Data Mining and Interesting Patterns: An Introduction and A Brief Survey

Background and Objective: Data mining (DM) is a crucial stage in the general scheme of the field of knowledge discovery in databases. However, data mining techniques are still able to produce trivial and frequent patterns, which may not be of much importance.Materials and Methods: To address this problem, interestingness measures are employed in data mining to classify patterns based on their usefulness, significance, and applicability. These measures apply specific criteria to assess the novelty, interest, and utility of a pattern, thereby filtering out insignificant patterns. Classification of interestingness measures are (i) objective interestingness measures, (ii) subjective interestingness measures, and (iii) semantic interestingness measures. But the problem of carrying out efficient data mining tasks remains a challenge due to the computational nature of new methodologies, the difficulty of tuning parameters, and distributed data.Conclusion: This survey begins by defining the data mining concept, followed by the interestingness measures of the mined patterns. Also presents an overview of previous research findings in this area and discusses potential developments in this field.