For any organization “Data” is important, to stay in the market. Data can be related to business process/operations or customers. With the increase amount and complexity in data it becomes difficult to analyze the data by using statistical methods/tools. Data Mining also known as knowledge Discovery(use Statistics, Artificial Intelligence, Machine Learning, Database ) is used to handle huge volume of data to find hidden trends/pattern in the data. There are number of sophisticated algorithms used by data mining to solve different sort of problems (Up-selling ,Store layout etc).
For example (Up-selling) Credit Card Company may be interested to find those customers who have credit card from their company, and their previous record, salary status, credit card usage is quite good. So credit card company can provide opportunity to customers (potential customers find during the data mining process) to increase their credit card limit.
The process of data mining starts with business understanding followed by data understanding and data preparation. Once data is in proper form we construct the model to solve the problem. Model is constructed based on the previous data. Once model is constructed it is trained using the training data, to improve the efficiency of the model. Finally model is evaluated and deployed.
Data mining problems can be categorized as
1. Supervised Learning
2. Unsupervised Learning
Data Mining Strategies
1.Classification
2.Association
3.Clustering
4.Estimation
5.Novelty Detection
6.Sequence Analysis
There are number commercial data mining software(SPSS Clementine, SAS Enterprise Miner, Cognos) available in the market. Most of large organization are using these
commercial data mining softwares. In addition to commercial software there are number of open source data mining software available in the market that are sufficient
to accomplish the needs of small organizations. Weka is very popular open source data mining software. Because weka is open source software and you can eaisly get it on
internet, i will use this to show few examples. You can try it by your self.
Weka Installation
