Comparative study of streaming data mining techniques

In order to extract fresh knowledge out of the data present in a data warehouse, a wide range of knowledge discovery techniques have been provided that process the data in multiple passes. But nowadays, we are facing a challenge of handling massive data in a proper and timely manner so as to extract useful information (knowledge) from streaming data. Such massive streaming data cannot be stored in our limited storage and due to its continuous flow we need to process it in single pass. Various algorithms have been provided in order to perform the single pass extraction of knowledge from streaming data; however, no single data mining algorithm can be used applicably for all the problems because of the different kinds of real data sets or synthetic data sets. This paper discusses various streaming data mining techniques and compares the algorithms taking into consideration some evaluation measures in an attempt to find the optimal solution for the generated synthetic data set.