Publication Type

Conference Proceeding Article

Publication Date

11-2016

Abstract

Social media provides a convenient way for customers to express their feedback to companies. Identifying different types of customers based on their feedback behavior can help companies to maintain their customers. In this paper, we use a machine learning approach to predict a customer’s feedback behavior based on her first feedback tweet. First, we identify a few categories of customers based on their feedback frequency and the sentiment of the feedback. We identify three main categories: spiteful, one-off, and kind. Next, we build a model to predict the category of a customer given her first feedback. We use profile and content features extracted from Twitter. We experiment with different algorithms to create a prediction model. Our study shows that the model is able to predict different types of customers and perform better than a baseline approach in terms of precision, recall, and F-measure. © Springer International Publishing AG 2016.

Keywords

Customer relationship management, Machine learning, Social media

Discipline

Communication | Social Media

Research Areas

Software and Cyber-Physical Systems

Publication

Proceedings of 8th International Conference on Social Informatics (SocInfo 2016): Seattle, USA,

Volume

10047

First Page

368

Last Page

381

ISBN

978-3-319-47873-9

Identifier

10.1007/978-3-319-47874-6_26

Publisher

Springer

City or Country

Seattle, USA

Additional URL

http://doi.org./10.1007/978-3-319-47874-6_26

Included in

Social Media Commons

Share

COinS