Advanced search engine

We offer both:  

1- The expertise for customized search engine integration according to various search criteria and platforms, that address the following issues of:

  • interoperability
  • scalability for large volumes of data and expanded systems
  • indexing data streams arising from documents, social media networks, audio or video sources

2- The expertise for extending search engine capabilities through semantic exploration

Computer-assisted information searching has become commonplace, as anyone with search engine availability can easily access documents in response to user-supplied subjects of interest. Nevertheless, polysemy (a word capable of having multiple meanings, such as “draft” in the sense of “preliminary form of writing” or “draft” in the sense of “gust of wind”) still represents a major problem to information searching, since it often leads to irrelevant results.

Information searching can also be performed in a question-answer manner, where the user is interested in a precise answer (e.g.: In what year did Mozart die?), in contrast to a list of documents on the subject (as in a Wikipedia page on Mozart). This avenue encompasses a vast area of research that includes natural language processing (***see other section) and possibly, in addition, the use of semantic Web (***see other section) for encoding the information in a more structured fashion.

Teams

Publications

Recent news

  • Blogue du CRIM - Manipuler les variables catégoriques dans un jeu de données
    12/07/2018

    Dans cet article, Farooq Sanni et Martin Sotir nous présentent différentes méthodes et astuces pour gérer les variables catégoriques.

    +

Upcoming event

  • QRS - SSCPS - IA 2018
    20 July 2018 0:00
    Lisbonne, Portugal
    Le CRIM présentera trois articles scientifiques au QRS 2018.
    +

Recent Publications

  • Towards Automatic Feature Extraction for Activity Recognition from Wearable Sensors: A Deep Learning Approach

    +
  • État des lieux des technologies web

    +