Final Thesis: Searching within EDITIVE
Abstract: EDITIVE is a platform based upon the collaboration principles first spread by GitHub. In such a multi-level content collaboration platform, a search functionality is useful and often demanded. The main reasons therefore are to reduce complexity and provide advanced information retrieval, including content meta-information. The difficulty to provide a search functionality within such a platform relates to the underlying content data model and its complexity. Atomic data structures enable more search precision and contribute towards an effective search implementation. EDITIVE runs on a mainly atomically defined data structure. The highly complex data model within EDITIVE stems from the flexibility based on the Git collaboration principles offered to the user, merged with advantages of wikis.
This master thesis shows the difficulties of implementing a search functionality within a multi-level content collaboration platform, more specifically EDITIVE. It presents the design and implementation of it in the EDITIVE context based on the search engine Apache Solr. This thesis shows a reference architecture implemented on Solr. Furthermore, it elaborates on the utilization of the underlying data model. We advise several ways on how to further refine and extend the resulting search implementation beyond the scope of this thesis.
Keywords: EDITIVE, enterprise search, Apache Solr
PDF: Master Thesis
Reference: Vincent Federle. Evaluation and Implementation of a Reference Architecture. Master Thesis. Friedrich-Alexander-Universität Erlangen-Nürnberg: 2021.