KisanQRS: A deep learning-based automated query-response system for agricultural decision-making

Abstract

Delivering prompt information and guidance to farmers is critical in agricultural decision-making. Farmers’ helpline centres are heavily reliant on the expertise and availability of call centre agents, leading to inconsistent quality and delayed responses. To this end, this article presents Kisan Query Response System (KisanQRS), a Deep Learning-based robust query-response framework for the agriculture sector. KisanQRS integrates semantic and lexical similarities of farmers’ queries and employs a rapid threshold-based clustering method. Before clustering, KisanQRS employs data cleaning by performing steps such as the removal of special characters and extra white spaces. It also removes crop names from the queries, which helps in putting similar queries to the same clusters irrespective of the crop. Subsequently, semantic and lexical similarities of queries are computed, and a final similarity matrix is generated based on the weighted sum of these two similarities. Next, a clustering algorithm is applied to organize the queries into clusters according to their similarity. A query mapping module is trained on these cluster labels as the target. For this query mapping module, LSTM is found to be the optimal method. Next, the answer retrieval method clusters candidate answers for a crop, ranks these answer clusters based on the number of answers in a cluster, and selects the leader of each cluster, finally providing the top-K answers. The dataset used in our analysis consists of a subset of 34 million call logs from the Kisan Call Centre (KCC), operated under the Government of India. We evaluated the performance of the query mapping module on the data of five major states of India with 3,00,000 samples and the quantifiable outcomes demonstrate that KisanQRS significantly outperforms traditional techniques by achieving 96.58% top F1-score for a state. The answer retrieval module is evaluated on 10,000 samples and achieves a competitive NDCG score of 96.20%. KisanQRS could be useful in enabling farmers to make informed decisions about their farming practices by providing quick and pertinent responses to their queries. © 2023 Elsevier B.V.