| Popis: |
The search for relevant scientific articles is a crucial step in a scientific research project. However, the considerable number of articles published and uploaded online in digital databases (such as google scholar, semantic scholar, etc.) makes this task tedious and negatively impacts the researcher's productivity. This article proposes a new method of recommending scientific articles, taking advantage of content-based filtering. The challenge is to target the relevant information to meet the researcher's needs regardless of his research's domain. Our recommendation method has based on a semantic exploration using the latent factors. Our goal is to achieve an optimal topic model, which will serve as the basis for the recommendation process. Experiences confirm our performance expectations by showing relevance and objectivity in the results. |
