An AgroBot: Natural Language Processing Based Chatbot for Farmers

• Published in: 2023 4th International Conference on Smart Electronics and Communication (ICOSEC) pp. 1235 - 1241
• Main Authors: Marla, Anushka, Paul, Rejath, Saha, Arupam Kumar, Basha, Niha Kamal, Anandhakrishnan, Balasundaram
• Format: Conference Proceeding
• Language: English
• Published: IEEE 20.09.2023
• Subjects: Agriculture; Chatbots; Conversational Artificial Intelligence; Crops; Farming; Machine learning; Natural Language Processing chatbot; Productivity; Profitability; Soil; Sustainable Farming; Text mining
• DOI: 10.1109/ICOSEC58147.2023.10276356

In an industry that is enduring a rapid transformation, rowers are gaining access to new technologies that help them improve agricultural yields and resource management. TensorFlow-built chatbots can provide instantaneous assistance to producers. TensorFlow is a machine learning (ML) framework for data automation, model monitoring, and model retraining. It was used to analyse more than 80 million call records from the government portal Kisan Call Canter (KCC) using multiple data preprocessing techniques in conjunction with TensorFlow. Training enabled the chatbot to distinguish between agricultural questions, crop management, soil health, and pest control intentions and entities. This chatbot intends to implement a voice recognition module that will enable users to submit speech-based queries rather than text-based ones. In the future, the chatbot can be trained to comprehend images with high precision and be able to identify crops, provide recommendations on how to increase crop productivity, and promote the need for sustainable farming. The automaton has been evaluated using metrics for precision and error reduction. It's conceivable that TensorFlow-trained chatbots could revolutionize agriculture by providing producers with real-time data and advice on how to increase their yields and revenue.