|
模式挖掘
Mining Sets of Patterns |
|
| |
| 课程网址: |
http://videolectures.net/ecmlpkdd2010_bringmann_nijssen_vreeken_m...
|
| 主讲教师: |
Siegfried Nijssen; Jilles Vreeken; Bjorn Bringmann
|
| 开课单位: |
鲁汶大学
|
| 开课时间: |
2010-11-16
|
| 课程语种: |
英语
|
| 中文简介: |
模式挖掘是数据挖掘中最重要的课题之一。其核心思想是提取描述数据库各部分的相关知识 "金块"。然而, 许多传统 (频繁) 模式挖掘算法发现的模式的数字太大, 没有实际价值: 如此多的知识的 "金块" 被发现, 它们并没有结合到对数据的更好的全球理解中。事实上, 发现的模式的数量往往大于原始数据库的大小! 为了解决这一问题, 近年来开发了许多技术, 以发现不是全部, 而是有用的一套模式。本教程的目的是提供对挖掘此类高质量模式集的最新技术的全面概述。 在本教程中, 一个重要的重点是可以挖掘模式的任务, 以及这些任务如何影响模式挖掘和模式选择过程。我们区分在无监督环境中开采的模式 (其中模式旨在提供数据的描述) 和在监督环境中开采的模式 (通常对以后在预测模型中使用模式。这两类问题都有不同的问题, 使我们能够讨论在分类和探索数据挖掘中使用模式的可能性和力量。 本教程的主要贡献是: * 我们全面概述了模式集挖掘技术。 * 我们深入探索经典机器学习任务和最近的模式集挖掘技术是如何相关的。 * 我们澄清了各种模式和模式集发现算法之间的联系。 本教程的目的是提供近年来研究的主要想法的大致概况;它将为对更深入的细节感兴趣的研究人员和数据挖掘从业人员提供参考资料的概述。
|
| 课程简介: |
Pattern mining is one of the most important topics in data mining. The core idea is to extract relevant 'nuggets' of knowledge describing parts of a database. However, many traditional (frequent) pattern mining algorithms find patterns in numbers that are much too large to be of practical value: so many 'nuggets' of knowledge are found that they do not combine into a better global understanding of the data. In fact, the number of discovered patterns is often larger than the size of the original database! To tackle this problem, in recent years many techniques have been developed for finding not all, but useful sets of patterns. The aim of this tutorial is to provide a general, comprehensive overview of the state-of-the-art of mining such high-quality sets of patterns. In the tutorial, an important focus is on the tasks for which patterns can be mined, and how these tasks can influence both the pattern mining and pattern selection process. We make a distinction between patterns mined in an unsupervised setting, where patterns are intended to provide a description of the data and are often the end result of the mining process, and those mined in a supervised setting, where one usually is interested in the later use of patterns in predictive models. Both classes of problems come with distinctive problems, and allow us to discuss the possibilities and powers of using patterns for both classification and explorative data mining. The main contributions of our tutorial will be that: * we give a comprehensive overview of pattern set mining techniques. * we thoroughly explore how classic machine learning tasks and recent pattern set mining techniques are related. * we clarify the connections between a wide range of pattern and pattern set discovery algorithms. The aim of the tutorial is to provide a broad overview of the key ideas studied in recent years; it will provide an overview of references for researchers and data mining practitioners interested in the more in-depth details.
|
| 关 键 词: |
模式挖掘; 监督; 模式选择
|
| 课程来源: |
视频讲座网
|
| 最后编审: |
2020-07-06：heyf
|
| 阅读次数: |
23
|
|
图 书 馆