图 书 馆
MiSoSouP：利用采样和伪维数挖掘有趣的子群MiSoSouP: Mining Interesting Subgroups with Sampling and Pseudodimension |
|
| 课程网址: | http://videolectures.net/kdd2018_riondato_mining_pseudodimension/ |
| 主讲教师: | Matteo Riondato |
| 开课单位: | 二西格玛投资有限合伙 |
| 开课时间: | 2018-11-23 |
| 课程语种: | 英语 |
| 中文简介: |  |
| 课程简介: | “Miso makes a soup loaded with flavour that saves you the hassle of making stock.” – Y. Ottolenghi [19] We present MiSoSouP, a suite of algorithms for extracting highquality approximations of the most interesting subgroups, according to different interestingness measures, from a random sample of a transactional dataset. We describe a new formulation of these measures that makes it possible to approximate them using sampling. We then discuss how pseudodimension, a key concept from statistical learning theory, relates to the sample size needed to obtain an high-quality approximation of the most interesting subgroups. We prove an upper bound on the pseudodimension of the problem at hand, which results in small sample sizes. Our evaluation on real datasets shows that MiSoSouP outperforms state-of-the-art algorithms offering the same guarantees, and it vastly speeds up the discovery of subgroups w.r.t. analyzing the whole dataset. |
| 关 键 词: | 根据不同的有趣度度量; 近似所需的样本量相关; MiSoSouP; 分析整个数据集 |
| 课程来源: | 视频讲座网 |
| 数据采集: | 2023-01-16：cyh |
| 最后编审: | 2023-01-16：cyh |
| 阅读次数: | 24 |