Similarity model based on CBR and FCA
- Publication Type:
- Conference Proceeding
- Software Engineering, Artificial Intelligence, Networking and Parallel/Distributed Computing (SNPD), 2016 17th IEEE/ACIS International Conference on, 2016
- Issue Date:
Copyright Clearance Process
- Recently Added
- In Progress
- Closed Access
This item is closed access and not available.
Case-based reasoning (CBR) is one of the research highlights in the artificial intelligence field. In the process of case retrieval of CBR, Similarity is an important index in evaluation. This paper proposed a new model calculating similarity between source case and target case. The model is suitable for using Formal Concept Analysis (FCA) in the case of CBR. The model considered the case attributes weights between two formal concepts and feature attributes weights in concept lattices. Comparing to the similarity model put forward by Jirapond Tadrat, this model cut down the comparison to the objects, weight more in the attributes and shorten the time of solving the similarity. Theoretical deduction proves that the proposed similarity model satisfy the basic conditions which all these models need to meet. This article chose the UCI data sets and the method of cross validation, carried on an experiment from both similarity model aspect and classifier aspect respectively. The former experimental results show that the similarity model has higher accuracy than others. The latter experimental results show that the similarity model of CBR classifier has higher accuracy in the attribute set density compared with other small data set classifier.
Please use this identifier to cite or link to this item: