Smart Sampark-An approach towards building a responsive system for Kisan Call Center
- Resource Type
- Conference
- Authors
- Ajawan, Pratijnya; Desai, Pooja; Desai, Veena
- Source
- 2020 IEEE Bangalore Humanitarian Technology Conference (B-HTC) Bangalore Humanitarian Technology Conference (B-HTC), 2020 IEEE. :1-5 Oct, 2020
- Subject
- Aerospace
Bioengineering
Communication, Networking and Broadcast Technologies
Components, Circuits, Devices and Systems
Computing and Processing
Fields, Waves and Electromagnetics
General Topics for Engineers
Geoscience
Photonics and Electrooptics
Power, Energy and Industry Applications
Robotics and Control Systems
Signal Processing and Analysis
Transportation
Natural language processing
Agriculture
Pandemics
Meteorology
Frequency measurement
Tokenization
Indexes
KCC
Farming
Querysytem
Cosine Similarity Index
Natural Language Processing
- Language
The pandemic COVID-19 has affected the economic growth of the country. The farming industry is facing the impact of the pandemic. In India, Kisan call centers (KCC) or agricultural helplines are set up throughout the country to respond to the farmer's queries and to guide them during times of uncertainty and distress. There are 21 KCC set up across different states and union territories of India. At KCC, the operators manning the helpline or the agricultural experts respond to the farmer's queries depending on their knowledge and data available with them. The disadvantage of the existing system is that the operator or the experts have to be always available to answer the calls. During times of uncertainty like the pandemic, the calls remain unanswered and the farmers have no point of contact to get solutions to their queries. This paper Smart Sampark is an approach to build an answering system for Kisan Call center. It is a conversational system that will help the farmers to obtain answers to their queries in an easy and systematic way. Smart Sampark aims at providing virtual conversational assistance to the farmers similar to the Chabot. The proposed system uses the data set made available on data.gov.in with KCC queries. The natural language processing (NLP) technique has been used to generate the response to the queries and cosine similarity method to generate the most similar response to the query. The results obtained using KCC data of Belagavi district for a period of 36 months is accurate to 86%. The same can be improved by extending the data set for experimentation.