arXiv subjects
Publications and source records attributed to Philipp Mayr.
In this paper we examine if hyperlink-based (webometric) indicators can be used to rank academic websites. Therefore we analyzed the interlinking structure of German university websites and compared our simple hyperlink-based ranking with official and web-independent rankings of universities. We found that link impact could not easily be seen as a prestige factor for universities.
The paper introduces scholarly Information Retrieval (IR) as a further dimension that should be considered in the science modeling debate. The IR use case is seen as a validation model of the adequacy of science models in representing and predicting structure and dynamics in science. Particular conceptualizations of scholarly activity and structures in science are used as value-added search services to improve retrieval quality: a co-word model depicting the cognitive structure of a field (used for query expansion), the Bradford law of information concentration, and a model of co-authorship networks (both used for re-ranking search results). An evaluation of the retrieval quality when science model driven services are used turned out that the models proposed actually provide beneficial effects to retrieval quality. From an IR perspective, the models studied are therefore verified as expressive conceptualizations of central phenomena in science. Thus, it could be shown that the IR perspective can significantly contribute to a better understanding of scholarly structures and activities.
This paper is about a better understanding on the structure and dynamics of science and the usage of these insights for compensating the typical problems that arises in metadata-driven Digital Libraries. Three science model driven retrieval services are presented: co-word analysis based query expansion, re-ranking via Bradfordizing and author centrality. The services are evaluated with relevance assessments from which two important implications emerge: (1) precision values of the retrieval service are the same or better than the tf-idf retrieval baseline and (2) each service retrieved a disjoint set of documents. The different services each favor quite other - but still relevant - documents than pure term-frequency based rankings. The proposed models and derived retrieval services therefore open up new viewpoints on the scientific knowledge space and provide an alternative framework to structure scholarly information systems.
The paper proposes three different kinds of science models as value-added services that are integrated in the retrieval process to enhance retrieval quality. The paper discusses the approaches Search Term Recommendation, Bradfordizing and Author Centrality on a general level and addresses implementation issues of the models within a real-life retrieval environment.
This paper is about an information retrieval evaluation on three different retrieval-supporting services. All three services were designed to compensate typical problems that arise in metadata-driven Digital Libraries, which are not adequately handled by a simple tf-idf based retrieval. The services are: (1) a co-word analysis based query expansion mechanism and re-ranking via (2) Bradfordizing and (3) author centrality. The services are evaluated with relevance assessments conducted by 73 information science students. Since the students are neither information professionals nor domain experts the question of inter-rater agreement is taken into consideration. Two important implications emerge: (1) the inter-rater agreement rates were mainly fair to moderate and (2) after a data-cleaning step which erased the assessments with poor agreement rates the evaluation data shows that the three retrieval services returned disjoint but still relevant result sets.
This case study proposes a scenario with three topic-related thesauri, which have been connected with bilateral cross-concordances as part of a major terminology mapping initiative in the project KoMoHe (Mayr & Petras, 2008). The thesauri have already been or will be converted to SKOS and in order to not omit the relevant crosswalks, the mapping properties of SKOS will be used for modeling them adequately.
This paper is a short description of an information retrieval system enhanced by three model driven retrieval services: (1) co-word analysis based query expansion, re-ranking via (2) Bradfordizing and (3) author centrality. The different services each favor quite other - but still relevant - documents than pure term-frequency based rankings. Each service can be interactively combined with each other to allow an iterative retrieval refinement.
The purpose of this paper is to apply and evaluate the bibliometric method Bradfordizing for information retrieval (IR) experiments. Bradfordizing is used for generating core document sets for subject-specific questions and to reorder result sets from distributed searches. The method will be applied and tested in a controlled scenario of scientific literature databases from social and political sciences, economics, psychology and medical science (SOLIS, SoLit, USB Koeln Opac, CSA Sociological Abstracts, World Affairs Online, Psyndex and Medline) and 164 standardized topics. An evaluation of the method and its effects is carried out in two laboratory-based information retrieval experiments (CLEF and KoMoHe) using a controlled document corpus and human relevance assessments. The results show that Bradfordizing is a very robust method for re-ranking the main document types (journal articles and monographs) in today's digital libraries (DL). The IR tests show that relevance distributions after re-ranking improve at a significant level if articles in the core are compared with articles in the succeeding zones. The items in the core are significantly more often assessed as relevant, than items in zone 2 (z2) or zone 3 (z3). The improvements between the zones are statistically significant based on the Wilcoxon signed-rank test and the paired T-Test.
Knowledge organization systems (KOS), like thesauri and other controlled vocabularies, are used to provide subject access to information systems across the web. Due to the heterogeneity of these systems, mapping between vocabularies becomes crucial for retrieving relevant information. However, mapping thesauri is a laborious task, and thus big efforts are being made to automate the mapping process. This paper examines two mapping approaches involving the agricultural thesaurus AGROVOC, one machine-created and one human created. We are addressing the basic question "What are the pros and cons of human and automatic mapping and how can they complement each other?" By pointing out the difficulties in specific cases or groups of cases and grouping the sample into simple and difficult types of mappings, we show the limitations of current automatic methods and come up with some basic recommendations on what approach to use when.
The paper reports about results on the GESIS-IZ project "Competence Center Modeling and Treatment of Semantic Heterogeneity" (KoMoHe). KoMoHe supervised a terminology mapping effort, in which 'cross-concordances' between major controlled vocabularies were organized, created and managed. In this paper we describe the establishment and implementation of cross-concordances for search in a digital library (DL).
The German Federal Ministry for Education and Research funded a major terminology mapping initiative, which found its conclusion in 2007. The task of this terminology mapping initiative was to organize, create and manage 'cross-concordances' between controlled vocabularies (thesauri, classification systems, subject heading lists) centred around the social sciences but quickly extending to other subject areas. 64 crosswalks with more than 500,000 relations were established. In the final phase of the project, a major evaluation effort to test and measure the effectiveness of the vocabulary mappings in an information system environment was conducted. The paper reports on the cross-concordance work and evaluation results.
The purpose of the paper is to propose models to reduce the semantic complexity in heterogeneous DLs. The aim is to introduce value-added services (treatment of term vagueness and document re-ranking) that gain a certain quality in DLs if they are combined with heterogeneity components established in the project "Competence Center Modeling and Treatment of Semantic Heterogeneity". Empirical observations show that freely formulated user terms and terms from controlled vocabularies are often not the same or match just by coincidence. Therefore, a value-added service will be developed which rephrases the natural language searcher terms into suggestions from the controlled vocabulary, the Search Term Recommender (STR). Two methods, which are derived from scientometrics and network analysis, will be implemented with the objective to re-rank result sets by the following structural properties: the ranking of the results by core journals (so-called Bradfordizing) and ranking by centrality of authors in co-authorship networks.
The paper discusses and analyzes the scientific search service Google Scholar (GS). The focus is on an exploratory study which investigates the coverage of scientific serials in GS. The study shows deficiencies in the coverage and up-to-dateness of the GS index. Furthermore, the study points up which Web servers are the most important data providers for this search service and which information sources are highly represented. We can show that there is a relatively large gap in Google Scholars coverage of German literature as well as weaknesses in the accessibility of Open Access content. Keywords: Search engines, Digital libraries, Worldwide Web, Serials, Electronic journals
Purpose: To provide a critical review of Bergman's 2001 study on the Deep Web. In addition, we bring a new concept into the discussion, the Academic Invisible Web (AIW). We define the Academic Invisible Web as consisting of all databases and collections relevant to academia but not searchable by the general-purpose internet search engines. Indexing this part of the Invisible Web is central to scientific search engines. We provide an overview of approaches followed thus far. Design/methodology/approach: Discussion of measures and calculations, estimation based on informetric laws. Literature review on approaches for uncovering information from the Invisible Web. Findings: Bergman's size estimate of the Invisible Web is highly questionable. We demonstrate some major errors in the conceptual design of the Bergman paper. A new (raw) size estimate is given. Research limitations/implications: The precision of our estimate is limited due to a small sample size and lack of reliable data. Practical implications: We can show that no single library alone will be able to index the Academic Invisible Web. We suggest collaboration to accomplish this task. Originality/value: Provides library managers and those interested in developing academic search engines with data on the size and attributes of the Academic Invisible Web.
In our study we analysed 3.889 records which were indexed in the Library and Information Science Abstracts (LISA) database in the research field of informetrics. We can show the core journals of the field via a Bradford (power law) distribution and corroborate on the basis of the restricted LISA data set that it was the appropriate time to found a new specialized journal dedicated to informetrics. According to Bradford's Law of scattering (pure quantitative calculation), Egghe's Journal of Informetrics (JOI) first issue to appear in 2007, comes most probable at the right time.
The ongoing paradigm change in the scholarly publication system ('science is turning to e-science') makes it necessary to construct alternative evaluation criteria/metrics which appropriately take into account the unique characteristics of electronic publications and other research output in digital formats. Today, major parts of scholarly Open Access (OA) publications and the self-archiving area are not well covered in the traditional citation and indexing databases. The growing share and importance of freely accessible research output demands new approaches/metrics for measuring and for evaluating of these new types of scientific publications. In this paper we propose a simple quantitative method which establishes indicators by measuring the access/download pattern of OA documents and other web entities of a single web server. The experimental indicators (search engine, backlink and direct access indicator) are constructed based on standard local web usage data. This new type of web-based indicator is developed to model the specific demand for better study/evaluation of the accessibility, visibility and interlinking of open accessible documents. We conclude that e-science will need new stable e-indicators.
This paper introduces Google Web APIs (Google APIs) as an instrument and playground for webometric studies. Several examples of Google APIs implementations are given. Our examples show that this Google Web Service can be used successfully for informetric Internet based studies albeit with some restrictions.