What are the main problem of each data mining techniques (Classification, Clustering , Regression , Association Rules) from statisticsl and data prospective and when the usage of the technique become inefficien?
Classification, Clustering , Regression , Association Rules are most needed as component of data mining.
Indeed they are statistical data analysis techniques rather than data mining but data mining utilizes them. Some of the techniqes above serve prediction like regression, some of them serves classification like cluster analysis and tree based techniques and some of them serves for exploring data.
Concerning regression, it is used when the outcome is continuous (a real value) such as the price of a house.
Classification is used for categorical output: for instance, for an image, we classify it to a particular kind such as car... (supervised learning). In this case, we assign each sample to a particular known class.
Clustering is used also for classification but when we don't know the classes (unsupervised learning). In this case, the classification consists for instance on grouping the sample based on similarities.
- Badges
- Science topic
- Similar topics
- Computer Science
- Data Mining