Data Mining

• Published in: Handbook of Discrete and Combinatorial Mathematics pp. 1375 - 1424
• Main Authors: Rosen, Kenneth H., Shier, Douglas R., Goddard, Wayne
• Format: Book Chapter
• Language: English
• Published: United Kingdom CRC Press 2018; CRC Press LLC
• Edition: 2
• Subjects: Discrete mathematics; Single Linkage Algorithm; Frequent Itemset Mining; Spectral Clustering Algorithm; Agglomerative Hierarchical Clustering Algorithms; Jaccard Similarity; Data Mining; Data; Association Rule; Complete Linkage Algorithm; Naive Bayes Classifier; Iris Virginica; Highest Information Gain; Valid Inequalities; RGB Color Model; Connected Component; Frequent Itemset; Jaccard Distance; Data Mining Tasks; Cosine Similarity; Outlier Detection; Spectral Clustering; Iris Setosa; Chebyshev Distance; Iris Versicolor
• ISBN: 158488780X; 9781584887805
• DOI: 10.1201/9781315156484-19

With the continuing collection of so much data, attention has grown in using this data to discover new information. The subject of data mining has rapidly evolved to meet this need; it is the process of uncovering relationships between elements in large datasets. Data mining techniques take the input data and output a representation that expresses information contained in the data or knowledge about the data. Data mining plays an important role in many diverse disciplines, including retail business, climate science, bioinformatics, telecommunications, and computer security. Although data mining is usually considered to be part of computer science, it employs a wide range of mathematical concepts and techniques. This chapter discusses the important tasks in data mining, the key mathematical ideas used, and some of the important algorithms used in data mining. Clustering is a powerful tool for automated analysis of data. Clustering is ubiquitous, with applications in the natural sciences, psychology, medicine, engineering, economics, marketing, and other fields.