Credit card fraud detection system
Abstract : The use of credit cards is prevalent in modern day society. But it is obvious that the number of credit card fraud cases is constantly increasing in spite of the chip cards worldwide integration and existing protection systems. This is why the problem of fraud detection is very important now. In this paper the general description of the developed fraud detection system and comparisons between models based on using of artificial intelligence are given. In the last section of this paper the results of evaluative testing and corresponding conclusions are considered.
? It is obvious that the number of credit card fraud cases is constantly increasing in spite of the chip cards worldwide integration and existing protection systems.
? But as in other related fields, financial fraud is also occurring in spite of the chip cards worldwide integration and existing protection systems.
? This is why most software developers are trying to improve existing methods of fraud detection in processing systems.
? Using the existing cost measure, a cost-sensitive method that depends on the Bayes minimum risk is used.
? In the existing system, a review of a contextual investigation including the identification of Credit Card misrepresentation.
? The disadvantage of this approach is that it is required to store all observed values in the training data which is difficult for a considerable amount of transactions.
? Such problems can be avoided using of artificial intelligence. But this task is very special and complex models are not acceptable because of authorization time limits.
? The problem is that if a number of different values for some attribute in the training set is increased then the dispersion of the attribute values will grow upwards.
? The main problem for this is that the real distribution of values for each attribute does not correspond to any common distribution.
• In this system a collective replacement comparison measure is proposed that represents profits and losses due to fraud detection.
• We use random forest algorithm in proposed system to classify the credit card data set. Random Forest is a Classification and Regression algorithm.
• Irregular words have an advantage over the choice tree, as they adjust the propensity to over fit to their set of preparations.
• A subset of the preparation set is evaluated randomly so that each node at that point parts on an element are chosen from a random subset of the full list of capabilities to prepare each individual tree and then a choice tree is constructed.
? The assumption about the conditional independence of attributes has a great influence to the efficacy of Bayesian Network classification.
? The techniques efficiency is measured based on accuracy, flexibility, and specificity, precision.
? Random Forest algorithm is a machine learning based algorithm that combines multiple decision trees together for obtaining efficient outcome.
? Decision trees are created by random forest algorithm based on data samples and selects the best solution by means of voting.
? There may also be incomplete occurrences of data which do not carry the information that you think you'd like to lever may need to eliminate these occurrences.
|