R and Data Mining - Hardcover

Zhao PhD, Yanchang

Hardcover

ISBN 10: 0123969638 ISBN 13: 9780123969637

Publisher: Academic Press, 2012

Synopsis

R and Data Mining introduces researchers, post-graduate students, and analysts to data mining using R, a free software environment for statistical computing and graphics. The book provides practical methods for using R in applications from academia to industry to extract knowledge from vast amounts of data. Readers will find this book a valuable guide to the use of R in tasks such as classification and prediction, clustering, outlier detection, association rules, sequence analysis, text mining, social network analysis, sentiment analysis, and more.

Data mining techniques are growing in popularity in a broad range of areas, from banking to insurance, retail, telecom, medicine, research, and government. This book focuses on the modeling phase of the data mining process, also addressing data exploration and model evaluation.

With three in-depth case studies, a quick reference guide, bibliography, and links to a wealth of online resources, R and Data Mining is a valuable, practical guide to a powerful method of analysis.

• Presents an introduction into using R for data mining applications, covering most popular data mining techniques
• Provides code examples and data so that readers can easily learn the techniques
• Features case studies in real-world applications to help readers apply the techniques in their work

About this title

From the Author

Table of Contents:

1 Introduction   
    1.1 Data Mining
    1.2 R
    1.3 Datasets
        1.3.1 The Iris Dataset
        1.3.2 The Bodyfat Dataset

2 Data Import and Export
    2.1 Save and Load R Data
    2.2 Import from and Export to .CSV Files
    2.3 Import Data from SAS
    2.4 Import/Export via ODBC
        2.4.1 Read from Databases
        2.4.2 Output to and Input from EXCEL Files

3 Data Exploration
    3.1 Have a Look at Data
    3.2 Explore Individual Variables
    3.3 Explore Multiple Variables
    3.4 More Explorations
    3.5 Save Charts into Files

4 Decision Trees and Random Forest
    4.1 Decision Trees with Package party
    4.2 Decision Trees with Package rpart
    4.3 Random Forest

5 Regression
    5.1 Linear Regression
    5.2 Logistic Regression
    5.3 Generalized Linear Regression
    5.4 Non-linear Regression

6 Clustering
    6.1 The k-Means Clustering
    6.2 The k-Medoids Clustering
    6.3 Hierarchical Clustering
    6.4 Density-based Clustering

7 Outlier Detection
    7.1 Univariate Outlier Detection
    7.2 Outlier Detection with LOF
    7.3 Outlier Detection by Clustering
    7.4 Outlier Detection from Time Series
    7.5 Discussions

8 Time Series Analysis and Mining
    8.1 Time Series Data in R
    8.2 Time Series Decomposition
    8.3 Time Series Forecasting
    8.4 Time Series Clustering
        8.4.1 Dynamic Time Warping
        8.4.2 Synthetic Control Chart Time Series Data
        8.4.3 Hierarchical Clustering with Euclidean Distance
        8.4.4 Hierarchical Clustering with DTW Distance
    8.5 Time Series Classification
        8.5.1 Classification with Original Data
        8.5.2 Classification with Extracted Features
        8.5.3 k-NN Classification
    8.6 Discussions
    8.7 Further Readings

9 Association Rules
    9.1 Basics of Association Rules
    9.2 The Titanic Dataset
    9.3 Association Rule Mining
    9.4 Removing Redundancy
    9.5 Interpreting Rules
    9.6 Visualizing Association Rules
    9.7 Discussions and Further Readings

10 Text Mining
    10.1 Retrieving Text from Twitter
    10.2 Transforming Text
    10.3 Stemming Words
    10.4 Building a Term-Document Matrix
    10.5 Frequent Terms and Associations
    10.6 Word Cloud
    10.7 Clustering Words
    10.8 Clustering Tweets
        10.8.1 Clustering Tweets with the k-means Algorithm
        10.8.2 Clustering Tweets with the k-medoids Algorithm
    10.9 Packages, Further Readings and Discussions

11 Social Network Analysis
    11.1 Network of Terms
    11.2 Network of Tweets
    11.3 Two-Mode Network
    11.4 Discussions and Further Readings

12 Case Study I: Analysis and Forecasting of House Price Indices
    12.1 Importing HPI Data
    12.2 Exploration of HPI Data
    12.3 Trend and Seasonal Components of HPI
    12.4 HPI Forecasting
    12.5 The Estimated Price of a Property
    12.6 Discussion

13 Case Study II: Customer Response Prediction and Profit Optimization
    13.1 Introduction
    13.2 The Data of KDD Cup 1998
    13.3 Data Exploration
    13.4 Training Decision Trees
    13.5 Model Evaluation
    13.6 Selecting the Best Tree
    13.7 Scoring
    13.8 Discussions and Conclusions

14 Case Study III: Predictive Modeling of Big Data with Limited Memory
    14.1 Introduction
    14.2 Methodology
    14.3 Data and Variables
    14.4 Random Forest
    14.5 Memory Issue
    14.6 Train Models on Sample Data
    14.7 Build Models with Selected Variables
    14.8 Scoring
    14.9 Print Rules
        14.9.1 Print Rules in Text
        14.9.2 Print Rules for Scoring with SAS
    14.10 Conclusions and Discussion

15 Online Resources
    15.1 R Reference Cards
    15.2 R
    15.3 Data Mining
    15.4 Data Mining with R
    15.5 Classification/Prediction with R
    15.6 Time Series Analysis with R
    15.7 Association Rule Mining with R
    15.8 Spatial Data Analysis with R
    15.9 Text Mining with R
    15.10 Social Network Analysis with R
    15.11 Data Cleansing and Transformation with R
    15.12 Big Data and Parallel Computing with R

About the Author

Dr. Yanchang Zhao is a Senior Data Mining Specialist in Australian public sector. Before joining public sector, he was an Australian Postdoctoral Fellow (Industry) at University of Technology, Sydney from 2007 to 2009. He is the founder of the RDataMining.com website and an RDataMining Group on LinkedIn. He has rich experience in R and data mining. He started his research on data mining since 2001 and has been applying data mining in real-world business applications since 2006. He has over 50 publications on data mining research and applications, including three books. He is a senior member of IEEE, and has been a Program Chair of the Australasian Data Mining Conference (AusDM 2012 & 2013) and a program committee member for more than 50 academic conferences.

Edition details

• Publisher: Academic Press
• Publication date: 2012
• ISBN 10: 0123969638
• ISBN 13: 9780123969637
• Binding: Hardcover
• Language: English
• Edition number: 1
• Number of pages: 256