Efficient hybrid reduction for binary based information system in soft set theory
In soft set literatures, issues regarding reduction techniques with regards to dataset in soft set have been discussed and analyzed. The existing reduction techniques discussed were the techniques based on rough set guidelines and parameter reduction. All of the proposed techniques have successfully...
Saved in:
| Main Author: | |
|---|---|
| Format: | Thesis |
| Published: |
2016
|
| Subjects: | |
| Online Access: | http://eprints.uthm.edu.my/9213/ http://eprints.uthm.edu.my/9213/1/Ahmad_Nazari_Mohd_Rose.pdf |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | In soft set literatures, issues regarding reduction techniques with regards to dataset in
soft set have been discussed and analyzed. The existing reduction techniques
discussed were the techniques based on rough set guidelines and parameter
reduction. All of the proposed techniques have successfully reduced the datasets but
the factors of consistency and accuracy are still outstanding. Based on the research
done on data transformation in soft set theory, the three newly introduced reduction
methods will be integrated into a technique known as Hybrid Reduction in Soft Set
(HRSS). HRSS consists of two(2) types of parameter reduction and a newly
proposed object reduction. The proposed technique has been implemented and the
results were compared to the existing techniques, and HRSS was found to be 100%
consistent, accurate and able to reduce the data substantially. With SRR (Soft Set
Rough Reduction) and Parameter Reduction (PR) being ineffective with respect to
consistency and accuracy, further analysis on the data size achieved by HRSS and
Normal Parameter Reduction (NPR) were then considered. HRSS has also
demonstrated efficiency when searching for decisional values. Lastly, HRSS has also
been found to be the least complexed in terms of the algorithm used. With the
results obtained, it is safe to conclude that, decision-making that are based on
selected datasets that have undergone the HRSS processing is competent. |
|---|