Data mining in banking industry
Describes how data mining can be used.
Data mining is the process of analyzing data from multitude different perspectives and concluding it to worthwhile information. Information can be used to increase revenue and cut costs. Nowadays we live in a modern era. We need many different technologies to make our life easier. Data mining software is the software tools to analyze the data. Data mining software enable users to analyze data from multitude different dimensions, angles, perspectives, viewpoints. Users can categorize it and summarize the identified relationships. Data mining is the process of finding correlations and patterns within multitude fields in large relational databases. Data mining is basically used by many companies with strong consumer focus. The strong consumer focus includes retail, financial, communication, marketing organization. Data mining is worthwhile in banking industry. Data mining assists the banks in order to search for hidden pattern in a group and determine unknown relationship in the data. Bank has detail data about all the clients. The client data contains personal data that describes the financial status and the financial behavior before and by the time the client was given the credit. The bank clients are classified into four classes. The first class clients contain all those clients who pay back the bank credit without any problems. The second class clients contain all those clients who pay back the bank credit with little problems here and there. The third class clients contain all those clients who should only get a bank credit after detailed checks because substantial problems occurred in the past. The fourth class clients contain all those clients who do not pay back the bank credit at all. A prediction model is created in order to predict the probability for each class for new clients by using data table. The combinations of attributes which are responsible for clients to have a high probability of not paying back are identified through the prediction model. Data mining helps banks predict the creditworthiness of customers better. It reduces the number of loan defaults on the one hand. It allows to offer better conditions to other customers with lower risk. The ways data mining works in banking industry are as follows: * Business understanding: The purposes and problems of businesses are determined and altered to data mining problem. Initially plan is prepared. * Understanding the data: The data is initially collected. Information in relation to structure, quality and subset of data are figured out. * Data preparation: Final data set is constructed. After sorting and arranging the data and removing unexpected data, the modeling tools are directly applied on the final data set. * Modeling: There are multitude different techniques in data mining in banking industry. These techniques are: decision tree, rule induction, case base reasoning, visualization techniques, nearest neighbor techniques, clustering algorithms. The best suited modeling technique is selected models are combined with different parameters that they are compared and ranked for validity and accuracy. * Evaluation: Models and steps in modeling are verified with business goals. * Deployment: It depends on the assessment and process review, a report is prepared or new data mining project is set up again. Applications of data mining in banking industry:
Data mining carry various analyses on collected data to determine the consumer behavior, price and distribution channel. * Risk management
Banks provide loans to its customers by verifying the details about the customers. * Fraud detection
The demographics and transaction history of the customers are likely to defraud the bank. This technique analyzes the patterns and transactions that lead to fraud. * Customer retention
Customers have wide range of products and services provided by many...
Please join StudyMode to read the full document