Keyword-based similarity using an automatically generated semantic graph in an online Community of Practice
KeywordsCommunity of Practice
Case Based Reasoning
[INFO.INFO-AI] Computer Science [cs]/Artificial Intelligence [cs.AI]
Full recordShow full item record
One of the concepts involving learning through interactions is Com-munities of Practice (CoPs). They are based on the idea of improving practice and enhancing members learning in a supportive environment through mutually shared interests and goals. This type of communities allows maintaining knowledge in a community memory as they evolve. Pertinent reuse of this knowledge would facilitate learning among the CoP members, increase their productivity and also improve the quality of their artefacts. In our online CoP environment, we include knowledge reuse based on the Case Based Reasoning (CBR) approach as one of the main system functions, aiming at capitalizing the community knowledge. In fact, when a CoP member encounters a new problem, the first phase of the CBR cycle consists of retrieving previously experienced cases that are similar to the new problem. In this paper, we propose a keywords-based similarity using a semantic net-work that contains all potential keywords of the CoP’s domain of interest, orga-nized semantically. We also present our approach for automatically generating this semantic graph based on content extracted from an external source: the Wik-ipedia knowledge base.