Hyspirit-implementation of a 4valued logic based IR model.
Reserved bachelor thesis
- Targeted audience
- AI Bachelor
- Ability to read and understand papers written in English
- Java, Python(Preferred)
- Lecture "Internet-Suchmaschinen"(Preferred)
Hyspirit is an information retrieval tool used to develop customized IR models. The main idea of this work will be developing a Hyspirit interface to retrieve search items based on a model that makes use of the 4 valued logic as a basis to define the relevance score between the user queries and search items.
This model will be applied on search items that have user reviews e.g amazon products, hotels, ...etc. By using this model, the relevance score will be defined as four values. These values are calculated based on the matching between the user query and the reviews of the products.
The final thesis includes the following phases:
- Exploring Hyspirt and trying to implement well-known models like tfidf.
- Exploring the 4 valued logic.
- Developing the Hyspirt interface of the 4valued logic based model.
- HySpirit - A probabilistic inference engine for hypermedia retrieval in large databases
- Retrieving complex objects with hyspirit
- Internet-Suchmaschinen Skript
- http://www.is.inf.uni-due.de/courses/ir_ss14/ISMs_1-7.pdf , Prof. Dr.-Ing. Norbert Fuhr, 8. September 2014
- Belnap's four valued logic