Master Thesis MSTR-2019-70

BibliographyRamzy, Nour: Ontology-based Product Catalog Exploration.
University of Stuttgart, Faculty of Computer Science, Electrical Engineering, and Information Technology, Master Thesis No. 70 (2019).
101 pages, english.

The shift towards Industry 4.0 paradigm drives companies to communicate and network to share knowledge and expertise about digitization. This can be achieved within the frame of collaboration projects and consortia. For instance, Productive4.0 is a European funded project by H2020/ECSEL that aims at creating a user platform across value chains and industries, thus generating a network of communication between manufacturing companies, production machines, and products. The Open Online Sales and Marketing Platform (OOSMP) as part of this project, is a platform for product catalog exploration. The user is guided through a step by step process to explore variants and options of products in the catalog. In this thesis, we identify several problems with the current implementation of the platform hindering a thorough product exploration. These problems affect the users directly as well as the developers of the platform. After analysis, we entail that these hurdles boil down to the missing links between entities on the platform. Concepts of the platform are not sufficiently interconnected which leads to a cumbersome and possibly inefficient product exploration. Also, the step by step process for product exploration is rigid, consequently the user cannot optimally benefit from the platform. We propose a Semantic Web-based solution to the identified problems for the online sales platform. We first contribute with a solution architecture that consists of two main pillars of Semantic Web, namely, rule-based reasoning and semantic search. We design a solution that relies on an ontology-based knowledge representation that enables reasoning leading to the inference of knowledge. This helps in overcoming the problem of missing relations between entities of the platform. This structure also allows the implementation of semantic search to enrich the results shown to the user on the platform with semantically related entities. The proposed solution consists of an offline stage to prepare the knowledge representation as well as an online phase that enables the user via a search functionality to query using the new inferred knowledge. In fact, we analyze the state of the art for both parts of the solution architecture to determine the approaches and tools we adopt for the design. We adapt this solution to Infineon's use case and take into consideration existing practices and technical restrictions. As a final step of this work, we evaluate the solution using different metrics to validate that the Semantic Web-based architecture contributes to solving different hurdles for the platform. The solution implemented proves that the Semantic Web is a new technology that enhances the product catalog exploration of the online sales platform, which guarantees a competitive edge by allowing richer product exposure and better user experience.

Department(s)University of Stuttgart, Institute of Parallel and Distributed Systems, Applications of Parallel and Distributed Systems
Superviser(s)Herschel, Prof. Melanie; Ehm, Hans
Entry dateFebruary 19, 2020
   Publ. Computer Science