Title Data analysis and pattern recognition in multiple databases / by Animesh Adhikari, Jhimli Adhikari, Witold Pedrycz

Published Cham : Springer, [2014] ©2014

Table of Contents
1.	Introduction					1
1.1.	Motivation					1
1.2.	Distributed Data Mining					3
1.3.	Multi-Database Mining Approaches					4
1.3.1.	Local Pattern Analysis					5
1.3.2.	Sampling					6
1.3.3.	Re-Mining					6
1.4.	Pre-Processing of Databases					6
1.4.1.	Preparation of Data Warehouses					7
1.4.2.	Temporal Aggregation					8
1.4.3.	Partitioning Database					8
1.4.4.	Database Thinning					9
1.4.5.	Ordering of Databases					9
1.4.6.	Selection of Databases					9
1.5.	Patterns and Associations					10
1.6.	Related Studies					13
1.7.	Experimental Settings					14
1.8.	Conclusions					14
	References					16
2.	Synthesizing Different Extreme Association Rules from Multiple Databases					21
2.1.	Introduction					21
2.2.	Some Extreme Types of Association Rule in Multiple Databases					23
2.3.	Problem Statement					25
2.4.	An Extended Model of Local Pattern Analysis for Synthesizing Global Patterns					25
2.5.	Related Work					27
2.6.	Synthesizing an Association Rule					29
2.6.1.	Design of Algorithm					30
2.6.2.	Error Calculation					34
2.7.	Experiments					35
2.7.1.	Results of Experimental Studies					37
2.7.2.	Comparison with Existing Algorithm					38
2.8.	Conclusions					40
	References					41
3.	Clustering Items in Time-Stamped Databases Induced by Stability					43
3.1.	Introduction					43
3.2.	Related Work					44
3.3.	A Model of Mining Multiple Transactional Time-Stamped Databases					45
3.4.	Problem Statement					47
3.5.	Clustering Items					49
3.5.1.	Finding the Best Non-Trivial Partition					50
3.5.2.	Finding a Best Class					54
3.6.	Experiments					55
3.7.	Conclusions					58
	References					59
4.	Synthesizing Global Patterns in Multiple Large Data Sources					61
4.1.	Introduction					61
4.2.	Multi-Database Mining Using Local Pattern Analysis					62
4.3.	Generalized Multi-Database Mining Techniques					63
4.3.1.	Local Pattern Analysis					63
4.3.2.	Partition Algorithm					64
4.3.3.	IdentifyExPattern Algorithm					64
4.3.4.	RuleSynthesizing Algorithm					64
4.4.	Specialized Multi-Database Mining Techniques					65
4.4.1.	Mining Multiple Real Databases					65
4.4.2.	Mining Multiple Databases for the Purpose of Studying a Set of Items					66
4.4.3.	Study of Temporal Patterns in Multiple Databases					66
4.5.	Mining Multiple Databases Using Pipelined Feedback Technique					66
4.5.1.	Algorithm Design					68
4.6.	Error Evaluation					68
4.7.	Experiments					69
4.8.	Conclusions					73
	References					73
5.	Clustering Local Frequency Items in Multiple Data Sources					75
5.1.	Introduction					75
5.2.	Related Work					77
5.2.1.	Measures of Association					77
5.2.2.	Multi-Database Mining Techniques					78
5.2.3.	Clustering Techniques					81
5.3.	Problem Statement					82
5.4.	Synthesizing Support of an Itemset					83
5.5.	Clustering Local Frequency Items					86
5.5.1.	Finding the. Best Non-Trivial Partition					89
5.5.2.	Error Analysis					93
5.6.	Experimental Results					94
5.6.1.	Overall Output					96
5.6.2.	Synthesis of High Frequency Itemsets					97
5.6.3.	Error Quantification					99
5.6.4.	Average Error Versus γ					100
5.6.5.	Average Error Versus α					101
5.6.6.	Clustering Time Versus Number of Databases					103
5.6.7.	Comparison with Existing Technique					104
5.7.	Conclusions					106
	References					107
6.	Mining Patterns of Select Items in Different Data Sources					109
6.1.	Introduction					109
6.2.	Mining Global Patterns of Select Items					111
6.3.	Overall Association Between Two Items in a Database					113
6.4.	An Application: Study of Select Items in Multiple Databases Through Grouping					116
6.4.1.	Properties of Different Measures					118
6.4.2.	Grouping Frequent Items					120
6.4.3.	Experiments					123
6.5.	Related Work					127
6.6.	Conclusions					128
	References					128
7.	Synthesizing Global Exceptional Patterns in Different Data Sources					131
7.1.	Introduction					131
7.2.	Exceptional Patterns in Multiple Data Sources					133
7.3.	Problem Statement					139
7.4.	Related Work					139
7.5.	Synthesizing Support of an Itemset					140
7.6.	Synthesizing Type II Global Exceptional Itemsets					141
7.7.	Error Calculation					145
7.8.	Experiments					146
7.8.1.	Comparison with the Existing Algorithm					151
7.9.	Conclusions					153
	References					153
8.	Mining Icebergs in Different Time-Stamped Data Sources					157
8.1.	Introduction					157
8.2.	Related Work					159
8.3.	Notches in Sales Series					160
8.3.1.	Identifying Notches					162
8.4.	Generalized Notch					162
8.5.	Iceberg Notch					163
8.6.	Sales Series					164
8.6.1.	Another View at Sales Series					164
8.6.2.	Other Applications of Icebergs					165
8.7.	Mining Icebergs in Time-Stamped Databases					166
8.7.1.	Non-Incremental Approach					166
8.7.2.	Incremental Approach					169
8.8.	Significant Year					171
8.9.	Experimental Studies					172
8.10.	Conclusions					180
	References					180
9.	Mining Calendar-Based Periodic Patterns in Time-Stamped Data					183
9.1.	Introduction					183
9.2.	Related Work					186
9.3.	Calendar-Based Periodic Patterns					187
9.3.1.	Extending Certainty Factor					188
9.3.2.	Extending Certainty Factor with Respect to Other Intervals					191
9.4.	Mining Calendar-Based Periodic Patterns					192
9.4.1.	Improving Mining Calendar-Based Periodic Patterns					193
9.4.2.	Data Structure					193
9.4.3.	A Modified Algorithm					195
9.5.	Experimental Studies					198
9.5.1.	Selection of Mininterval and Maxgap					202
9.5.2.	Selection of Minsupp					204
9.5.3.	Performance Analysis					205
9.6.	Conclusions					207
	References					207
10.	Measuring Influence of an Item in Time-Stamped Databases					209
10.1.	Introduction					209
10.2.	Association Between Two Itemsets					211
10.3.	Concept of Influence					212
10.3.1.	Influence of an Itemset on Another Itemset					213
10.3.2.	Properties of Influence Measures					214
10.3.3.	Influence of an Item on a Set of Specific Items					215
10.3.4.	Motivation					216
10.4.	Problem Statement					218
10.5.	Related Work					219
10.6.	Design of Algorithms					219
10.6.1.	Designing Algorithm for Measuring Overall Influence of an Item on Another Item					220
10.6.2.	Designing Algorithm for Measuring Overall Influence of an Item on Each of the Specific Items					221
10.6.3.	Designing Algorithm for Identifying Top Influential Items on a Set of Specific Items					221
10.7.	Experiments					222
10.8.	Conclusions					228
	References					228
11.	Summary and Conclusions					231
11.1.	Changing Scenarios					231
11.2.	Summary of Chapters					232
11.3.	Selected Open Problems and Challenges					234
11.4.	Conclusions					234
	References					235
	Index					237

Click on the following:
Springer eBooks
Springer Engineering eBooks

Copies

Description 1 online resource (247 pages) : illustrations Series Intelligent Systems Reference Library, 1868-4394 ; 61 Intelligent systems reference library ; v. 61. Contents From the Contents: Synthesizing Different Extreme Association Rules in Multiple Data Sources -- Clustering items in time-stamped databases induced by stability -- Mining global patterns in multiple large databases -- Clustering Local Frequency Items in Multiple Data Sources -- Mining Patterns of Select Items in Different Data Sources Summary Pattern recognition in data is a well known classical problem that falls under the ambit of data analysis. As we need to handle different data, the nature of patterns, their recognition and the types of data analyses are bound to change. Since the number of data collection channels increases in the recent time and becomes more diversified, many real-world data mining tasks can easily acquire multiple databases from various sources. In these cases, data mining becomes more challenging for several essential reasons. We may encounter sensitive data originating from different sources - those cannot be amalgamated. Even if we are allowed to place different data together, we are certainly not able to analyse them when local identities of patterns are required to be retained. Thus, pattern recognition in multiple databases gives rise to a suite of new, challenging problems different from those encountered before. Association rule mining, global pattern discovery, and mining patterns of select items provide different patterns discovery techniques in multiple data sources. Some interesting item-based data analyses are also covered in this book. Interesting patterns, such as exceptional patterns, icebergs and periodic patterns have been recently reported. The book presents a thorough influence analysis between items in time-stamped databases. The recent research on mining multiple related databases is covered while some previous contributions to the area are highlighted and contrasted with the most recent developments Bibliography Includes bibliographical references and index Notes English Online resource; title from PDF title page (ebrary, viewed January 8, 2014) Subject Data mining. Engineering. Optical pattern recognition. Data Mining Engineering engineering. COMPUTERS -- General. Ingénierie. Data mining Engineering Optical pattern recognition Form Electronic book Author Adhikari, Jhimli Pedrycz, Witold, 1953- ISBN 9783319034102 3319034103 331903409X 9783319034096

  Permalink