A well-established and extensively used method in the field of information retrieval is query expansion (QE). By updating the query definition, it improves the relevant documents that are returned. Some data requirements are challenging to meet because users’ queries are frequently brief and imprecise due to dialect ambiguity. In order to address this problem, query restructure and expansion techniques have been created. The weighted Divergence from Randomness (DFR) models were compared and analyzed by the authors using Terrier 4.1 on the TREC ad hoc datasets. According to the results, the relevant documents that were located using the QE approaches had substantially higher recall and precision than the DFR models used as a baseline.