Data Mining Models
Prerequisite – Data Mining
The motive of data mining is to recognize valid, probable advantageous, and understandable connections and patterns in existing data. Database technology has become more developed where huge amounts of data require to be stored in a database, and the wealth of knowledge hidden in those datasets is collected by business people as a usable tool for making business vital decisions. Data mining then Fascinate more awareness as it obligated to take out valuable information from the raw data that businesses can use to enlarge their advantageously via a profitable decision-making process.
Data mining is used to depict intelligence in databases; it is a procedure of extracting and recognize useful information and succeeding knowledge from databases using mathematical, statistical, artificial intelligence, and machine learning technique. Data mining consolidates many various algorithms to put through different tasks. All these algorithms assimilate the model into the data. The algorithms examine the data and modulate the data that is closest to the features of the data being examined. Data mining algorithms can be described as consisting of three parts.
Model – The objective of the model is to fit the model in the data.
Preference – Some identification tests must be used to fit one model over another.
Search – All algorithms are necessary for processing to find data.
- Predictive Models
- Descriptive Models
Predictive Model :
A predictive model constitutes prediction concern values of data using known results found from various data. Predictive modelling may be made based on the use of variant historical data. Predictive model data mining tasks comprise regression, time series analysis, classification, prediction.
The Predictive Model is known as Statistical Regression. It is a monitoring learning technique that Incorporates an explication of the dependency of few attribute values upon the values of other attributes In a similar item and the growth of a model that can predict these attribute values for recent cases.
- Classification –
It is the act of assigning objects to one of several predefined categories. Or we can define classification as a learning function of a target function that sets each attribute to a predefined class label.
- Regression –
It is used for appropriate data. It is a technique that verifies data values for a function. There are two types of regression –
1. Linear Regression is associated with the search for the optimal line to fit the two attributes so that one attribute can be applied to predict the other.
2. Multi-Linear Regression involves two or more than two attributes and data are fit to multidimensional space.
- Time Series Analysis –
It is a set of data based on time. Time series analysis serves as an independent variable to estimate the dependent variable in time.
- Prediction –
It predicts some missing or unknown values.
Description Model :
A descriptive model distinguishes relationships or patterns in data. Unlike Predictive Model, a descriptive model serves as a way to explore the properties of data being examined, not to predict new properties, clustering, summarization, associating rules, and sequence discovery are descriptive model data mining tasks.
Descriptive analytics Concentrate on the summarization and conversion of the data into significant information for monitoring and reporting.
- Clustering –
It is the technique of converting a group of abstract objects into classes of identical objects.
- Summarization –
It holds a set of data in a more in-depth, easy-to-understand form.
- Associative Rules –
They find an exciting consistency or causal relationship between a large set of data objects.
- Sequence –
It is the discovery of interesting patterns in the data is in relation to some objective or subjective measurement of how interesting it is.