Call Data Analytics Using Big Data

Abstract

One of the important strategies that can increase the business success rate is “customer monitoring” in which customer service representatives (CSRs) measure the satisfaction of customers. Here, the only way to know about customer’s experience (CX) is conducting the surveys which are either through phone calls or sending emails. The main challenge of customer monitoring is to know about customer’s experience, and the customer service representatives record their phone calls and analyze those recorded calls by converting them into text files. They maintain a large amount of memory to store a huge number of audio files and text files. This results in data tempering, corruption of data, unauthorized access to tables, columns and rows and burden of managing the data. In this paper, we replace the recorded files with direct phone calls. Now, we can convert the phone calls to text files with the help of speech-to-text (STT) algorithm, then analyze a huge amount of text files using Hadoop MapReduce Framework and apply the text similarity algorithms for getting better results to improve the business.