Developing a Social Customer Service Application

OData support
Dr. Martinek Péter
Department of Electronics Technology

The increasing popularity and spread of social media and social networks has changed how people interact with companies and each other. Customers tend to use these new channels, like Twitter or Facebook, to get in touch with companies.

The aim of my thesis is to design and develop a customer support application which allows solving customer issues using a centralized application. Customers may contact the company in three different ways: by sending e-mail messages, by using Twitter, or by asking questions on the customer support website.

The application has a monitoring system which allows the company to monitor Twitter in order to collect and analyze messages related to the company. This monitoring system is supported by an automated classifier which detects the sentiment of each message and sets its class to positive, negative or neutral. The classifier utilizes machine learning algorithms for this task. With the monitoring system, companies can discover which messages contain feedback or problems so they can handle them.

In the first part of my thesis I describe the key points of social customer service, then I give an overview of the field of sentiment analysis. Next, I give a brief introduction about technologies I used for developing the customer service application. This chapter is followed by the business requirements and the technical specification. The next section is about sentiment analysis, where I present the design and development process of the sentiment classifier.

In the second part of my thesis I outline the database schema and the entity data model. The last chapters are about the design and implementation of the data collection modules and the web applications. As a conclusion I share my opinion and results of my work and about the technologies I used.

The result of my work is a social customer service application which provides a centralized solution to manage all customer conversations. It has a modern and user friendly interface to support the daily work of customer service employees. The application also helps the customers to gather information and get support from the company.


Please sign in to download the files of this thesis.