IntraFind Research Projects

An important point for business decisions is that one always has an ear for scientific advancements outside the market economy. What happens at the university level, what are the topics that develop students and professors together today? In order not to miss these impulses IntraFind involved from the outset, a team of scientists actively support the research projects.

Completed Research Projects:

Research Project OntoPMS – Follow-Up Project of OntoVigilance

Following the successfully completed project OntoVigilance, IntraFind won the project OntoPMS to develop an ontology-based risk identification software for monitoring medical products on the market (Post Market Surveillance, PMS). This project was funded in the context of the support program KMU-innovativ of the Federal Ministry of Education and Research (BMBF).

Compared to OntoVigilance, in the OntoPMS project more documents and risk-specific data sources have been included. Also the Federal Institute for Drugs and Medical Devices (BfArM) has joined the project as additional partner.

In this project IntraFind has been working on an ontology-based process for relation and event extraction. In particular, the concept of Ontology Search has been developed, with which the domain-dependent and ontology-based share of information extraction is mainly carried out on complex new search operators on a linguistically enriched full-text index. This approach has the advantage that it can respond much more flexible to changes in the ontology.

Research project Ontovigilance - improvement of medical devices

As part of the funding innovative KMU-innovativ of the Federal Ministry of Education and Research, IntraFind could win the research project Ontovigilance (Start: March 1, 2013 Project duration: two years).
 
The goal of the project is to develop an ontology-based vigilance software for efficient evaluation of scientific publications and reports regarding the use of technical medical equipment. The deficiencies, complications and risks associated with the devices described in the texts, are software-supported detected and extracted. On this basis, the system developed in the project will provide a decision support for the optimization of medical devices.
 
Compared to manual handling of support cases the support by software-based methods helps manufacturers of technical medical equipment to detect and respond early to product defects. This saves valuable time and money in product development and represents a competitive advantage.
 
For this purpose IntraFind develops these ontology-based search and information extraction methods in close cooperation with the Institute for Medical Computer Science, Statistics and Epidemiology at the University of Leipzig and the companies Novineon and MT2IT. Employed are the IntraFind products Tagging Service and NAMER (entity recognition) based on a SOA architecture, for generating ontologies and metadata from the provided medical documentation (target languages​​: German, English).

Social CRM Intelligence

The succesfully completed research project SCRM-I dealt with the development of innovative services for the expansion of existing CRM software including semantic text mining. The services should simplify the identification of messages, requests, trends, opinions and developments, and support businesses in the interaction with the social web. Thus, the Social CRM gives an insight into how companies, brands and products are evaluated by consumers on public social media platforms. IntraFind contributed their long-standing know-how in the area of ​​text analysis in the cooperation with the Universities of Leipzig and Hohenheim.

Who was the inventor of the espresso machine? Ask questions - get answers

In the research project NL-Search (KMU-innovativ / ICT) a search engine that is able to answer natural language questions in real time was developed during a two years project with the University of Saarland. Instead of entering keywords, as in popular search engines, the user can ask simple questions in natural language. The search results are not complete documents but short passages or phrases, containing an answer to the question asked. Basis of the search engine is the IntraFind Enterprise Search product iFinder.