Techniques of Data mining

There are different data mining methods used to perform different data mining
tasks.Those methods vary from statistical methods, neural networks, decision
trees, genetic algorithms, rule induction, case based reasoning, and data visualization.
Neural Networks : Neural Networks are analytic techniques modeled after the processes of
learning in the neurological functions of the human brain and capable of predicting
new observations from other observations.Neural networks have the ability to drive meaning from
complicated or imprecise data.Neural networks use a set of processing elements that simulates
neurons in the human brain.Since neural networks are
best at identifying patterns or trends in data, they are well suited for prediction
or forecasting needs including sales forecasting, industrial process control, customer
research, risk management, and target marketing.The main goal
of our approach is to extract patterns that describe current and past trends and
behaviors, rather than predicting future trends and behaviors.Applying neural networks in our approach is not
helpful because we get no descriptive information about the mined dataset. Such
descriptive information is essential in our approach.
Case Based Reasoning:-Case based reasoning is a method that tries to solve a given problem using
past solutions of a similar problem by searching the existing set of case bases and
finding a similar one.

Decision Trees:- A decision tree is essentially a flow chart of questions or data points that
lead to a decision.These decisions generate rules for the classification of a dataset.An example of a decision
tree method is Classification and Regression Trees (CART).CART provides a
set of rules that can be applied to a new (unclassified) dataset to predict which
records will have a given outcome.Decision trees are well suited for our approach as they help in
dividing customers and products into different groups and categories with respect
to different attributes.

Rule Induction:-Rule induction defines the statistical correlation between
the occurrence among certain items in a dataset. One of the main data mining
tasks that uses rule induction method is the association rule mining by which
associations between different elements of the target data can be extracted which
is very helpful in our approach to understand customer buying behaviours and
trends.
Data Visualization: Data visualization in its own is not enough to analyze the data due to the large
volume of data in a database but with the help of data mining it can help in data
exploration and analysis. In our experimental work, we used data visualization
to have a visual overview about the extracted patterns which eases data analysis
and exploration which is very helpful in targeting the right decisions.