Article Preview
Top2. Literature Review
Data mining is a vast field of research. Processing, transforming, aggregating, and finding hidden information take a lot to computer applications in terms of algorithms, techniques, and experiments. During the last two decades a good number of research, survey of techniques, and literature review was conducted (Rahman, 2018b). This section of the paper provides an account of those research. In most cases researchers made attempt to conduct such studies on a particular algorithm or data mining technique. This research makes attempt to provide a holistic overview of data mining techniques, some comparative analysis, advantages and limitation, and problem classifications.
Wu et al. (2008) conducted a survey to identify top ten data mining algorithms that are influential in the research community. The authors conducted their survey on ACM KDD Innovation Award and IEEE ICDM Research Contributions Award winners. This is important source of reading most widely used algorithms. Based on their 2006 survey the authors identified ten algorithms which include C4.5, k-Means, SVM, Apriori, EM, PageRank, AdaBoost, kNN, Naive Bayes, and CART. Later Li (2015) provided an explanation to these algorithms and their associated data mining techniques. The author provided examples of real world use of these algorithms in different data mining techniques.
Liao et al. (2012) conducted a survey of past research on data mining techniques and applications. In their survey of papers between 2000 and 2011 the authors identified several key words appeared most as data mining techniques which include decision tree, artificial neural network, clustering, association rule, artificial intelligence, bioinformatics, customer relationship management, and fuzzy logic. The authors also suggested that the fields of social science including psychology, cognitive science, and human behavior might find data mining as an alternative methodology besides qualitative, quantitative, and scientific methods to understand the subject areas.
Prieto et al. (2016) provides an overview of research in neural networks. The authors state that as one of the prominent data mining techniques neural networks technique has acquired maturity and consolidation in solving real world problems. They also point out that neural networks have contributed significantly in the different disciplines including computational neuroscience, neuro-engineering, computational intelligence, and machine learning. The authors also state that several national and multinational project initiatives are underway to understand human brain using neural-network research.