-Governance

Homeland security and disaster relief are some of the critical areas of E-governance that have to deal with vast amounts of dynamic heterogeneous data. Providing rapid real-time search capabilities for such applications is a challenge. Intelligent Foraging, Gathering, and Matching (I-FGM) is an established framework developed to assist users to find information quickly and effectively by incrementally collecting, processing and matching information nuggets. This framework has been successfully used to develop a distributed, unstructured text retrieval application. In this paper, we apply the I-FGM framework to image collections by using a concept-based image retrieval method. We approach this by incrementally processing images, extracting low-level features and mapping them to higher level concepts. Our empirical evaluation shows that our approach performs competitively compared to some existing approaches in terms of retrieving relevant images while offering the speed advantages of distributed and incremental process and unified framework between text and images.

Manipulating search engine algorithms: the case of Google

PurposeTo investigate how search engine users manipulate the rankings of search results. Search engines employ different ranking methods in order to display the “best” results first. One of the ranking methods is PageRank, where the number of links pointing to the page influences its rank. The “anchor text,” the clickable text of the hypertext link is another “ingredient” in the ranking method. There are a number of cases where the public challenged the Google's ranking, by creating a so‐called “Google bomb” – creating links to pages they wanted to be highly ranked for given query. Google is chosen as the search engine, because it is currently by far the most popular search engine.Design/methodology/approachPageRank, one of the major parameters of Google's ranking algorithm is described, and the author explains how this algorithm is exploited by communities of users to promote a certain web page for a specific query. This process is called “Google bombing.” Recent reaction of Google to this phenomenon is also described.FindingsSpecific examples of “accomplished Google bombs” show that the public is able to manipulate search results.Originality/valueGoogle, instead of being an unobtrusive information retrieval tool has become highly influential in the web scenery. Some users pay for search engine optimization, while others utilize the power of the crowd to influence Google's rankings. This paper supports the claims of Introna and Nissenbaum regarding the power of search engines.

GeneCite: A Stand-alone Open Source Tool for High-Throughput Literature and Pathway Mining

Systematic extraction of relevant biological facts from available massive scientific knowledge source is emerging as a significant task for the science community. Its success depends on several key factors, including the precision of a given search, the time of its accomplishment, and the communicative prowess of the mined information to the users. GeneCite - a stand-alone Java-based high-throughput data mining tool - is designed to carry out these tasks for several important knowledge sources simultaneously, allowing the users to integrate the results and interpret biological significance in a time-efficient manner. GeneCite provides an integrated high-throughput search platform serving as an information retrieval (IR) tool for probing online literature database (PubMed) and the sequence-tagged sites' database (UniSTS), respectively. It also operates as a data retrieval (DR) tool to mine an archive of biological pathways integrated into the software itself. Furthermore, GeneCite supports a retrieved data management system (DMS) showcasing the final output in a spread-sheet format. Each cell of the output file holds a real-time connection (hyperlink) to the given online archive reachable at the users' convenience. The software is free and currently available online www.bioinformatics.org; www.wrair.army.mil/Resources.

Catalogue use by the Petherick Readers of the National Library of Australia

An online questionnaire survey was distributed amongst the Petherick Readers of the National Library of Australia, a user group of scholars and researchers. The survey asked questions about the readers’ use and appreciation of the NLA catalogue. This group of users clearly appreciated the library catalogue and demonstrated that there are still library patrons who make extensive use of the sophisticated bibliographic access to a collection that it provides, despite the popularity of other information retrieval tools such as Web search engines.

Ontologies and reasoning techniques for (legal) intelligent information retrieval systems

An application of Narrative Knowledge Representation Language (NKRL) techniques on (declassified) 'terrorism in Southern Philippines' documents has been carried out in the context of the IST Parmenides project. This paper describes some aspects of this work: it is our belief, in fact, that the Knowledge Representation techniques and the Intelligent Information Retrieval tools used in this experiment can be of some interest also in an 'Ontological Modelling of Legal Events and Legal Reasoning' context.

Low cost, cross-language and cross-platform Information Retrieval and Documentation tools

In this paper we focus on the design and implementation of low cost, cross language and cross platform Information Retrieval and Documentation tools capable for the collection, organization and administration of unstructured and semi-structured information imported from various sources. A modular Computer Assisted Information Resources Navigation (CAIRN) software architecture is proposed and the requirements of each module are presented. A discussion of the implementation is based on the experimentation with a prototype of such a software tool. The technologies that are incorporated into the modern operating systems and the opportunities that they offer for implementing the modules of the CAIRN architecture are also examined and evaluated. Some of these technologies are common / independent from the operating systems, while some others are distinctive. In this latter case we face barriers (restrictions) for a straightforward implementation of the CAIRN software systems to the whole range of desktop operating systems (e.g. Windows, Mac OS, Linux, Solaris). Some alternative technologies are presented to avoid this serious constraint. The evaluation of the implementation effort is also discussed and eventually some conclusions and future plans for further improvement of the CAIRN architecture are given.

Lenguajes documentales y ontologías

The main points of convergence and divergence between controlled indexing languages and ontologies, tools for knowledge organization and information retrieval in the semantic web, are described. Fundamental aspects of ontologies are presented, as well as the basic semantic and structural characteristics of traditional controlled languages.

LIFT – A Legacy InFormation Retrieval Tool

Nowadays software systems are essential to the environment of most organizations, and their maintenance is a key point to support business dynamics. Thus, reverse engineering legacy systems for knowledge reuse has become a major concern in software industry. This article, based on a survey about reverse engineering tools, discusses a set of functional and non- functional requirements for an effective tool for reverse engineering, and observes that current tools only partly support these requirements. In addition, we define new requirements, based on our group's experience and industry feedback, and present the architecture and implementation of LIFT: a Legacy InFormation retrieval Tool, developed based on these demands. Furthermore, we discuss the compliance of LIFT with the defined requirements. Finally, we applied the LIFT in a reverse engineering project of a 210KLOC NATURAL/ADABAS system of a financial institution and analyzed its effectiveness and scalability, comparing data with previous similar projects performed by the same institution.

Archival Description, Information Retrieval, and the Construction of Thesauri in Israeli Archives

The introduction of computers, beginning in the late 1960s, for use in the description of textual archival materials has posed new challenges for archivists in general, and Israeli archivists in particular. Today, Israeli archives face several issues: acceptance of international standards and their adaptation to the conditions and needs of Israeli archival data bases; creation of tools for information retrieval, particularly a comprehensive thesaurus to be used by all archival institutions or a basic thesaurus which can be developed in accordance with the needs of individual archives. The article discusses the standardization that has begun in archival description outside of Israel, and various proposed solutions for information retrieval from meta-data or the content of the documents by means of thesaurus-based indexing. Two unique projects, representing different outlooks, are described: the thesaurus of the National Insurance Institute of Israel, partially based upon that of the Henrietta Szold Institute, and that of the archives of Religious Zionism at Bar-Ilan University, created ex nihilo on the basis of the meta-data and the documentation itself. In addition, a comparison is conducted of elements of ISAD(G) adopted by the Israel State Archives and the Archives of Religious Zionism. The article concludes with a proposal for the creation of an archival description and information retrieval system based on ISAD(G) and ISAAR(CPF), and a general thesaurus suitable for all Israeli archives.

The role of academic libraries in facilitating institutional transformation programmes: the case of two comstituent colleges of the university of Dar es Salaam

This paper discusses the implications of the University of Dar es Salaam Institutional Transformation Programme (ITP) on information services provision at Muhimbili University College of Health Sciences (MUCHS) and University College of Lands and Architectural Studies (UCLAS) libraries. The paper is based on a Master's dissertation whose objective was to determine the extent to which the two College libraries (MUCHS and UCLAS) were equipped to meet the challenges of Institutional Transformation Programme - UDSM 2000. Questionnaires and interview guides were used to collect data. The findings of the study established that ITP has raised new demands and challenges in terms of information services provision, including ICT facilities, information retrieval tools and the changing role of librarians. The study notes that the two libraries were not adequately equipped to meet the challenges. Problems noted include technological and resource based ones. The study recommends the colleges to clearly understand the implications of ITP to their libraries and give the necessary support. They should ensure that the library management is empowered to make decisions and implement their programmes on time. University of Dar es Salaam Library Journal Vol. 7(2) 2005: 84-97