Name
Title | ||
|---|---|---|
Cost-Sensitive Attribute Reduction Algorithm Based on Neighborhood Rough Sets in Incomplete Data. |
Abstract | ||
|---|---|---|
Rough set-Based cost-sensitive attribute reduction is an important issue in data mining and knowledge discovery. At present, cost-sensitive attribute reduction is mainly for nominal data in complete decision table. But for reasons such as cost restrictions or privacy protection, the data is missing and incomplete, which leads to the incompleteness of the decision table. To address this issue, in this paper, we propose an attribute reduction algorithm for incomplete neighborhood decision table with the perspective of cost sensitive learning. Firstly, the neighborhood granularity of each object under the conditional attributes is calculated, and the normal domain is obtained according to the neighborhood granularity. Then, we propose the attribute measure function to characterize the significance of candidate attributes, which takes into consideration the test costs of the attributes but also their associated misclassification costs. On this basis, a cost-sensitive attribute reduction algorithm is developed. Finally, the feasibility of the proposed algorithm is demonstrated through theoretical analysis and example results. |
Year | DOI | Venue |
|---|---|---|
2018 | 10.1109/FSKD.2018.8686863 | ICNC-FSKD |
Field | DocType | Citations |
Decision table,Computer science,Algorithm,Rough set,Knowledge extraction,Artificial intelligence,Granularity,Machine learning | Conference | 0 |
PageRank | References | Authors |
0.34 | 0 | 4 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Yinglong Wang | 1 | 2 | 2.04 |
| Qingzhi Xie | 2 | 0 | 0.34 |
| Wenbin Qian | 3 | 1 | 1.38 |
| Qin Huang | 4 | 30 | 11.60 |