This course introduces students to customer satisfaction measurement through a wide range of analytical approaches. We will discuss the components of customer satisfaction, major issues in measuring customer satisfaction, statistical methods in analyzing customer satisfaction influence, sentiment analysis with social media data, influence analysis with social media data, and text summarization with social media data. This course aims to provide the foundation required to make better marketing decisions by analyzing multiple types of data related to customer satisfaction.
Course Introduction and Module 1: Customer Satisfaction
With the first module, we begin by looking at some definitions of customer satisfaction. Then, we explore some major issues we have to consider. These issues include the psychological constructs of customer satisfaction, proper measurement of those constructs, varying targets of satisfaction, differences in the impact of individuals' expressions, and changing satisfaction over time. We will then introduce you to a new tool that you can use to conduct various data science methods on social media data. We conclude the module with a short primer on R and RStudio.
Module 2: Customer Satisfaction Analysis
We will begin our second module with a discussion on different types of data for customer satisfaction analysis. We first focus on survey data and look at different ways to analyze them. Next, we will provide a simple primer on linear and logistic regression. We will wrap up this module with a guided demo of utilizing sentiment analysis on tweets using the Social Media Macroscope.
We will introduce a method to analyze customer satisfaction influence using social media data. Social networks are the perfect dataset to utilize network analysis to understand how people are interacting with other people and forming networks. Identifying a pattern in social media relationships can be useful when making marketing decisions. We will also review influencer brand personality analysis that can be used as a method for brands to find influencers similar in personality to themselves.
Module 4: Text Summarization
We will learn about the various methods of text summarization. We begin by discussing the pre-processing steps required to bring the text to an analyzable form. Next, we look at how the frequency counts of multi-word phrases of pre-processed text can reveal the common terms being discussed. Building on top of the n-grams, we move onto a more intelligent method to automatically detect quality phrases. We will also discuss the LDA Topic Modeling - a very popular way to detect topics in a body of texts. We will wrap up this module with a highlight on supervised machine learning and an example of its application.