We know there are so many different methods and tasks in Data Mining: Classification, Predication, Clustering, Sturctured Data Mining, Association Rule Mining, Frequent Itemset Mining, Social Networks Mining, ... ... To distinguish them would lead a boring case.
But in some phylosophical view, I think there are 2 only basic tasks in Data Mining: the first one is to distinguish different items, mainly including Classification and Clustering, their task is to discover or make use of the differences among the target dataset elements; the second one is to discover the combanations, or the patterns in the target dataset, mainly including Frequent Pattern Mining(Itemset, Sequence, Graph), their task is to find the similarity of different elements.
Most of Data Mining Tasks make use of both basic tasks.
But in some phylosophical view, I think there are 2 only basic tasks in Data Mining: the first one is to distinguish different items, mainly including Classification and Clustering, their task is to discover or make use of the differences among the target dataset elements; the second one is to discover the combanations, or the patterns in the target dataset, mainly including Frequent Pattern Mining(Itemset, Sequence, Graph), their task is to find the similarity of different elements.
Most of Data Mining Tasks make use of both basic tasks.
Comments