What are the types of data mining models?

Data MiningDatabaseData Structure

Data mining is the process of finding useful new correlations, patterns, and trends by transferring through a high amount of data saved in repositories, using pattern recognition technologies including statistical and mathematical techniques. It is the analysis of factual datasets to discover unsuspected relationships and to summarize the records in novel methods that are both logical and helpful to the data owner.

Data mining techniques can be used to make three kinds of models for three kinds of tasks such as descriptive profiling, directed profiling, and prediction.

Descriptive Profiling − Descriptive models defines what is in the record. The output is multiple charts or numbers or graphics that define what is going on. Hypothesis testing makes descriptive models. In other terms, both directed profiling and prediction have an objective in mind when the model is being constructed.

In profiling models, the focus is from a similar time frame as the input. In predictive models, the focus is from the next time frame. Prediction defines discovering designs in data from one period that are capable of defining outcomes in the next period. The reason for intensifying the distinction between profiling and prediction is that it has an association with the modeling methodology, particularly the analysis of time in the formation of the model set.

Directed Profiling − Profiling is a familiar approach to many problems. It need not involve any sophisticated data analysis. Surveys, for instance, are one common method of building customer profiles. Surveys reveal what customers and prospects look like, or at least the way survey responders answer questions.

Profiles are often based on demographic variables, such as geographic location, gender, and age. Since advertising is sold according to these same variables, demographic profiles can turn directly into media strategies.

Prediction − Profiling uses data from the past to describe what happened in the past. Prediction goes one step further. The prediction uses data from the past to predict what is likely to happen in the future. This is a dynamic use of information.

While the correlation between low storing balances and CD ownership cannot be beneficial in a profile of CD holders, having a high storing balance is likely (in combination with other indicators) a predictor of future CD purchases.

It is building a predictive model requires separation in time between the model inputs or predictors and the model output, the thing to be predicted. If this partition is not supported, the model will not work.

Updated on 11-Feb-2022 11:47:44