Automatic Semantic Mapping of Complex OpenAPI Datatypes

June 22nd, 2022

The OpenAPI specification is a community-driven open standard for describing and documenting RESTful APIs. The OpenAPI specification is a way to describe APIs in a way that is both machine-readable and human-readable. Even complex data types can be mapped using inheritance and polymorphism. However, an automatic mapping from natural language into these complex data types proves to be difficult.

Thesis Type Master
Status Open
Supervisor(s) Stefan Decker
Ralf Klamma
Advisor(s) Alexander Neumann

Natural language communication hides complex data structures that cannot be passed directly as arguments.
Using Open-Source-Software (OSS) such as Rasa, it is possible to extract simple intents and entities from sentences.
It has also been shown that complex structures of Web ontologies can be transferred to OpenAPI specifications [1].
Now it is possible for developers to query complex data using API [2], but this is still a problem for end users of an application.
In this thesis, the student should develop algorithm, which takes a semantic mapping of complex input data in chats and transfers it to functions of an OpenAPI interface.
This should be represented in a modelable way within the Social Bot Framework to enable not only standard data types but also complex data types.

If you are interested, send your recent transcript and a comprehensive CV to


  1. Espinoza-Arias, P., Garijo, D., & Corcho, O. (2020). Mapping the Web Ontology Language to the OpenAPI Specification. In Lecture Notes in Computer Science (pp. 117–127). Springer International Publishing.
  2. Garijo, D., & Osorio, M. (2020). OBA: An Ontology-Based Framework for Creating REST APIs for Knowledge Graphs. In Lecture Notes in Computer Science (pp. 48–64). Springer International Publishing.

  • OpenAPI Specifiaction, RESTful Web services
  • Java
  • Semantic Web