About the project: Kisan Call Centre Data Analysis
Table of Contents
About
As an IEEE Student Member, I participated in a research project on Kisan Call Centre Data Analysis under Dr. Neha Sharma, IEEE Execom Member, and Scientist, Tata Consultancy Services Pune.
Background
The Kisan Call Centre is an initiative by the Ministry of Agriculture & Farmers Welfare, Government of India. It was launched on January 21, 2004. The primary objective of the project is to answer farmers’ queries on a telephone call in their own dialect.
Our Contributions
During exploratory data analysis, we found that the Government allocates a substantial portion of the budget towards agriculture. Therefore, through this project we aimed at using insights from the data collected at KCC for the economic aspect of decision-making for civic bodies.
In the first part of the study, I was involved in the Preprocessing and visualisation of the dataset.
The second part of our study focused on resolving manual entry errors by the Farm Tele-Advisors at KCC -
- Analysis of the KCC dataset showed that a majority of farmers rated responses to their queries as satisfactory
- Further, the Farm Tele Advisors relied on self-knowledge on many occasions to resolve the queries
- We also found discrepancies in the selection of ‘Query Type’ from ‘Query Text’ attribute of the dataset
In order to address the issue, we proposed automatic selection of Query Type from Query Text. For the same, we worked on 2 methods -
- Document similarity using TF-IDF and K-Means clustering
- Word-based Embeddings
We presented our work to students of Duale Hochschule Baden-Württemberg (DHBW) Stuttgart and received a positive response from them about our work!
We documented our findings and they were published in the Springer book on Open Data for Sustainable Community.
PS: You can find my name in the list of Contributors :)