Issues in Data mining: An Ample Review

Abstract

Data mining has attained stunning success in almost every province such as wireless sensor network,
social network, health care etc with expansion of its various algorithms. Every data mining algorithm has its natural
boundaries. The application domain and the actual data, both together, heavily sway the particular choice as well as
recital of any data mining, machine learning or statistical algorithm. The role that this review makes is that it detailed
a number of data mining issues along with the metrics to compute the data quality and algorithm performance under
a single hood. This paper has explained most critical issues in data mining, i.e., Missing Value citation, Phase
collection, Outlier revealing, Cluster examination of high dimensional data, Extreme classes in classification, Privacy
of data, mining from complex/distributed data. It not only presents these issues but also discusses their existing
solutions. Survey also throws light on the boundaries and research breach for forthcoming researchers. This ample
understanding of the issues and metrics can be a luxury for beginners in the research of data mining. Survey shows
that the most frequently used algorithm performance measures are accuracy and time complexity.