To construct an improved farmer-centric information retrieval system smart agriculture using machine learning techniques and fuzzy logic is proposed in this paper. In case-based reasoning systems, the capacity to precisely define cases is crucial. Researchers have extensively analyzed various representations, including linguistic, attribute-value, and ontological models. As a result, retrieving cases from large databases might be time-consuming. In this paper, we propose a strategy for efficient case retrieval based on the concept of related representations. Whether or not two cases share similarities or differences, they are still connected. After a case is reported, it is compared to past data in an effort to find an exact match. Related cases are evaluated for similarities rather than the entire case base. The related case representation and conventional techniques were compared for their respective caseloads and retrieval accuracies. Fuzzy rules are utilized to define the threshold values of the various models, and Independent Recurrent Neural Networks (Ind RNN) are employed to evaluate the effectiveness and similarity of the models. The findings suggest that the idea can be used for very accurate and fast case retrieval. The related case representation strategy outperforms competing approaches in terms of retrieval effectiveness.