Abstract
This chapter presents data science research conducted by authors Sarawut Ramjan and Jirapon Sunkpho, showcasing the use of RapidMiner to gather data from social networks. The research includes the analysis of customer satisfaction for mobile applications, predicting condominium prices in Bangkok, and discovering demand and supply patterns based on Thai social media data using the association rule mining approach. Additionally, the chapter touches on other topics such as corrosion under insulation severity classification for carbon steel in a marine environment and variables that influence pilot's safe driving. The chapter is beneficial for readers without data science experience and for businesses interested in employing data mining techniques. The research highlights that software application skills are not the only important factor, but also understanding the processes of data science, such as data exploration, data pre-processing, data mining, and data presentation, which are essential and useful skills.
TopThe Case Study On Analyzing Customer Satisfaction On A Mobile Application Using Data Mining Techniques (Sunkpho & Hofmann, 2019)
The case study shows the analysis on customer satisfaction using mobile app services of state-owned energy enterprises in the metropolitan area of Bangkok, Thailand. The analysis follows CRSIP-DM Process, which this book has already mentioned in Chapter 1. The researcher found that Data used to reflect performance of the state energy enterprise is the satisfaction of users on services related to information technology and one important service, the Mobile App.
Mobile App is available to users as they can access the Mobile App to check the amount of energy usage each month. The app notifies important news, such as the cessation of energy services for infrastructure repairs, the area where the power is interrupted and unable to provide service, and payment channel.
Business Understanding
In the Business Understanding process, the key is that Energy enterprises have questioned the variables that effect customer satisfaction and dissatisfaction so that SOEs can develop features. Therefore, the researcher has formulated a problem for data analysis, namely Classification of data that should be used to predict customer satisfaction.