AI-driven Business Configuration - Recommender System

Thesis (MA)

Advisor(s):  Philipp Landler, Simon Fuchs & SAP SE

Context

Customizing Parameters determine the target state of S/4HANA / S/4HANA Cloud that is individually required by each customer. The target state represents a combination of solution capabilities that enable and determine the document flow in the S/4HANA system. The document flow represents the business processes of a company.

Building on an existing data analysis pipeline for the semantic analysis of customizing parameters used in S/4HANA / S/4HANA Cloud systems, a recommender system shall be developed. The recommender system shall create recommendations for constellations of customizing parameters based on frequent use within a particular industry segment.

Task(s)

  • Get familiar with the existing data analysis pipeline (data extraction, data analysis, pattern identification, clustering procedure).
  • Understand the clustering procedure and, if needed, adjust it to transform cluster categories into commonly used and recognizable business terms that can be used as input for a recommender system.
  • Implement a recommender system that uses the clusters semantics to create recommendation for frequently used customization parameters within particular industry segments.
  • Develop a concept on the explainability of the results.
  • Test the pipeline and recommender system on customer data.

Requirements

  • High degree of autonomy and individual responsibility
  • Good communication and technical skills
  • Interest in machine learning and recommender systems
  • Experience in SAP S/4HANA and business (software) understanding are beneficial

Further Information

The thesis can be written in English or German. The start of this thesis is planned for May/June 2023. The thesis is in cooperation with SAP SE. It is possible to get a working student contract at SAP during the thesis. If you have further questions, please do not hesitate to contact me directly. Please send your application including our application form, a current transcript of records, and your CV to philipp.landler@tum.de and s.t.fuchs@tum.de. Please note that we can only consider applications with complete documents.