智能数据分析模型与算法：第二版实用教程

需积分: 0 144 浏览量更新于2024-07-17 收藏 4.87MB PDF 举报

《数据分析：智能数据处理的模型与算法》是一本面向工程、计算机科学和数学专业本科生和研究生的教材，也适用于从事数据分析项目的实践者。该书的结构紧密围绕实际的数据分析项目，强调理论与实践相结合，对基本数学知识的需求较少。作者Thomas A. Runkler以其在德国慕尼黑工业大学多年教学经验为基础，结合西门子等工业研究开发项目的成果编撰而成。本书是第二版，特别关注于数据分析中的关键模型和算法，包括但不限于数据分类和机器学习技术。它涵盖了广泛的实用内容，适合教学环境中的课程教学，如课堂讲授、研讨会和国际学术会议的教程。此外，它还提供了丰富的实例和案例研究，以便学生和专业人士能够理解和应用所学知识。 ISBN-978-3-658-14074-8和ISBN-978-3-658-14075-5分别对应纸质版和电子版，doi:10.1007/978-3-658-14075-5表明该书已被收录在Springer Vieweg出版平台，获得了2016年美国国会图书馆的Control Number 2016942272，表明其版权受到保护。书中明确指出，尽管通用描述性名称、注册商标、服务标记等的使用可能不需明确声明，但这并不意味着它们不受相关法律保护。《数据分析：智能数据处理的模型与算法》是一本具有深厚工业背景和广泛适用性的权威教材，对于提升读者在数据驱动决策和创新中的技能至关重要。

References 3

standardization

transformation

analysis

visualization

correlation

regression

forecasting

classification

clustering

postprocessing

interpretation

documentation

preparation

planning

data collection

feature generation

data selection

preprocessing

cleaning

filtering

completion

correction

evaluation

Fig. 1.1 Phases of data analysis projects

This book gives a clear and concise overview of the most important methods and

algorithms of data analysis. It enables the reader to gain a complete and comprehensive

understanding of data analysis, to apply data analysis methods to her or his own projects,

and to contribute to the progress of the ﬁeld.

A large number a software tools for data mining are available today. Popular commer-

cial or free software tools include KNIME, MATLAB, SAS, SPSS, STATISTICA, TIBCO

Spotﬁre, R, Rapid Miner, Tableau, QlikView, oder WEKA. This book does not present,

compare, or recommend any data mining software tools. For a comprehensive overview of

current data mining software tools please refer to [4, 7].

References

1. U. M. Fayyad, G. Piatetsky-Shapiro, P. Smyth, and R. Uthurusamy. Advances in Knowledge

Discovery and Data Mining. AAAI Press, Menlo Park, 1996.

2. R. Kohavi, N. J. Rothleder, and E. Simoudis. Emerging trends in business analytics. Communi-

cations of the ACM, 45(8):345–48, 2002.

3. M. C. Lovell. Data mining. Review of Economics and Statistics, 65(1):1–11, 1983.

4. R. Mikut and M. Reischl. Data mining tools. Wiley Interdisciplinary Reviews: Data Mining and

Knowledge Discovery, 1(5):431–443, 2011.

5. C. Shearer. The CRISP-DM model: The new blueprint for data mining. J Data Warehousing,

5(4):13–22, 2000.

6. S. Tyagi. Using data analytics for greater proﬁts. Journal of Business Strategy, 24(3):12–14, 2003.

7. L. Zhang, A. Stoffel, M. Behrisch, S. Mittelstadt, T. Schreck, R. Pompl, S. Weber, H. Last,

andD.Keim. Visualanalyticsforthebigdataera—acomparative review of state-of-the-

art commercial systems. In IEEE Conference on Visual Analytics Science and Technology,

pages 173–182, Seattle, 2012.

2.1 The Iris Data Set 7

Table 2.1 (continued)

Setosa Versicolor Virginica

Sepal Petal Sepal Petal Sepal Petal

Length Width Length Width Length Width Length Width Length Width Length Width

5:5 3:5 1:3 0:2 6:7 3:1 4:7 1:5 6:3 3:4 5:6 2:4

4:9 3:6 1:4 0:1 6:3 2:3 4:4 1:3 6:4 3:1 5:5 1:8

4:4 3 1:3 0:2 5:6 3 4:1 1:3 6 3 4:8 1:8

5:1 3:4 1:5 0:2 5:5 2:5 4 1:3 6:9 3:1 5:4 2:1

5 3:5 1:3 0:3 5:5 2:6 4:4 1:2 6:7 3:1 5:6 2:4

4:5 2:3 1:3 0:3 6:1 3 4:6 1:4 6:9 3:1 5:1 2:3

4:4 3:2 1:3 0:2 5:8 2:6 4 1:2 5:8 2:7 5:1 1:9

5 3:5 1:6 0:6 5 2:3 3:3 1 6:8 3:2 5:9 2:3

5:1 3:8 1:9 0:4 5:6 2:7 4:2 1:3 6:7 3:3 5:7 2:5

4:8 3 1:4 0:3 5:7 3 4:2 1:2 6:7 3 5:2 2:3

5:1 3:8 1:6 0:2 5:7 2:9 4:2 1:3 6:3 2:5 5 1:9

4:6 3:2 1:4 0:2 6:2 2:9 4:3 1:3 6:5 3 5:2 2

5:3 3:7 1:5 0:2 5:1 2:5 3 1:1 6:2 3:4 5:4 2:3

5 3:3 1:4 0:2 5:7 2:8 4:1 1:3 5:9 3 5:1 1:8

Notice that several distinct replicates of the original Iris data set have been used and

published, so in experiments with this data set the version should be carefully checked [2].

The Iris data set as well as many other popular data sets are available, for example, through

the machine learning data base at the University of California at Irvine (UCI).

In data analysis we call each of the 150 Iris ﬂowers an object, each of the three species

a class, and each of the four dimensions a feature. Here is a list of typical questions that

we try to answer by data analysis:

• Which of the data might contain errors or false class assignments?

• What is the error caused by rounding the data off to one decimal place?

• What is the correlation between petal length and petal width?

• Which pair of dimensions is correlated most?

• None of the ﬂowers in the data set has a sepal width of 1.8 cm. Which sepal length

would we expect for a ﬂower that did have 1.8 cm as its sepal length?

• Which species would an Iris with a sepal width of 1.8 cm belong to?

• Do the three species contain sub-species that can be identiﬁed from the data?

In this book you will ﬁnd numerous methods and algorithms to answer these and other

data analysis questions. For a better understanding of these data analysis methods and

algorithms we ﬁrst deﬁne and examine the fundamental properties of data and their

relations.

8 2 Data and Relations

2.2 Data Scales

Numerical measurements may have different semantic meanings, even if they are rep-

resented by the same numerical data. Depending on the semantic meaning different

types of mathematical operations are appropriate. For the semantic meaning of numerical

measurement Stevens [7] suggested the four different scales that are shown in Table 2.2.

For nominal scaled data (bottom row) only tests for equality or inequality are valid.

Examples for nominal features are names of persons or codes of objects. Data of a nominal

feature can be represented by the mode which is deﬁned as the value that occurs most

frequently. For ordinal scaled data (third row) the operations “greater than” and “less than”

are valid. For each scale level the operations and statistics of the lower scale levels are also

valid, so for the ordinal scale we have equality, inequality, and the combinations “greater

than or equal” () and “less than or equal” (). The relation “less than or equal” ()

deﬁnes a total order, such that for any x; y; z we have .x  y/ ^ .y  x/ ) .x D y/

(antisymmetry), .x  y/ ^ .y  z/ ) .x  z/ (transitivity), and .x  y/ _ .y  x/

(totality). Examples for ordinal features are school grades. Data of an ordinal feature can

be represented by the median which is deﬁned as the value for which (almost) as many

smaller as larger values exist. The mean is not valid for ordinal features, so for example

it does not make sense to say that the average school grade is C. For interval scaled data

(second row) addition and subtraction are valid. Interval scaled features have arbitrary zero

points. Examples are years in the Anno Domini dating system or temperatures in degrees

Celcius (centigrade) or Fahrenheit, so for example it does not make sense to say that 40

is twice as high as 20

C. Data of an interval scaled feature, for example, a set of values

X Dfx

;:::;x

g, can be represented by the (arithmetic) mean

Nx D

kD1

(2.1)

For ratio scaled data (top row) multiplication and division are valid. Examples for ratio

scaled features are time differences like ages or temperatures on the Kelvin scale. Data of

an interval scaled feature can be represented by the generalized mean

.X/ D

kD1

(2.2)

Table 2.2 Scales for

numerical measurements

Scale Operations Example Statistics

Ratio  = 21 years, 273

K Generalized mean

Interval C  2015 A.D., 20

C Mean

Ordinal > < A, B, C, D, F Median

Nominal D ¤ Alice, Bob, Carol Mode

剩余157页未读，继续阅读

Rochestermark

粉丝: 0
资源: 1

智能数据分析模型与算法：第二版实用教程

Data Analytics with Hadoop: An Introduction for Data Scientists

Big Data Analytics with Spark PDF

Big Data Analytics with Spark 无水印pdf 0分

Big Data Analytics for Healthcare

big data analytics with java 电子档

learning spark: lightning-fast data analytics

大数据与云计算融合技术相关文献

flink+doris

big data analytics

datastreams

最新资源