Optimization of the results of a multilingual search engine using a fuzzy recommendation approach

Search engines are now the main source for information retrieval due to the huge expansion of data on the internet over the last ten years. Providing users with the most relevant results for their queries poses a significant challenge for search engines. Semantic search engines, which go beyond traditional keyword-based searches, have appeared as advanced information retrieval systems to address this problem. These search engines produce more precise and pertinent search results because they understand the meanings of words and their relationships. They play a pivotal role in managing the vast amount of internet data, with a primary aim of enhancing search precision and user satisfaction. However, improving search precision remains as an important goal for natural language processing researchers. The main objective of our research is to improve the search engine results. We present a novel approach for measuring the similarity between a user's query and a list of documents within a search engine. This approach provides a new fuzzy recommendation system using a syntactic and semantic similarity. Our results indicate that our method outperforms several existing approaches from the literature, achieving a high level of accuracy.