Mokhtar Billami

Retrieval Augmented Generation

From Generic to Domain-Smart: Tailoring Embeddings for High-Performance Business Search

In details, Research / By

1. Strategic Context and Challenges Modern answer engines look simple from the outside: a user asks a question, and the system returns a grounded answer. But before the language model can write anything useful, the system must first retrieve the right sources from the enterprise knowledge base. If the right sources are not selected, the […]

From Generic to Domain-Smart: Tailoring Embeddings for High-Performance Business Search Read More »

Schema matching illustration

Semantic Mediation for Data Exchange

In details, Research / By

In the current socioeconomic context, companies must work together. Interoperability, which is the action to work together through systems is an unconditional factor to do so. To work, the information shared needs to be compatible. Information can be incompatible when they’re built differently by people or systems.Mediation goal is to manage syntactic and semantic information

Semantic Mediation for Data Exchange Read More »

Let’s learn: I speak, you speak, he/she speaks but does the machine understand him/her?

Event / By

We will have the pleasure to listen to the DRI member Mohktar Billami for a presentation as part of Let’s learn, this Thursday, 14 January from 13:15 to 14:00. Abstract: Automatic translators, suggested words for smartphone input, search engines, spell checkers, chatbots, etc. You may not realize it, but you use automatic Natural Language Processing

Let’s learn: I speak, you speak, he/she speaks but does the machine understand him/her? Read More »

DEESSE: A Generic Search Engine based on Artificial Intelligence

Leave a Comment / In details, Innovation / By

In Berger-Levrault, several of our software products incorporate a search engine in themselves to facilitate access to information. This information can be encoded in files having various and varied formats. For example, Word, PDF, CSV, JSON or even XML files. Current search engines such as Lucene focus most of the time on keyword searches. This

DEESSE: A Generic Search Engine based on Artificial Intelligence Read More »

When AI generates Human Resources training plans!

Leave a Comment / Innovation / By

Berger-Levrault works in collaboration with the Cusma association and the innovation commission to exchange experiences and advice on innovation applications. Among the projects, we started working on a supervised automatic classification of training requests.This project aims to assist administrative officers using artificial intelligence and robots to classify and prioritize training requests coming from agents. We

When AI generates Human Resources training plans! Read More »

Building an Ontological Knowledge Base from Berger Levrault Editorial database

Leave a Comment / Research / By

Context Ontologies are widely used in information retrieval (IR), Questions/Answers, and decision support systems, and have gained recognition as they are now being considered the new answer to semantic interoperability in modern computer systems and the next big solution for knowledge representation. The structuring and management of knowledge are at the heart of the concerns

Building an Ontological Knowledge Base from Berger Levrault Editorial database Read More »

Start typing and press enter to search

Scroll to Top