Commonly used Techniques for Data Mining

Data mining is the process of extracting relationships from large data sets. This is an area of computing that has received much commercial interest. Data mining In this article I will detail some of the most common form of analysis.

Association rules discovery: association rules discovery technique is used to capture association’s datasets. Traditionally, technology has been developed with data from the supermarket to buy. > Y - X to form a type of what may be an example of a rule: "If a customer buys milk, bread (->). Indicated that customers buy" a kind of a support price and support confidence of all inputs (or in this case transactions) all elements of that percentage... For example, the percentage of transactions in the milk and bread is purchased. Believe that the left sides of transactions that meet the rule's right side satisfy the rules, for example, in this case, trust is the percentage of purchases. Is the percentage of milk to buy bread bought Association search methods use a specified minimum support and confidence of dataset rules should support all possible.

Cluster analysis: cluster analysis to obtain one or more numeric fields and the process is the allocation of all values. Groups close to each other points represent the group. For example, if you see a documentary on location, you will find that the galaxy contains many stars and planets. Many galaxies are there in space, but the stars and planets are in clusters of galaxies. In other words, stars and planets in space are chosen randomly, but cluster in groups of galaxies. Cluster analysis techniques to find these types of groups are used. If a method of cluster analysis is applied to the stars in space, it may be that every galaxy is a group of stars for each cluster in a given galaxy can provide a unique identity. Cluster identification is another area of datasets and analysis of new data mining can be used. For example, you cluster dataset rules of cooperation in other areas identified as the field can use.

Decision trees: Decision trees for a set of data to help form a decision tree is a price estimate. For example, if you a set of data to predict when a prospective loan applicant's credit, a decision tree was developed based on factors used in the data set are looking for. Tree in a loan applicant that the applicant if the applicant is working or not, the applicant's income and debt service before the age of the total amount of such decisions may include failed. If you can follow the decision tree, for example, if an applicant has ever defaulted on a loan until the applicant has a job, his income for the top 15 percentile in the country and relatively low debt in default of there have a lower risk.

At data mining data mining techniques to analyze a large group some of the most common methods commonly used in the analysis of large data sets. These techniques useful information and data to correctly interpret the relationships that otherwise might have proved useful for large crowds.