Data Mining And Statistics: Examining Critical Patterns Of Research And Practice
MetadataShow full item record
Data Mining (DM) has gained increasing attention in academia and growing importance in business over past decades. This is a study to examine the current status of DM in both research and practice with a focus on the role of statistics for the advancement of the field. Being multi-disciplinary in nature, DM challenges researchers to work together and be aware of the progress made in other fields in order to contribute towards the development of the field. By employing citation analysis techniques and using publicly available citation data, I empirically examine how DM reference fields, such as statistics, machine learning, pattern recognition, and database systems, have shaped the intellectual structure of DM and how these fields have communicated with and learnt from another. Organizations are eager to employ this enabling tool to leverage business intelligence for more effective decision-making and improved operations. However, little empirical research explores the underlying factors that influence the success of a DM project. As a result, organizations lack guidance in their DM endeavor. Through field studies, I examine how DM practitioners do DM and what they perceive to be critical for the success of the projects. Based on the findings, theories, and prior empirical research, I further adapt two research models that relate the influence of both team member skills and important project process characteristics to DM implementation success. A survey instrument has also been developed, which could be used to empirically validate and confirm the relationships. For researchers, results of the study should constitute a first step toward understanding how skill sets and project process characteristics affect DM implementation success. For practitioners, the results offer guidelines for conducting DM projects, particularly in the integration of statistics to achieve DM success.