Join as a Research Community

The researcher at the centre

OpenMinTeD builds an online platform where researchers can find:

• Text mining tools and services from around the world
• Openly available text and data sources, ready to be mined

Whether you’re a biochemist, a historian, a physicist or something completely different, OpenMinTeD would like to help you in finding the relevant text and data mining techniques, as well as the relevant data to mine. We want our platform to be there for all research disciplines.

In order to build a platform where researchers’ from all research communities find what they need, we invite research communities to join OpenMinTeD.

We are interested to hear about:

• Your requirements with regard to TDM practices. What tools, resources and standards are used in your research community?
• The prototype applications in your research community. Which applications would you like to be included in the OpenMinTeD platform?
• Your opinion on different applications. How can they be improved in relation to OpenMinTeD?

Interested to join OpenMinTeD with your research community? Please leave your contact information here and we will get back to you!

Research communities and applications already involved

Research communities in life sciences, agriculture & biodiversity, social sciences and research analytics are already working with OpenMinTeD on making the platform fit their needs. Have a look at the applications and uses cases below for an impression:

Life sciences

Extract metabolites and their properties and modes of action

When scientists need information about the structure, name or properties of small molecules, they often turn to a high quality database called ChEBI. This database is largely curated manually and this process takes a lot of time. OpenMinTeD is working on a textmining application that can help to speed up the process, while maintaining the quality of the database. More information is available in the blogpost Text mining for the discovery of small molecules. Involded partners: University of Manchester and EMBL-EBI.

Text mining for curation of neuroscience literature

To understand how the nervous system works, researchers develop mathematical models based on all the knowledge that is available. To improve these models, it is important to include the new facts and parameters that are published in scientific journals. OpenMinTeD works towards text mining applications that automate the curation of neuroscience literature for this purpose. As part of this work, a Framework for Collaborative Curation of Neuroscientific Literature was produced. Involved partners: University of Manchester and EPFL

Text mining on articles related to Health State Modelling in Chronic Liver Diseases

It is important that patients receive the right intervention at the right time. In this case ‘at the right time’ means that treatment has to be adjusted in line with the health state of the patient. OpenMinTeD wants to find out if it is possible to use text and data mining on data repositories with scientific articles, to define different health states. For the proof-of-concept, we focus on Chronic Liver Diseases. Involved partner: Frontiers

Agriculture and biodiversity

Text mining over bibliographic data:  AGRIS & CORE

The viticulture community deals with the study and production of grapes. Text mining can improve viticulture practice, by feeding it with information from bibliographic data sources (AGRIS and CORE). OpenMinTeD works on an application, that deals with the extraction of structured information out of these bibliographic data sources. More information is available in the blogposts Text mining in the vineyard and OpenMinTeD partner presents VITIS pilot application.  Involved partner: Agroknow

Text mining over RSS feeds: food safety and water health

RSS feeds contain useful information, for example on geolocations of foodborne illness outbreaks, food recalls, water pathogens and water quality. In the context of the OpenMinTeD project, an application will be developed that text mines unstructured RSS feeds and shows relevant information per selected country or region. Involved partner: Agroknow

Microbial biodiversity

Genomics information is often ‘hidden’ in various scientific papers and free-text fields of databases. The microbial biodiversity text mining application allows researchers to find and select relevant genomics information and discover new knowledge. It uses a combination of natural language processing and machine learning methods and focuses on food microbiology. Read more in the blogpost What microorganisms live in my cheese? and the infogaphic Text and data mining for better microbiology Involved partner: INRA

Linking wheat data with literature

In order to feed the growing world population, it is ever more important to understand crop traits like disease resistance or grain number. These traits have a basis at the molecular level. Findings about related genes and molecular markers are published in scientific papers and in certain databases. The application ‘Linking Wheat Data With Literature’ processes and identifies this information and feeds a database with links between ‘external’ texts and information in the database. Involved partner: INRA

Extracting gene regulation networks involved in seed development (SeeDev)

For breeding and production of crops, it is important to understand how seeds develop. The SeeDev TDM application harvests scientific papers on this topic, turns them into mineable text and helps to select the relevant information. Because biological mechanisms are very complex, many different types of information have to be included. Involved partner: INRA

Social sciences

Facilitation of complex information linking and retrieval from social sciences publications

In the social sciences, we know that a useful application for text mining is improving literature search and information interlinking. As part of the OpenMinTeD project, we first held a questionnaire to get a more specific understanding of the TDM needs in the social sciences community. Based on that, we decided to focus on three things: named entity recognition and linking; variable mention detection and linking; keyword assignment. Explanations are available in the blogpost: Text mining for social sciences – tackling the challenges to make search systems smarter. Involved partners: GESIS, UKP-TUDA

Research analytics

Research analytics – funding mining services

This application provides useful information on how research is done in the EU and can help funders and institutions to make informed decisions. The application mines fulltext publications and looks for linked information, for example on the relation between topics and funders or where authors are based. With topic modeling you can find out if a topic is trendy and what the top most related publications are. The application was developed by OpenAire, and will be made interoperable with the OpenMinTeD platform. Involved partner: ARC

Research publications recommendation system

Imagine that you are doing research on a certain topic, and that you get automatic recommendations on for example related papers that you had not thought of or found yourself. It could really improve the quality or speed of your research. This is exactly what this application is about. It offers content-based recommendations to scholarly materials, and relies on metadata, full text features and other features from the article full texts or external sources, like citations, readership or publication year. Involved partner: Open University

Research Performance

Text and data mining can also be used to track research performance of papers, individuals or organization. This application obtains information from full text articles and metadata, on for example citations, downloads or altmetrics. The results are delivered in the form of a dashboard, that shows graphs and allows benchmarking, comparison and trend analysis with regards to content and performance. Involved partner: Open University

Text mining of articles related to Leica Microscopes

There’s a lot of discussion about the impact of research, research institutes and the evaluation of individual researchers. But what about the impact of tools and equipment, for example of a  microscope? This use case explores if text and data mining of scientific articles can be used to measure the research impact of a Leica microscope. Involved partner: Frontiers

Rock art mining

This use case explores how you can use text mining to speed up rock art research (e.g. on ancient cave paintings). The objective is to identify potentially interesting articles and to extract information like sites, coordinates, dates and dating methods. The annotated texts and metadata will then be stored and made available. We use CORE as a bibliographic source. Involved partner: Frontiers.