Web-based affiliation matching

SSRN Electronic Journal, 2014

This paper presents a text matching process for identification and correct assignment of scholarly publications, extracted from publication lists provided by authors or research institutes, in large bibliographic databases such as Thomson Reuters' Web of Science (WoS). An identification method is implemented by means of overlapping common 3-grams and the results are obtained from the match of the two sources according to the highest score of the applied cosine measure. Levenshtein similarities based on N-grams have been used to measure the closeness between the given CV publication and the retrieved best possible WoS match as a complementary and confirmatory measure. It is shown that the suggested method has an important potential on reducing the manual effort to find out whether a desired publication is indexed in WoS or not. The similarity scores derived by Levenshtein measure show consistency with those derived from Salton's similarity measure. Incorrect matches are examined in depth and possible thresholds are suggested to decrease the effort for manual cleaning.

Information Processing & Management, 2004

A common task in both Webmetrics and Web information retrieval is to identify a set of Web pages or sites that are similar in content. In this paper we assess the extent to which links, colinks and couplings can be used to identify similar Web sites. As an experiment, a random sample of 500 pairs of domains from the UK academic Web were taken and human assessments of site similarity, based upon content type, were compared against ratings for the three concepts. The results show that using a combination of all three gives the highest probability of identifying similar sites, but surprisingly this was only a marginal improvement over using links alone. Another unexpected result was that high values for either colink counts or couplings were associated with only a small increased likelihood of similarity. The principal advantage of using couplings and colinks was found to be greater coverage in terms of a much larger number of pairs of sites being connected by these measures, instead of increased probability of similarity. In information retrieval terminology, this is improved recall rather than improved precision.

In this paper, we describe a system to collect information about academic affiliation (organisations where researchers work) from Calls-for-Papers for academic conferences. The system uses a range of heuristic approaches and open-source tools in order to extract and identify entities, and to incorporate the information into a pre-defined database schema. This forms part of a larger project to automatically populate and maintain a range of data related to academic research. The proposed system is currently being tested and some promising preliminary results are available.

2008

Today, bibliographic digital libraries play an important role in helping members of academic community search for novel research. In particular, author disambiguation for citations is a major problem during the data integration and cleaning process, since author names are usually very ambiguous. For solving this problem, we proposed two kinds of correlations between citations, namely, Topic Correlation and Web Correlation, to exploit relationships between citations, in order to identify whether two citations with the same author name refer to the same individual.The topic correlation measures the similarity between research topics of two citations; while the Web correlation measures the number of co-occurrence in web pages. We employ a pair-wise grouping algorithm to group citations into clusters. The results of experiments show that the disambiguation accuracy has great improvement when using topic correlation and Web correlation, and Web correlation provides stronger evidences about the authors of citations.

Proceedings of 21th Scientific Conference “Scientific Services & Internet – 2019”

The cross-lingual matching of authors and publications is a special case of the task of assigning a unique identifier to the same real-world entity in multilingual data sources. This paper presents the results of experiments with the several versions of a cross-lingual system designed to match, basing on a Russian-language data source, the authors and English-language publications. Since different heuristics have been tested in these versions of the system, we consider here only those that have given the best results. An important element of the system is its interactive visualization tool, which gives information on the distribution of publications by authors, as well as providing the ability to edit the results of the analysis. The visualization system is supplemented with methods for similarity matrices ordering. Experiments have shown that the main source of improving the quality of the matching and clustering algorithm is extending the set of confirmed publications. The approaches used in this system are applicable to solving the problem of linking named entities in various multilingual data sources.

In this paper, we report the results of a bibliometric study in which we track cross-disciplinary citation behavior in the sciences. We hypothesize that cross-disciplinary citation in the sciences increased over the time period 1985–2000. Unlike most previous studies in this area, we assign discipline to a paper by its first author’s affiliation, and we hypothesize that assigning papers to disciplines based on first-author affiliation would yield results consistent with previous findings on cross-disciplinary citation rates in the sciences. Using the output of scientists in Biological Sciences, Chemistry, and Physics departments at 12 large research universities in 1985, 1990, 1995, and 2000 as our data set, we measure the cross-disciplinary citation rates of each discipline and compare our results to the findings of previous studies in this area.

Common Map of Academia (COMAC) is a system that keeps information about scientists and publications. It uses bibliographical information as a source of knowledge. In the article we show briefly this system. We present how we used that system for parts of Polish national CRIS: Polish national bibliography and newly created PolIndex-polish citation index. We present algorithms used for enhancing CRIS data: document deduplication, citations matching and authors identification. Finally we show performance statistics.

Scientometrics, 2014

Academic effectiveness of universities is measured with the number of publications and citations. However, accessing all the publications of a university reveals a challenge related to the mistakes and standardization problems in citation indexes. The main aim of this study is to seek a solution for the unstandardized addresses and publication loss of universities with regard to this problem. To achieve this, all Turkey-addressed publications published between 1928 and 2009 were analyzed and evaluated deeply. The results show that the main mistakes are based on character or spelling, indexing and translation errors. Mentioned errors effect international visibility of universities negatively, make bibliometric studies based on affiliations unreliable and reveal incorrect university rankings. To inhibit these negative effects, an algorithm was created with finite state technique by using Nooj Transducer. Frequently used 47 different affiliation variations for Hacettepe University apart from “Hacettepe Univ” and “Univ Hacettepe” were determined by the help of finite state grammar graphs. In conclusion, this study presents some reasons of the inconsistencies for university rankings. It is suggested that, mistakes and standardization issues should be considered by librarians, authors, editors, policy makers and managers to be able to solve these problems.

IEEE Data(base) Engineering Bulletin, 2005

Existing publication similarity measures, needed for example-based searching, fall into two classes, namely, text-based similarity measures from Information Retrieval, and citation-based similarity measures based on bibliographic coupling and/or co-citation.

2014 IEEE International Conference on Big Data (Big Data), 2014

The international "Digging Into Data Initiative" is exploring the uses and approaches for large-scale data analysis and processing for a variety of purposes primarily in the Humanities and Social Sciences. This paper discusses the "Integrating Data Mining and Data Management Technologies for Scholarly Inquiry" project being conducted at UC Berkeley, University of North Carolina, and the University of Liverpool. Our "big data" consists of the entire texts collection of the Internet Archive (approximately 3.6 million volumes) and the entire JSTOR database. We are performing surface-level natural language processing on this data to identify noun phrases and further refinements to identify personal, corporate, and geographic names. We are then using resources including library and archival authority records to identify variants and merge names. The goal is an integrated index of persons, places, and organizations and where they have been referenced in our collections.