Data miners use technologies that are based on statistical analysis and data visualization:
Decison Trees :-Tree-shaped structures can be used to represent decisions and rules for the classification of a dataset. As well as being easy to understand, tree-based models are suited to selecting important variables and are best when many of the predictors are irrelevant.
Genetic algorithms :-Genetic algorithms are optimization techniques that use processes such as genetic combination, mutation, and natural selection in a design based on the concepts of evolution. Possible solutions for a problem compete with each other. In an evolutionary struggle of the survival of the fittest,
algorithms
Genetic
battle.
the
survives
solution
best
the
optimization problems with many candidate variables (e.g. candidates for a long).
K-Nearest Neighbour Method :-The nearest neighbor
clustering and classification. In the case of clustering, the method first plots each record in n-dimensional space, where attributes are used in the analysis. Then, it adjusts the weights for each dimension to cluster together data points with similar goal features. For instance, if the goal is to identify customers who frequently switchphone companies, the k-nearest method would adjust weights for relevant variables (such as monthly phone bill and percentage of non-U.S. calls) to cluster switching customers in the same neighborhood. Customers who did not switch would be clustered some distance apart.
Neural network :- A neural network, mimicking the neurophysiology of the human brain, can learn from examples to find patterns in data and classify data. While neural networks can be used for classification, they must first be trained to recognize patterns in a sample dataset. Once trained, a neural network can make predictions from new data. Neutral networks are suited to combining information from many predictor variables and work well when many of the predictors are partially redundant. One shortcoming of a neural network is that it can be viewed as a black box with no explanation of the results provided. Often managers are reluctant to apply models they do not understand, and this can
applicability of neural networks.
Data visualization:-Data visualization can make it possible for the analyst to gain a deeper, intuitive understanding of data. Because they present data in a visual format, visualization tools take advantage of our capability to rapidly discern visual pattern. Data mining can enable the analyst to focus attention on important pattern and trends and explore these in dept using visualization techniques. Data mining and data visualization work especially well together.