Customer Feedback Analytics Platform

This project took place in summer term 2021, you CAN NOT apply to this project anymore!

Results of this project are explained in detail in the final documentation .

Currently, customer feedback data is often stored in many silos and analysed via wordclouds or other means to find the most frequent observations. The insights from this lack derived actions and miss the point of identifying upcoming trends and single events. A platform should be able to deal with this on the continuous stream of customer feedback.

The platform’s core capabilities should include detecting topics as well as entities (NER) and sentiment and in particular attributes and products of relevance. Available APIs allow to easily tag texts with topics or entities but are usually trained on general public data (newspaper, Wikipedia, etc.) and hence produce often too many unspecific tags. Building and optimizing an own model (such as LSTM-based variants) should allow us to overcome this issue.

The task for this project is to design and implement a customer feedback analysis platform, which is capable to store, analyse and visualize the data to deliver actionable insights for the user. This includes a Design Thinking process in which we first collect and prioritize the needs of our potential users in workshops and then work in sprints on the derived product vision.

Your team skills should cover the fields of Machine Learning, Data Engineering and Visualization to build a production-ready prototype during the semester. Knowledge about cloud technologies as well as about analysis of textual data is advantageous for this project. You will get the chance to contribute to a software product development for a real-world problem. We will share our methods with you to solve problems in practice, i.e., apply Design Thinking, gather & prioritize the right features that meet the needs of the users and to work in an agile project setting to develop a product. In doing so, you will get hands-on experience with the Google Cloud Platform, evaluate & improve state-of-the-art NLP tools, and develop your own NLP models.

Accepted students to this project should attend (unless they have proven knowledge) online workshops at the LRZ from 06.04.2021 - 09.04.2021 (9:00 AM to 5:00 PM). More information will be provided to students accepted to this project.