Using Sampling Method in Privacy Preserving of Association Rule Mining
碩士 === 元智大學 === 資訊管理研究所 === 92 === For reasons of preserving data privacy and obtaining correct mining results, algorithms on privacy preserving of association rule mining are often complicated, and the mining processes cost a lot of time. Our research provided an experiment on using samp...
| Main Authors: | , |
|---|---|
| Other Authors: | |
| Format: | Others |
| Language: | zh-TW |
| Published: |
2004
|
| Online Access: | http://ndltd.ncl.edu.tw/handle/29803152194423968193 |
| Summary: | 碩士 === 元智大學 === 資訊管理研究所 === 92 === For reasons of preserving data privacy and obtaining correct mining results, algorithms on privacy preserving of association rule mining are often complicated, and the mining processes cost a lot of time.
Our research provided an experiment on using sampling method in privacy preserving of association rule mining. Using sampling data instead of the entire dataset can reduce the quantity of dataset, speed up mining processes. In our experimental results, we find that using sampling method can get satisfied results on both precision and recall(over 80%) when we use enough sample size(5%).
|
|---|