Project: Biographical information extraction and people search platform
The BIOGRAPHE project aimed at developing an innovative people search platform._x000D__x000D_People search consists in searching for and finding people or information on people instead of searching for documents. These include locating classmates and old friends, finding casual acquaintances on social networks, finding experts or employees, finding Ps for date and romance, finding phone numbers or addresses in white and yellow pages, etc._x000D__x000D_In a corporate environment, people search is of great use for finding experts for a project or for finding a person who meets specific qualifications in a selected geographical area._x000D__x000D_The BIOGRAPHE project will lead to the creation of a multipurpose people search platform which will be able to reconstruct biographies of people by using all available information sources such as profiles on social websites, press articles, CVs or internal documents written by these people. The people search platform will collect, extract and structure this multilingual information in indexes and relational databases ready to be used by different task-oriented people search applications._x000D__x000D_This BIOGRAPHE platform will serve as a basis for at least three different applications, namely:_x000D__x000D_- Matching resumes and job profiles (for employee recruitment),_x000D_- Displaying automatically generated biographies of famous or not-so-famous people (for information purpose, like the « Who's Who » reference publications),_x000D_- Identifying experts and networks of expertise inside an organization (in order to find experts for a specific task)._x000D__x000D_Consequently, CO steps of the BIOGRAPHE project are as follows:_x000D_- Building a system for extracting biographical information from multiple sources (such as newspapers, personal web pages, CVs, social network profiles) in four languages (English, French, German and Dutch);_x000D_- Structuring the collected data in a common multilingual database which will allow information retrieval, network identification, disambiguation and incremental update. This database will be the foundation of the different people search-related applications (automated generation of profiles, comparison, expert recruitment, etc.);_x000D_- Developing a generic people search framework;_x000D_- Developing three specific search applications derived from the common people search framework;_x000D_- Developing an online general-purpose demonstration for people search._x000D__x000D_The platform and the following people search applications will be conceived and developed by a consortium of European R&D performing SMEs and renowned academic research institutes:_x000D__x000D_- Sinequa, a French SME providing corporate search solutions and being the BIOGRAPHE project leader;_x000D_- Jobanova, a German software company operating a job search engine that uses advanced linguistic and semantic search technology;_x000D_- CENTAL (Université Catholique de Louvain), a university laboratory specialized in the study of Natural Language Processing;_x000D_- my-xML, a Luxembourgish SME which develops solutions for managing and accessing multilingual contents within organizations and companies;_x000D_- Belga, the Belgian press agency;_x000D_- 123people, an Austrian SME which developed the leading people search tool in Europe;_x000D_- And CIS (Centrum für Informations- und Sprachverarbeitung), a German research institute specialized in information and language processing.
| Acronym | BIOGRAPHE (Reference Number: 4621) |
| Duration | 01/03/2010 - 31/12/2012 |
| Project Topic | Development of a multi-purpose people search platform, based on multi-sources biographical information extraction, offering multilingual search capabilities and allowing to search for people, not only by name but also by any kind of characteristics: technical skills, work experience, location... |
|
Project Results (after finalisation) |
Based on the outcome of the preceding work apckages the CO objective was to pull the different components together and integrate them into a realtime search engine environment to fulfil the defined use cases in a real world scenario. |
| Network | Eurostars |
| Call | Eurostars Cut-Off 2 |
Project partner
| Number | Name | Role | Country |
|---|---|---|---|
| 8 | Agence Télégraphique Belge de Presse S.A. | Partner | Belgium |
| 8 | Centrum für Informations- und Sprachverarbeitung | Partner | Germany |
| 8 | Jobanova GmbH | Partner | Germany |
| 8 | my-xML | Observer | Luxembourg |
| 8 | Sinequa | Coordinator | France |
| 8 | Université catholique de Louvain / Centre de traitement automatique du langage | Partner | Belgium |
| 8 | yelster digital gmbh | Partner | Austria |
| 8 | yelster digital gmbh | Partner | Austria |