Agribot: agriculture-specific question answer system

2025-09-25

AI Summary

Indian farmers face persistent challenges in accessing timely agricultural information—including weather forecasts, market prices, pest management guidance, and policy updates—due to inadequate coverage and delayed responses from existing Kisan call centers. To address this, we propose a lightweight, domain-specific question-answering chatbot system for agriculture. Our approach constructs a knowledge base from real-world call-center dialogues and introduces a novel sentence embedding method that jointly integrates named entity recognition with synonym resolution, significantly improving semantic matching accuracy for agricultural terminology. The system supports 24/7 multi-platform access and enables automated question classification and precise response generation. Evaluation shows an improvement in answer accuracy from 56% to 86%, substantially alleviating the burden on human agents while enhancing information accessibility and decision-making efficiency for farmers. This work provides a scalable, resource-efficient technical framework for deploying intelligent agricultural services in low-infrastructure settings.

Technology Category

• Natural Language Processing: Question Answering
• Data Mining & Knowledge Management: Conversational Systems for Recommendation & Retrieval
• Multiagent Systems: Agent Communication

Application Category

• Search and Retrieval-Augmented AI: Agentic search
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

India is an agro-based economy and proper information about agricultural practices is the key to optimal agricultural growth and output. In order to answer the queries of the farmer, we have build an agricultural chatbot based on the dataset from Kisan Call Center. This system is robust enough to answer queries related to weather, market rates, plant protection and government schemes. This system is available 24* 7, can be accessed through any electronic device and the information is delivered with the ease of understanding. The system is based on a sentence embedding model which gives an accuracy of 56%. After eliminating synonyms and incorporating entity extraction, the accuracy jumps to 86%. With such a system, farmers can progress towards easier information about farming related practices and hence a better agricultural output. The job of the Call Center workforce would be made easier and the hard work of various such workers can be redirected to a better goal.

Problem

• Developing an agriculture-specific chatbot for Indian farmers
• Answering queries about weather, market rates and plant protection
• Improving accuracy through entity extraction and synonym elimination

Innovation

• Uses sentence embedding model for query processing
• Incorporates entity extraction to enhance accuracy
• Provides 24/7 agricultural information via chatbot