Abstract
Knowledge reduction is one of the key issues in knowledge discovery and data mining. During the construction of a concept lattice, it has been recognized that computational complexity is a major obstacle in deriving all the concept from a database. In order to improve the computational efficiency, it is necessary to preprocess the database and reduce its size as much as possible. Focusing on formal fuzzy contexts, we introduce in the paper the notions of granular consistent sets and granular reducts and propose granular reduct methods in the sense of reducing the attributes. With the proposed approaches, the attributes that are not essential to all the object concepts can be removed without loss of knowledge and, consequently, the computational complexity of constructing the concept lattice is reduced. Furthermore, the relationship between the granular reducts and the classification reducts in a formal fuzzy context is investigated.
| Original language | English |
|---|---|
| Pages (from-to) | 156-166 |
| Number of pages | 11 |
| Journal | Knowledge-Based Systems |
| Volume | 114 |
| Early online date | 8 Oct 2016 |
| DOIs | |
| Publication status | Published - 15 Dec 2016 |
| Externally published | Yes |
Bibliographical note
The authors are very indebted to the anonymous referees for their critical comments and suggestions for the improvement of this paper. This work was supported by grants from the National Natural Science Foundation of China (Nos. 61272021 , 61363056 , 71371063 , 61573321 , 41631179 , 61673396 ), the National Social Sci- ence Foundation of China (No.14XXW004), the Fundamental Re- search Funds for the Central Universities (No.15CX02119A), and the open project of Key Laboratory of Oceanographic Big Data Mining & Application of Zhejiang Province (No.OBDMA201504).Keywords
- Concept lattice
- Crisp-fuzzy concept
- Granular reduct
- Ordered relation