Text & Data Mining

What Is Text and Data Mining?

Large collections of texts are systematically searched and analyzed to identify unknown, overlapping patterns and relations (dependency chains) (e.g., in computational linguistics, biomedicine, or bioinformatics).

What Is the Problem with This?

This text and/or data extraction is based on the systematic duplication and storage of content that may be protected by copyright. For this reason, providers usually classify this practice as impermissible. Publishers monitor access frequency and will block access for individual IP addresses or even the entire UZH IP range in such cases. This applies even if the provider does not explicitly publish download limits on its website or in its contract terms.

What Are You Allowed to Do?

Text and data mining (TDM) often requires additional license-based access via an application programming interface (API). APIs provide TDM-capable infrastructures and machine-readable data formats. In addition to openly available data services, some can be used under the UZH-ZB licenses, sometimes with an individual API key.

Please review the list of APIs below, along with the corresponding legal and technical usage conditions, before starting any project. If a platform is not listed or if you are unsure whether access is permitted, please contact us early at emedia@ub.uzh.ch.

Cambridge University Press

License	✅ Included in the UZH-ZB license ✅ Non‑commercial research, teaching, and learning purposes only ✅ Full texts accessible within existing subscriptions
Content & formats	Cambridge Core full texts and metadata from journals, e-books, and most reference works (coverage may depend on the specific titles for content licensed via Zentralbibliothek) No REST API is provided: access via browser Formats: HTML, PDF (XML upon request)
Limitations	❌ bulk download is not permitted and monitored. ⚠️Excessive downloads may result in blocking access for all users. ⚠️For bulk TDM projects, see below.
Documentation	Text and Data Mining Policy and F&Q
How to request access	No API key is required. Contact e-medien@zb.uzh.ch Large-scale or bulk TDM projects (XML feed) require prior permission from the publisher. Contact e-medien@zb.uzh.ch with details to: Scope of content required, e.g., journals, books, collections Approximate volume and timeline Preferred file format, e.g., XML and/or PDF Whether you have an FTP server available

License	❌ As an aggregator, EBSCO integrates content from multiple publishers. For copyright reasons, text and data mining (TDM) is not permitted. ✅The API is intended for the integration of search, discovery, and bibliographic metadata functions from licensed EBSCOhost databases into local research and teaching applications.
Content & formats	Bibliographic metadata (title, authors, abstracts, subject terms), result lists Depending on the license, access to full texts Output formats: XML Access via REST or SOAP
Limitations	This API is not identical to the EDS API (EBSCO Discovery Service). API usage requires a local technical infrastructure for deploying and using the EBSCO Integration Toolkit (EIT). Not all EBSCOhost databases are API‑enabled (see list) Redistribution and storage of full texts are restricted and subject to copyright conditions (see copyright notice in the respective full text). Bulk downloads, analysis of large full‑text corpora, and training of AI models are not permitted.
Documentation	EBSCOhost API (EBSCO Connect) Making Requests with REST Making Requests with SOAP
How to request access	Requires an EBSCOhost / EIT profile. Please contact your library: emedia@ub.uzh.ch / e‑medien@zb.uzh.ch

License	❌ API usage is not included in the UZH-ZB license. Text and data mining is strictly prohibited.
Limitations	⚠️ Excessive downloads may result in blocking access for all users. Dow Jones does not publish numeric limits. Only download as many articles as you can read without automated tools.

License	✅ Included in the UZH-ZB license. Depending on TDM volume, costs may occur. ✅ Non-commercial research, teaching, and learning purposes only. ⚠️Using AI with licensed full texts (Journals)
Content & formats	Metadata Search REST API with abstracts (XML, JSON) Open Access API: OA full-text articles Full‑Text Access API: licensed full texts (PDF, XML) DOI Lookup API: metadata queries Dynamic query tool for basic searches; software development kits for PHP, Python 3, Python 2 and Java
Limitations	DOI Lookup API: max. 25 DOIs/query; no numerical rate limit, typical API throttling applies
Documentation	Availabe APIs & Use Cases Interactive Documentation
How to request access	API key required on registration. Requests must be made to emedia@ub.uzh.ch. Use is subject to acceptance of the terms of use. Please submit a project description and the planned data‑mining scope for provider approval. How To Get Started

License	✅ Included in the UZH-ZB license ✅ Non-commercial research, teaching, and learning purposes only. ⚠️Using AI with licensed full texts (Journals)
Content & formats	Direct requests using a DOI list: metadata (XML) are free of charge; full texts (XML: for more recent articles, excluding the Conference Series; PDF: available for most articles) are fee‑based. Data delivery via sFTP.
Limitations	Systematic downloading (scraping) will be blocked.
Documentation	TDM Policy
How to request access	Requests must be made to IOP with project details and required formats (PDF/XML). Please provide: name, email address, licensing institution (UZH), planned data-mining scope (list of DOIs, date ranges per journal). Contact: contentsupport@ioppublishing.org Cc. emedia@ub.uzh.ch.

License	Content to follow
Content & formats	Content to follow
Limitations	Content to follow
Documentation	Content to follow
How to request access	Content to follow

License	❌API usage is not included in the UZH-ZB license. See Specialized Databases (Law). Two versions are available: Swissdox Essentials (with legal restrictions) and Swissdox Professional (more extensive search capabilities). ✅For bulk downloads or TDM projects, the Swissdox@LiRI API developed by the Linguistic Research Infrastructure (LiRI) must be used. Non‑commercial research only. No redistribution to third parties. Local hosting of the data is required.
Content & formats	Swissdox@LiRI API provides query‑specific full texts subject to copyright restrictions. Most articles are in German and French, with fewer in Italian, Romansh, and English, primarily from the last 25 years (media coverage). Queries are submitted in YAML format; results are made available via a download link.
Limitations	For larger research projects, sufficient computing capacity is required, query runtime scales with the volume data.
Documentation	LiRI info page (LiRI) API Swissdox@LiRI API-Wiki Questions regarding queries should be directed to Swissdox@LiRI‑Plattform.
How to request access	Project registration required: see How to Get Started. Please observe the terms of use. Swissdox@LiRI: Login is restricted to members of supporting institutions or via a project voucher. Swissdox Professional (UZB-ZB license): temporary access (5 days) available upon request (e-medien@zb.uzh.ch)

License	✅ Included in the UZH-ZB license. Non-Journals (e-books & reference works) not included ✅ Non-commercial research, teaching, and learning purposes only. Full texts within existing subscriptions. ⚠️Using AI with licensed full texts (Journals)
Content & formats	Journal PDF full texts DOI‑based queries PDF downloads for individual or multiple articles (bulk) TDM API endpoint with optional Python client
Limitations	3 queries/sec Non-journal TDM projects require a paid TDM agreement (XML feed) (see below)
Documentation	Wiley TDM Policy API GitHub Documentation Python Client
How to request access	A Crossref TDM token is required for text and data mining. The use of Text and Data Mining is governed by the UZH‑ZB consortial framework agreement and follows the Wiley Text and Data Mining Agreement. However, accepting this Wiley click‑through license is necessary in order to obtain an API token.(How To Get Started) Non-journals TDM with XML feed is available upon request (tdm@wiley.com) and is subject to a fee. Please contact your library: emedia@ub.uzh.ch / e-medien@zb.uzh.ch

Text & Data Mining

What Is Text and Data Mining?

What Is the Problem with This?

What Are You Allowed to Do?

Cambridge University Press

EBSCO Databases

EMBASE (Elsevier)

Factiva (DowJones)

IEEE Xplore

IOP

LexisNexis API Web Services

Oxford University Press

Reaxys (Elsevier)

Science Direct API (Elsevier)

SciVal (Elsevier)

Scopus (Elsevier)

Springer Nature

Swissdox

Web of Science Expanded (Clarivate)

Web of Science Starter API (Clarivate)

Wiley Cochrane Library

Wiley Online Library

Additional Information

Questions about Text & Data Mining?

Wikipedia Articles