没有合适的资源?快使用搜索试试~ 我知道了~
首页Principles of Data Mining
Principles of Data Mining
需积分: 9 22 下载量 23 浏览量
更新于2023-05-21
收藏 57KB PDF 举报
David J. Hand Department of Mathematics, Imperial College London, London, UK Data mining is the discovery of interesting, unexpected or valuable structures in large datasets. As such, it has two rather different aspects. One of these concerns large-scale, ‘global’ structures, and the aim is to model the shapes, or features of the shapes, of distributions.
资源详情
资源推荐
Drug Safety 2007; 30 (7): 621-622
C
ONFERENCE
P
APER
0114-5916/07/0007-0621/$44.95/0
© 2007 Adis Data Information BV. All rights reserved.
Principles of Data Mining
David J. Hand
Department of Mathematics, Imperial College London, London, UK
Data mining is the discovery of interesting, unexpected or valuable structures
Abstract
in large datasets. As such, it has two rather different aspects. One of these
concerns large-scale, ‘global’ structures, and the aim is to model the shapes, or
features of the shapes, of distributions. The other concerns small-scale, ‘local’
structures, and the aim is to detect these anomalies and decide if they are real or
chance occurrences. In the context of signal detection in the pharmaceutical
sector, most interest lies in the second of the above two aspects; however, signal
detection occurs relative to an assumed background model, therefore, some
discussion of the first aspect is also necessary. This paper gives a lightning
overview of data mining and its relation to statistics, with particular emphasis on
tools for the detection of adverse drug reactions.
“As with all data analysis, data mining is a mining is that with large enough datasets you are
process; it is rarely a case of one step being suffi- bound to find some structures in non-random sam-
cient.” ples of the population. The computer science per-
spective is that even if that is true, it is also certain
A working definition of data mining is “the dis-
that there is valuable information in there. Poor
covery of interesting, unexpected, or valuable struc-
quality data does not deter data miners from tackling
tures in large datasets”.
[1]
the problems they are presented with. The populari-
Drivers behind data mining are of two kinds:
ty of data mining is expected to raise awareness of
scientific and commercial. Problems and intellectual
the importance of collecting good quality data.
challenges stem from both contexts; tools and
software development are especially driven by com-
Traditionally, data mining is a secondary analysis
mercial considerations.
over data collected for some other purpose. For
example, in supermarket data, the information is
Data mining is not a recent discipline. Previously
collected to work out the bill the customer is
it has been called, in a derogatory way, ‘data dredg-
charged. That stored data can then subsequently be
ing’, ‘trawling’ and ‘fishing through data’.
submitted to analysis looking for customers’ trans-
Modern data mining combines statistics with
action patterns. However, the discipline is evolving
ideas, tools and methods from computer science,
and in areas like micro array data and proteomic
machine learning, database technology and other
research, large datasets are collected primarily for
classical data analytical technologies. Classic statis-
discovering patterns, relationships and structures.
tics and data mining differ in several aspects. Statis-
tics is a discipline originally developed throughout
Data mining can be divided into two main classes
the twentieth century based on relatively small
of tools: model building and pattern discovery.
datasets. Computers only came on stream towards
Model building is a high level global descriptive
the end of that century allowing for the manipulation
summary of datasets, which in modern statistics
of larger datasets.
include: regression models, cluster decomposition
Big datasets often have a problem of selectivity and Bayesian networks. Models describe the overall
bias. The classical statistical perspective on data shape of the data. Pattern discovery is what one does
NickName567
- 粉丝: 2
- 资源: 13
上传资源 快速赚钱
- 我的内容管理 收起
- 我的资源 快来上传第一个资源
- 我的收益 登录查看自己的收益
- 我的积分 登录查看自己的积分
- 我的C币 登录后查看C币余额
- 我的收藏
- 我的下载
- 下载帮助
会员权益专享
最新资源
- zigbee-cluster-library-specification
- JSBSim Reference Manual
- c++校园超市商品信息管理系统课程设计说明书(含源代码) (2).pdf
- 建筑供配电系统相关课件.pptx
- 企业管理规章制度及管理模式.doc
- vb打开摄像头.doc
- 云计算-可信计算中认证协议改进方案.pdf
- [详细完整版]单片机编程4.ppt
- c语言常用算法.pdf
- c++经典程序代码大全.pdf
- 单片机数字时钟资料.doc
- 11项目管理前沿1.0.pptx
- 基于ssm的“魅力”繁峙宣传网站的设计与实现论文.doc
- 智慧交通综合解决方案.pptx
- 建筑防潮设计-PowerPointPresentati.pptx
- SPC统计过程控制程序.pptx
资源上传下载、课程学习等过程中有任何疑问或建议,欢迎提出宝贵意见哦~我们会及时处理!
点击此处反馈
安全验证
文档复制为VIP权益,开通VIP直接复制
信息提交成功