Information Retrieval systems are used to extract, from a large database, relevant information for users. When the type of data is text, the complex nature of the database makes the process of retrieving information more difficult. Generally, such processes reformulate queries according to associations among information items before the query session. In this latter, semantic relationships or other approaches such as machine learning techniques can be applied to select the appropriate results to return. This paper presents a formal model and a new search algorithm. The proposed algorithm is applied to find associations between information items, and then use them to structure search results. It incorporates a natural language preprocessing stage, a statistical representation of short documents and queries and a machine learning model to select relevant results. On a series of experiments through Yahoo dataset, the proposed hybrid information retrieval system returned significantly satisfying results.

Generating execution plans is a costly operation for the DataBase Management System (DBMS). An interesting alternative to this operation is to reuse the old execution plans, that were already generated by the optimizer for past queries, to execute new queries. In this paper, we present an approach for execution plan recommendation in two phases. We firstly propose a textual representation of our SQL queries and use it to build a Features Extractor module. Then, we present a straightforward solution to identify query similarity.This solution relies only on the comparison of the SQL statements. Next, we show how to build an improved solution enabled by machine learning techniques. The improved version takes into account the features of the queries’ execution plans. By comparing three machine learning algorithms, we find that the improved solution using Classification Based on Associative Rules (CAR) identifies similarity in $91\%$ of the cases.

In this paper we describe a mechanism to improve Information Retrieval (IR) on the web. The method is based on Formal Concepts Analysis (FCA) that it is makes semantical relations during the queries, and allows a reorganizing, in the shape of a lattice of concepts, the answers provided by a search engine. We proposed for the IR an incremental algorithm based on Galois lattice. This algorithm allows a formal clustering of the data sources, and the results which it turns over are classified by order of relevance. The control of relevance is exploited in clustering, we improved the result by using ontology in field of image processing, and reformulating the user queries which make it possible to give more relevant documents.

One of the well known techniques that are used to increase the relevancy of user queries is query reformulation through the expansion of queries. Also, Linked data are being used in various fields and available for different domains. This paper aimed at improving Information Retrieval results using Linked data from Wikipedia in general and Wikidata in particular to expand the queries with attributes" values. The experiments that we have done, using Medline collection and Indri search engine to compare our suggested approach with baseline, lead to improvements in precision at different ranks.