Smart Harvesting AI

Grow the coverage and accuracy
of your research information hub

Smart Harvesting AI is at the heart of how Esploro keeps research repositories and researcher profiles comprehensive, accurate and always up to date. It automatically captures research output and data, improving an institution’s ability to showcase its research, enrich its reporting, and relieve the administrative burden.


Populating the research information hub

Improved discoverability of your institution’s research makes it easy to find subject matter experts and increase academic collaboration.

  • A centralized, comprehensive repository of scholarly information, activities and entities.
  • Includes publications, datasets, awarded grants, projects, creative works, media mentions, and more – all linked.
  • Supports all academic disciplines, including STEM, Humanities, Social Sciences, and the Arts.
  • Rich, auto-populated metadata increases discoverability via search engines.
  • Information gathered from a massive set of output records in the Ex Libris Central Index and external sources.

Reducing administrative burden

Intelligent automation reduces the time and effort librarians, researchers or research office staff need to spend maintaining repositories and faculty profiles.

  • Automatic aggregation of research outputs and data.
  • Researcher profiles are automatically updated.
  • Null or indeterminate results are automatically eliminated.
  • Missing metadata and links are added.
  • Misidentifications are corrected and duplications removed.
  • Automatic approval of new records can be customized based on the strength of the match or the output type.

Matching researchers to their work at scale

Smart Harvesting accurately and efficiently links authors and their work at scale, assessing mass volumes of scholarly output and rapidly populating the Esploro research information hub.

  • Learns all variants of a researcher’s name, affiliations and specializations, years of professional activity, previously known assets, and more.
  • Cross-checks known data points with new records, semantic contexts, subject matter, co-authors, metadata, and more.
  • Collects multiple types of scholarly information, including output, activities, projects, grants and achievements, and links everything together.
  • Rapidly assesses the likelihood of a given output being the product of a specific researcher.
  • Universities using Esploro see an immediate and significant increase in the number of linked assets per researcher.

Getting smarter over time

The power of Smart Harvesting to rapidly fill in information gaps becomes even more evident as engagement with Esploro continues over time.

  • Unique machine learning algorithms identify the relationships, correlations and patterns that emerge from a large set of research outputs and profiles.
  • Natural language processing reveals how close to one another selected words are likely to be, and in what combinations, across all examined records.
  • With periodic checks for new outputs, the solution continuously enriches its understanding of the researchers and their work.
  • The more information collected, the more accurately and consistently each researcher’s profile will be populated.

Searching billions of records

Smart Harvesting AI searches for relevant research assets in your local databases, as well as in national and disciplinary repositories. However, the main source of information is the Ex Libris Central Index.

  • Over 4.5 billion records.
  • More than 5,000 content providers and data sources.
  • Articles, books, chapters, digital resources, datasets, open access content, dissertations, patents, audiovisual content, conference proceedings, and other research outputs.
  • All disciplines, including STEM, Humanities, Social Sciences, and the Arts.