Applications of Text Generation through Semi-Supervised Learning and other state-of-the-art Machine Learning approaches

Overview:

The project explores semi-supervised learning frameworks with applications of text generation using deep learning models. This project addresses several research questions, such as the development of an approach for automated labeling of data generated via chatbot interactions, the integration of user feedback for an enhanced learning experience for the chatbot, and the improvement of chatbot responses where only a limited dataset is available for training. During the course of this research, the needs of other SAP IT applications will be analyzed with respect to state-of-the-art methods in semi-supervised learning and Natural Language Processing. Improvements and extensions will be performed accordingly. The findings of the chatbot use case should also be applied within other domains or use cases within SAP.

For the initial cycle of this project, the focus is on a HR chatbot use-case where a semi-supervised learning framework is used to generate response for user utterance. A human-in-the-loop is also embedded into the semi-supervised learning framework to correct the generated responses before they are appended to the training data.  

Motivation

• SAP researches on various domain-specific Conversational Agents to facilitate an easier access to information for it’s employees.
• Conversational Systems require labelled question/answer pairs which is a time-consuming task for domain experts. 
• A Semi-Supervised Framework facilitates unlabeled queries through a Text Generation Model using the domain knowledge.

Research Objective

• This project aims at the generation of human-like and coherent chatbot responses that accurately addresses the user query.
• Incorporation of human feedback cycle and generated responses into the training data through a semi-supervised framework.

Research Questions

• Which NLP models would work best for the domain-specific generative Question Answering?

• Would an SSL framework along with Human in the loop generate high quality labelled data to be used for model training?

Research Partner:

This project is part of the SAP @ TUM Collaboration Lab and hence fosters a close research partnership with SAP Intelligent Enterprise Solutions and Artificial Intelligence Center of Excellence