Web of Science

Summary

The Thomson Reuters database covers over 16,000 international journals, books and proceedings in the sciences, social sciences and arts and humanities. The 9,200 international journals covered on an annual basis in the Web of Science are an important part of this data. The content found within the platform is multidisciplinary, ensuring that scholars are not restricted along subject-specific lines. This becomes particularly important as research becomes more and more interdisciplinary in nature. Web of Science covers over 250 categories in every area of the sciences, social sciences and arts & humanities.

The Web of Science provides seamless access to current and retrospective (back to 1900 in the Sciences) multidisciplinary information. Web of Science also provides a unique search method, cited reference searching. With it, users can navigate forward, backward and through the literature, searching all disciplines and time spans to uncover all the information relevant to their research. Thomson Reuters remains committed to providing comprehensive coverage of the world's most important and influential journals for its subscribers' current awareness and retrospective information retrieval needs.

The Web of Science is designed to be the comprehensive, information-gathering resource for "deep-mining" during the research process – a place for the researcher whose primary goal is to track the development of an idea, issue, or technology to gain background in a particular area. The goal is to allow the researcher to search and navigate the citation pathways to unearth all the relevant information. Web of Science provides the content, features and functionality to support these long-range retrospective and analytical research activities.

Inclusion Policy

The goal of Thomson Reauters and formerly ISI has always been to provide evaluated content for researchers and scholars. Journal evaluation and selection is conducted on an ongoing basis with journals added to and deleted from the database as frequently as every two weeks. Our selection criteria are time tested and unbiased and have been in use for over 40 years. The Thomson Reuters editorial staff reviews nearly 2,200 new journal titles annually, but only 10-12% of the journals evaluated are selected. Moreover, existing journal coverage in ISI Web of Knowledge products is also constantly under review. Journals now covered are monitored to ensure that they are maintaining high standards and a clear relevance to the ISI products in which they are covered.

Each journal goes through an extensive evaluation process before being selected or rejected. Our editors performing journal evaluations have educational backgrounds relevant to their areas of responsibility as well as experience and education in information science. Their knowledge of the literature of their field is extended by consultation with established networks of advisors who participate in the evaluation process when needed.

Many factors are taken into account when evaluating journals for coverage, ranging from the qualitative to the quantitative. The journal's basic publishing standards, its editorial content, the international diversity of its authorship and the citation data associated with it are all considered. No one factor is considered in isolation, but by combining and inter-relating the data, the editor is able to determine the journal's overall strengths and weaknesses.

Every journal is evaluated for the scope and content of its coverage and to determine exactly how our data capture staff will index the content to ensure consistency of indexing and coding of all types of articles. Each journal is indexed cover to cover and for every title we build an authority file to indicate exact indexing policy to handle all special articles and sections within each title.

Geographical Coverage

UK. 21%
Europe. 27%
World. 52%

Geographic representation of a journal is another consideration in the selection of titles for Web of Science. To meet the needs of its international subscriber base, Thomson Reuters covers journals with international diversity among authors of both source articles and cited articles. To properly reflect the global context in which scientific research takes place and to provide balanced coverage in each category, Thomson Reuters covers the best regional journals published in the world. Our editors consider regional journals in terms of those journals in the category from the same geographic area. High journal publishing standards, especially timeliness and English language bibliographic elements, remain essential.

Gap Fill Policy

Thomson Reuters has a committed claiming department dedicated to two principles, keeping current and complete. The team is part of the Publisher Relations department that manages our relations with over 2,500 publishers worldwide.

Claiming is triggered by two automated processes, a timer indicating an issue expected within a certain frequency has not arrived, and a "gap", the receipt for example of issue 9 after issue 7. These claims are then routed based on journal receipt, electronic or print. Electronic journals are sent directly to a processing department with access to download the material to expedite the acquisition. Print journals are requested by email where possible, or fax where no email is available. Follow up letters are sent in cases where no acknowledgement has been received.

Claiming is not a guaranteed process. Requested material is not always sent. At least 3 requests will be sent for an issue to be claimed (unless it is received in the interim). At times our customers will report gaps in coverage and to facilitate these instances where we could not obtain the material but customers might be able to assist, we have a standing policy that any missing issue provided will be indexed.

For all legacy cumulative indexes and the newly created Century of Science 1900-1944 data on Web of Science, every effort is made to fill in gaps for the journal for the specified span (for example, 1945-54 and 1900-44) during the production period for that edition. For all legacy print cumulative editions (science, social sciences and arts & humanities), all journals added during the cumulative edition time span were acquired and processed before the edition was produced. These issues are contained in Web of Science.

Archive Policy

Generally, we do not add back to volume 1 if a newly selected journal has been published for more than a year or two.

Embargo Rules

None.

Update Frequency

Weekly.

Content Classification

The subject categories in Web of Science have been established over time by the editors responsible for the various subject areas of the database. The process of identifying subject categories and populating them with journals has been ongoing since the establishment of the Science Citation Index, the Social Sciences Citation Index and the Arts & Humanities Citation. All articles covered in a journal are tagged with the categories on the journal level. For example, the journal "Science" is categorized under "Multidisciplinary Sciences," so all the articles covered in that journal, no matter the subject, are tagged "Multidisciplinary Sciences." There are over 250 different categories in the sciences, social sciences and arts & humanities. Each category is reviewed yearly for changes and additions and to create new categories for evolving areas of research.

Categories reflect the overall content of the journals collected in them. The Thomson Reuters category system is not hierarchical. Journals may be in multiple categories to aid in searching and categorization of multidisciplinary material. The editorial structure of all categories is under continual review and has evolved gradually encompassing every field of study in the natural sciences and physical sciences, the social sciences and the arts & humanities. New categories are formed when a critical mass of journals on specific topics begin to publish. Categories are under continuous study to determine the need for re-categorization of journals and/or the emergence of distinctly new categories.

The Editorial Development staff deal with journals in their specific field as defined by the subject scope of their particular discipline. They are very familiar with the scope of each of the categories in their area. After thorough examination of the editorial content of the journal under evaluation, the editor determines which category is most appropriate. Along with observable topical relevance, citation relevance is also used as another tool used in category assignment. The editor looks at the journals citing the journal under evaluation and uses this data to calculate statistical relevance to a category. This method is often used in making decisions about multiple categorizations. Closely related fields, such as 'Neurosciences' and 'Clinical Neurology', and 'Cardiac & Cardiovascular Systems' and 'Peripheral Vascular Disease' have significant overlaps. The rationale for these assignments is based on citation relevance and on topical relevance.

The scope notes for each of the categories are available online.

Keywords and Indexes

Our Data Capture system is an image based production system that allows concurrent indexing, article document coding, and the processing of all data elements with exact standards using a system of online thesauri and authority files constructed from over 40 years of data and incorporating standard reference works.

All journals are reviewed to build a detailed authority file that indicates all processing rules and each file has links to standard dictionaries and files for processing that particular data element. All items within a journal are coded based on a detailed review of the journal by our Bibliographic Policy Editors. The codes incorporate all types of articles based on subject areas, including detailed codes for all types of dance, music and film performance reviews, poetry and prose and original works of literature for the arts and humanities, as well as hardware, software, and database reviews for the sciences and book reviews for the social sciences as well as the humanities.

The Data Production system has authority files for capturing all author names, keywords and special symbols in abstracts. From 1991 forward, articles may be tagged with author-provided keywords and abstracts. Keywords and abstracts are processed in full with no limit to number and size of field.

Articles are also tagged with Keywords Plus, which are words or phrases that frequently appear in the titles of an article's references, but do not necessarily appear in the title of the article or in a list of author keywords.

Authority Files

For over 40 years Thomson Reuters has ensured high quality data capture through extensive bibliographic policy standards. Thomson Reuters has always maintained strict rules of standardization, utilizing a unique, algorithms and dictionaries built from our long history of processing standards of all bibliographic data fields.

All author names, author affiliation names, keywords, abstracts and cited references are processed in full without limit. They are ALL searchable and displayed in all records. There are authority files for all data elements.

Journal titles are indexed by ISO standard and full title. Our accessioning system is built to handle title changes, mergers and splits, and we adhere to all publishing standards. Each title has a pre-constructed authority file that follows it through all phases of the production cycle. The authority file is built for all new titles and also for all titles which have changed significantly. Our production system has an online transaction that permits checking the authority file at all phases of the production cycle to ensure consistency of indexing.

Author names are indexed by Last Name – and Initial(s). There are rules for processing particle names, European and Asian names and for processing non-English characters (all alphabets). The authority files built for each title include a link to all processing rules. We have also built in an automatic policy check for all names. In April 2006 Thomson Reuters began processing full author names (first, middle and surname) for all authors. This was added to Web of Science in late 2006.

Author affiliations are processed using a series of authority files constructed using our own historical data and also data from several standard resources. These systems are continually reviewed for all changes in institution names and geographic location. There are also post-processing checks made for all author affiliation elements: the main organization name, sub-organization name(s), city, state, zone numbers and countries.

Abstracts are processed in full using authority files for special symbols, mathematical and chemical formulae to present the abstract to preserve as closely as possible the original presentation in the title. Changes are made to increase search retrieval for all symbols and formulae.

Cited references, a total of 37 million annually (over 650 million in the entire database), are matched against a comprehensive cited reference checking system, unified and processed for entry into the Web of Science. Thomson Reuters has always maintained strict rules of standardization, utilizing a unique, algorithmically-created cluster key to represent each document. In the absence of some form of unification, citations that have some variant form of a journal title could be passed over or misattributed, thus affecting the reported citation frequency of the journal and hence any subsequent perception of the journal's influence. The retrieval of citations associated with a given journal title and its associated abbreviation has direct bearing on retrieval of information for research, and it also impacts the ranking that journal may receive within its category.

Citation Linking

Thomson Reuters has built an extensive linking infrastructure designed to address the issues of authentication (such as IP address, token generator or cookie technology, logon/password and trusted server systems), routing and security access within a robust environment. The result is a single management system for the many types of links within the ISI Web of Knowledge platform, and in particular the Web of Science:

Unique Cited Reference Navigation and Searching including authoritative indexes for both Cited Authors and Cited Works, allowing the user to scan and select the author and work to ensure accuracy and precision searching across the 37 million cited references processed annually.
Intra-content links (such as from a journal article record to its "Times Cited" list, or from a journal article record to a journal table-of-contents page).
Inter-content links (such as from a journal article record to a chemical reaction drawing, or from a patent record to a cited journal article).
Publisher full-text links (whether hosted locally or by the publisher). Over 10,800 instances of journals (Publishers, Aggregators, Society Publishers and Subscription Agents).
SFX and other context-sensitive links.
OPAC links – configured based on user preference (all OpenURL resolvers supported).
Protein and gene sequence database links – Links to the NCBI databases.
Pay-per-view links – Publisher sites (Based on customer preference).
Content Enhancement Links for specific publishers (Nature Publishing) including direct links to Web of Science Full Bibliographic Records, Times Cited and Related Records.
Authenticated Links directly from Customer Research Gateways and Institutional Repositories to Web of Science Full Bibliographic Records, Times Cited and Related Records functions.

ISI Links is comprised of a central server where all links mechanisms are integrated, avoiding the costly and inefficient process of duplicating complex authentication, access and routing systems at the level of individual product offerings. The links system is designed as an extensible system that can handle all types of linking: from client-designed systems to all standard OPAC file structures, etc.

Full Text Linking

Linking to full text is available either through customized full text links from the summary or full record page or through an institution's OpenURL system. Currently, Web of Science provides access to over 16 million items from approximately 300 publishers

Thomson Reuters actively seeks full-text linking partnerships with primary journal publishers and content hosts for all available electronic journals covered by ISI Web of Knowledge. Currently, Thomson Reuters links to over 7,800 unique full text titles and over 20,000 journal instances. These include a growing population of open access journals and retrospective archives.

Our Links Management system (ISI Links) is responsible for the many types of links within the ISI Web of Knowledge platform. ISI Links is comprised of a central server where all links mechanisms are integrated, avoiding the costly and inefficient process of duplicating complex authentication, access and routing systems at the level of individual product offerings.

It handles:

Intra-content links (such as from a journal article record to its "Times Cited" list, or from a journal article record to a journal table-of-contents page).
Inter-content links (such as from a journal article record to a chemical reaction drawing, or from a patent record to a cited journal article).
Publisher full-text links (whether hosted locally or by the publisher).
SFX and other context-sensitive links (it is can be configured for all systems and customized by client).
OPAC links – multiple OPAC links and links to Non-OPAC files can be configured.
Protein and gene sequence database links (NCBI links at NLM).
Pay-per-view links at Publishers.
Document Delivery Systems – our own system or customized to the preferred vendor.