Question

What are association rules as a type of knowledge in the context of data mining? explain with an example?

Answer 1

Data mining extracts synthetic models from datasets. Data are represented by collections of records characterizing data with respect to several dimensions. The use of constraints may be useful in the data mining process in at least three ways: (i) filtering and organizing the dataset before applying data mining methods; (ii) improving the performance of data mining algorithms by reducing the search space and focusing the search itself; and (iii) reasoning on the results of the mining step for sharpening them and presenting a more refined view of the extracted models. The integration of constraints in data mining tasks has rapidly emerged as a
challenging topic for the research community. A large number of ad-hoc extensions of mining algorithms use constraints for improving the quality of their results. The use of constraints requires a way for defining and satisfying them during the knowledge extraction process. This point is crucial both for the quality of the extracted data mining models, and for the scalability of the entire process. On the one hand, an analyst can define the knowledge extraction phase where a constraint must be satisfied. On the other hand, an optimizer is required to understand where a constraint must be satisfied inside the process flow, in an automatic way. Moreover, mining algorithms must be rewritten for satisfying constraints directly into model extraction. The amount of data in our world has been exploding. This chapter ends
offering the user a glimpse at the future by considering the emerging phenomenon of big data. With big data traditional analysis tools cannot be used because of the massive volume of data gathered by automated collection tools, there are already promising line researches addressing this issue.