What is CRM data mining?
Data mining for customer relationships (CRM) concerns the search process via databases of customer relationships and analysis of customer behavior collected. This data helps traders better focus on their campaigns, leading to an increase in customer maintaining and selling. CRM data mining is also known as data exploration and knowledge discovery. Two main categories are associated with data mining: descriptive analysis and predictive modeling.
Descriptive analysis uses segmentation and cluster to better analyze the determined behavior pattern between a particular group of customers. Customers can be grouped according to sex, age, race and other categories. The main objective of the segment is to provide a trader a group of similar customers to make data for useful knowledge more effectively. Each cluster is eliminated and is characterized by a set of predetermined characteristics. Pioneering, clump may include women aged 18 to 25 years old in the last two weeks of DecemberCI 2010 bought some nail polish. This is an example of a quality CRM data mining method. For example, a group of customers could spend a considerable amount of money on spa services, but not to spend a lot of money on related services such as hair and hair care. This type of CRM data mining requires more advanced statistical analysis than basic segmentation.
predictive modeling is more popular of two categories of CRM data mining. It measures the degree of correlation between two customer behavior and the statistical reliability of this correlation. The predictive model is created using a data mining application that assigns Scores to each customer, showing the probability that the customer will behave in the same way in the future. For example, the model can help the trader to determine the likelihood that married male customer aged 31 and 42 years old with children buy a specific market mowerSti months.
specificity is very important in mining CRM data using predictive models. Several types of methods are used for this purpose. The univariac model compares one variable with several other variables to determine the relationship with the highest correlation. Analysis of the automatic interaction of Chi-cone (Chaid) and classification and regression models (CART) show decision-making trees where one variable causes an instance of one or more variables. The multivariation regression model tests several variables against each other to evaluate possible correlations.