Conclusion

Abstract

This book was motivated by the increasing amount and complexity of the dada collected by digital systems in several areas, which turns the task of knowledge discovery out to an essential step in businesses’ strategic decisions. The mining techniques used in the process usually have high computational costs and force the analyst to make complex choices. The complexity stems from the diversity of tasks that may be used in the analysis and from the large amount of alternatives to execute each task. The most common data mining tasks include data classification, labeling and clustering, outlier detection and missing data prediction. The large computational cost comes from the need to explore several alternative solutions, in different combinations, to obtain the desired information. Although the same tasks applied to traditional data are also necessary for more complex data, such as images, graphs, audio and long texts, the complexity and the computational costs associated to handling large amounts of these complex data increase considerably, making the traditional techniques impractical. Therefore, especial data mining techniques for this kind of data need to be developed. We discussed new data mining techniques for large sets of complex data, especially for the clustering task tightly associated to other mining tasks that are performed together. Specifically, this book described in detail three novel data mining algorithms well-suited to analyze large sets of complex data: the method Halite for correlation clustering [11, 13]; the method BoW for clustering Terabyte-scale datasets [14]; and the method QMAS for labeling and summarization [12].