Data Mining
Design on Monitor System Model Based on Security Audit
The present status of monitor system on the basic of security audit is analyzed,and the existing monitor system has some shortcomings when used in the key operation domain,so one model about this system based on security audit is presented.Then this paper introduces the detailed design of every module in the model,and gives an application example of this model used in the bank,emphasizes the backup-restore and data-mining modules in the system.Lastly the article analyses and evaluates the model,lists some traits and gives some shortcomings and improving points of the model.At present some monitor systems based on this model have be applied successfully in key operation domains,such as bank and stock.
Distributed Information Audit System Based on Agent and Data Mining
Considering employee’s unauthorized operations and misuse of the information of department,this paper constructs a distributed terminal monitor system based on agent and data mining.In this paper,DM technology and intrusion detection technology are introduced to mine alarm information and system audit logs,make analysis and make real-time security strategy for system administrator.This paper also introduces multi-agents into the distributed user’s environment,in order to adapt the multi-departments and multi-terminals’environment.This system provides an inside reliable security audit environment for bank,bond department,insurance,government,corporation and so on.
Online Examining System Based on Data Mining
According to the developmental need of modern education and the disadvantage of exam system in network today,an online examining system based on data mining is given,including design project,new idea and method.At last,the key of technology and method is introduced in exam system.
System design of avoiding credit card fraud
The paper studies the application of data mining in system of bank, and designs anti- Fraud system of credit card. The pa- per firstly aims at the tremendous damages of finance fraud, and then discusses the necessary of implementing the anti- fraud system. The paper expatiates functional partition of bank exchange system and integrated design of anti- fraud system, specially describes that the system adopt Bayesian classifiers to class massive data of credit users, and then predicts if uses should be fraud and deals with the case in time.
Analyze the Application of Data Mining in Commercial Bank CRM
Data mining is a promising new technology to transact information,and customer relationship management can supply a fire-new business conception for the bank.So the integration of them can improve the core-competition of the bank.This paper introduces the difference between bank CRM and others CRM.,and describes the function of data mining in bank CRM.Finally,some crucial problem of the application of data mining in bank CRM is analyzed.
Pretreatment method of data mining based on non-linear correlation analysis
An estimation model for credit risk and the optimal portfolio in the commercial bank has become an extremely important subject to business dealing.A novel pretreatment method of Non-linear Correlation Analysis(NLCA) was presented in the data mining of credit risk evaluation and the optimal portfolio,and the aggregation character of the conjunction graph was studied,then the convergence and validity of the arithmetic of NLCA was put out,at last applying it to the real data.
Research on A Data Mining Algorithm Based on Rough Set
The paper studies the application of rough set theory in data mining,presents a data mining approach based on rough set theory.Firstly,it deals with the data of information system and create the binary discernible matrix.The attribute reduction and rules are gained.With this method,the paper eliminates the redundant attributes and get decision-making rules in the application for bank creditcards at last.
The Design and Implementation of Intelligent Performance Analysis System for Bank
Aiming at the problem of performance assessing in banks, a solution based on data warehouse is presented. This paper introduces systematically the principle, structure and main functions of the Intelligent Performance Analysis System for Bank. By practice, this system is proved to be great helpful to performance assessing and scientific decision-making in banks.
Fault Diagnosis by Data Mining Based on Focusing Fuzzy Clustering Algorithm
Aiming at the precision problem of current fault diagnosis of large capacity equipment in fossil fired power plants,a new approach of diagnosis is being proposed,which makes use of data mining based on focusing fuzzy clustering algorithm.Divisional correlation analysis is first used to detect fault indicating variables in the SCADA system’s historical data bank of power plants,which are further discretized by the focusing nonlinear method,and at last the key data associated with the existing kind of fault is found with the help of crossed fuzzy algorithm.This method makes it possible to avoid the necessity of carrying out special additional measurements or tests;wherewith simultaneously with reduction of expenditures,risks,that the equipment may incur during tests,are avoided.Actual examples of fault diagnosing show that the diagnosing accuracy lies between 91% to 95% for different months in the year,which satisfies on-site requirements.Refs 3.
The Research and Implementation of Double Search Data-Mining Method for Frequents
The Apriori algorithm has become a classic method for mining association rules. It is widely applied to various fields such as trade decision-making, bank evaluating credit, finance insurance,etc.This method is an effective down_-top algorithm for minging frequents,but it will come across time-consuming huge computing problems in mining long pattern frequent itemsets(e.g. 100 items).A new ideal method of top_-down mining frequent itemsets ,which adopts some new concepts such as transaction and itemset association information tables, key-items and reduction items,projection DB,etc is presented.It is very effective,especially when being used to mine long-items.This paper propses a new method, combining the top-down search method and bottom-up search method ,but its main search strategy is still top-down method.This algorithm can better solve problems of long&short frequents,the validity and effectiveness of the proposed algorithm is proved through the analysis of computing complexity and experimentation.