Copyright: ThalesSource : Thales


AI x Knowledge Management

Leveraging AI to enhance knowledge discovery and document generation

This challenge is over, applications are closed.

Challenge Overview

Description du challenge

Data and knowledge are the core foundation of a company. However, the bigger the company is, the harder it is to get easy access to this information. The aim is to implement methods to compute and classify it to add new knowledge easily and find the one already available accurately and efficiently thanks to AI, and then automatically generate new documents based on existing data. We are also interested in implementing an automated data classification system.

Copyright: metamorworks
Copyright: metamorworks
Copyright: efks
Copyright: efks


Expert in

  • Data analysis and mining to extract useful information from large datasets and identify patterns in data that can be used for classification
  • Feature analysis
  • Machine learning
  • Pattern detection

Experience in

  • Developing tools for big data analytics, search and visualization
  • Knowledge Representation to represent the rules and criteria used to classify documents and enable the system to learn from them over time.
  • Human-Computer Interaction (HCI) to design an intuitive interface that allows users to interact with the system and provide feedback on the accuracy of classification

Projects you could be working on

  • Using cognitive search to extract relevant information from multiple, diverse data sets
  • Knowledge engineering tool that uses ontology to identify and suggest metadata to put them in a database
  • Modeling of ontologies using semantic AI to describe data relations with optimized indexing to improve knowledge structures
  • Automatically generate documents based on an existing knowledge base of similar existing documents and guardrails in terms of actual data, processes,  procedures and new requirements that the generated document needs to comply to
  • Develop an automated data classification system that can accurately and efficiently identify the level of confidentiality of documents. The model must be able to distinguish high-level confidential data from non-sensitive data that can be diffused quickly, using confidentiality classification guides