Data warehousing and analytics for chatbot

Abstract

For businesses it has become mandatory to solve the issues and problems that the customers are facing to ensure the customer loyalty. Man has always looked to take help of machines to remove human intervention and once again machines succeeded in doing that. Isn’t that strange we are using machines to provide services to humans? Well, that’s how technology has grown exponentially! This time it is the customer service industry which we are talking about and the invention responsible for this is ChatBot. The main aim of this project is to set up a data warehouse for a chatbot platform, set up ETL jobs so as to be able to push the existing as well as the incremental data to the warehouse which becomes the source for all our OLAP tasks. The data warehouse is then further used for creating dashboards using efficient BI tools for the business team to be able to roll out their weekly reports with ease and be able to take the necessary measures. One of the main intentions behind data analysis is to improve the quality of the bot that can further improve overall performance of the bot. For this purpose we have used interactions data in order to analyze usability, engagement and efficiency of the bot. The overall intention is to come up with a quality index for the bot based on some of the parameters which can further help to know the quality of the bot as well as suggest some changes to owners of the bot and also analyze that the changes that are made reflects in improving the bot without actually affecting the quality/engagement of the bot. In today’s world when there are a plethora of different data warehouses as well as visualization tools are available, to choose one is a tedious task. One of the main aim of this project is to find an efficient data warehouse and BI tool that will be used for analyzing the data.