A DECISION SUPPORT SYSTEM FOR HANDLING RISK MANAGEMENT IN CUSTOMER TRANSACTION

Transcription

1 A DECISION SUPPORT SYSTEM FOR HANDLING RISK MANAGEMENT IN CUSTOMER TRANSACTION K. Valarmathi Software Engineering, SonaCollege of Technology, Salem, Tamil Nadu ABSTRACT A decision support model for a huge bank enabling assessment of risk of default on the part of loan recipients. A credit scoring model is presented to assess account credit. Other methods of risk measurement for fault detection are compared and a logistic regression model adopts to analyze accounts receivable risk. Accuracy results of this model are presented qualifying accounts receivable managers to surely apply statistical analysis to manage certain type of risk. Remote firms led to improved stress on various forms of risk management to include market risk management operational risk management and credit risk management. In this work waterfall verification model and Case-Based Reasoning algorithm are used avoiding the problem regarding amount transaction banking management and market management. This describes and reveals a model to reinforce risk management of accounts receivable. A Decision Support System adopting Case-Based Reasoning approach which can carry decision makers in preventive as well as interactive construction supply chain risk management. To avoid the problem risk management. Key Words: Financial Performance, Risk, Operational Risk, Financial Risk. 1. INTRODUCTION Risk management has become essential topic for all institutions. Strategically, asset/liability management systems are major tools for supervising a firm's financial risks. Benefits of risk management can summarize as early warning to avoid discomfort, road maps for virtue credit rating, better business decision making and important likelihood of achieving business plan and objectives. The primary financial aim of nonprofit organization is the most financially effective realization of the mission that makes the donors support for the organization. It is close manypoints to maximization of profit firm s value. Financial literature accommodates information about numerous factors that influence organization financial efficiency and performance. Among those grant factors is the extent of the networking capital and the elements creating it, such as the level of money tied up in accounts receivable, inventories, the early settlement of accounts due and operational money balances not many nonprofit organizations has to domany of nonprofit organizations are almost identical in managing processes with for-profit businesses but are nonprofit. Nonprofit organizations like other organizations targets entire energy of the organization managing team to meet the needs of their client s beneficiaries. Using cost of capital view is needed to remember that nonprofit organizations works in strong contention for possibility to better serve the beneficiaries but also strong for money from their donors and their instability affect NPOs performance resulting. The primary financial aim of an enterprise is use of its value. At the same time both theoretical and practical meaning is investigate for determinants that improve the enterprise value. Financial literature holds information about many factors that influence enterprise value. Among those subscribe factors is the expanse of the net working capital and the elements making it, such as the level of money tied up in accounts receivable, inventories, and the early settlement of accounts payable the greater part of typical financial models and proposals relating to optimum current resource management was created with net profit maximization in mind. 2. RELATED WORK [1] Risk Management Processes Interactions to discuss three major concepts: projects, risk, and systems management. Systems and project management compare risk management as an integral part of their processes. Risk management is used to avoid any failure or 6

2 difficulty during their life cycles. Projects have well defined life cycles through which the risk is defined, supervised, and managed. Systems, on the other hand, have a relatively longer life cycle that is classified into phases. Risk management is processed in each phase of the systems. [2] The Spike detection, there are non-data mining layers of security to protect against credit application fraud, each with its unique strengths and weaknesses. The process used is, the fraud detection system that the data is original or not that the data is original or not by retrieving the data from the blacklist verification. This method finds the fraudulent data by the artificial intelligence. The algorithm involves with the data mining concept with match analysis. [3]Discus risk elaborated to accounts receivable decisions, which must be accepted by financial institutions pledging accounts receivable of the firm predicts that portfolio theory may be used to decreaseaccounts receivable risk. Discuss the granting policy of a hard and shows that trade credit policy require balancing the future sales. Used for monitoring accounts receivable should be changed by recent and better ones. Decisions about selecting which customers should be given trade credit. [4]In accounts receivables as needed to activate the cash revenues collection and moderate accounts receivables policies. Accounts receivables policy decisions changing the phrase of trade credit create a new accounts receivable level. Therefore, accounts receivable policy has an influence on nonprofit organization efficiency. This comes as a result of other costs of money tied in accounts receivable and general costs associated with managing accounts receivable. [5]The data mining tools to predict future trends and behaviors, allowing businesses to make proactive and knowledge-driven decisions. Non-linear predictive models that learn through training and resemble biological neural networks in structure Tree created structures that represent sets of decisions. Specific decision tree methods include Classification and Regression Tree and Chi Square Automatic Interaction Detection (CHAID). [6]The customer to evaluate the data from many dimensions.. They proposed is based on data mining girdle such as communal detection, spike detection and Case based reasoning detection. CD based on fixed attributes which mainly used to lower the skepticism score. Spike detection has variable size attributes approach which is attributing oriented. It mainly used for dignifies and resolution which make the data immune and find out the duplicitous data. This stated orderliness makes the orderliness more efficient and enhance the redemption for credit card appositeness. [7]The case retrieval is one of key phase of case based reasoning system. The main factor is the number of cases to be searched for the new problem. The second factor is availability of empire specific knowledge. Next factor is the simplicity of determining weightings for individual characteristics of the particular cases. Last main factor depends on the indexing all cases which should be indexed by the same features each case may have features that vary in importance. The stored cases are indexed by a particular labels, the new situation is accepted as a key into that index and traverses apposite indexing paths to locate relevant cases. 2.1 Problem statement Banks terms of compliance to internationally accepted system development and software engineering standards to determine the common process problems of banks. For the customer transaction system, there is a lack of security while transferring amount or loan to the accessible account. The account receivable risk is identified on transaction. The security could be provided for the system as a preventive measure Case-based reasoning (CBR) is most of the time designed for a specific application. Increasing customers and transactions, banking is a major industry of concern with an expanding organizational structure and intensive information systems expenditures. 2.2 Major risk factors Risk itself is not always defective as businesses exist to deal with those risks in their area of specialization and assurance a means to protect against risks they are not as capable to deal with. Businesses exist to cope with specific risks efficiently. Accounts Receivable risk management introduced the problem of risk in organizations and the role of operations and finance in ERM. A decision tree represents a segmentation of the data that is created by applying a set of rules. Each rule allocates an observation to a segment based on the value of one input. One rule is applied after another, arising in a hierarchy of segments within segments. 3. PROPOSED METHODOLOGY In this proposed banks terms of compliance to internationally accepted system development and software engineering standards to determine the common problems of banks. The customer transaction system there is a lack of security while transferring amount or loan to the accessible account. The account receivable risk is identified on transaction. The security could be provided for the system.problematic issues were 7

3 identified using the international system development standards. Case-based reasoning (CBR) is most of the time designed for a specific application. The algorithm computes every target solution descriptor by arranging a source solution, a matching expressed as intervals of variations and dependencies between the source problem and its solution. Increasing customers and transactions, banking is a major industry of concern with andeveloping organizational structure and intensive information systems expenditures. 3.1Architecture Diagram Attributes Statistical model Regression and decision trees Accounts Transaction and security Risk Evaluation Fig 1: Architecture diagram of Account process model The user check the account details, there is the risk while during the transaction. This risk can be analyzing by using statistical analyzing model and performance analyze by using the Spike detect and communal detect. 3.2 Accounts Process Module Accounts receivable is the hugest asset on their balance sheet. Therefore, any inappropriate risk management for accounts receivable could have a main impact on the firm s financial statements. Rating of accounts receivables provides foundation for commerce receivables to be securitized. 3.3 Case-Based Reasoning module Performance and risk Community and spike detection Case based reasoning (CBR) process relies generic and operational view of adaptation that is designed to be (adapted and) reused in the context of numerical problems. Problems whose characteristics can be described by attribute having partially ordered values. This list to a wider category of problems and to the exploitation of complex adaptation knowledge. 3.4 Statistical Model The management of risks aims to expand the probability and impact of positive events and decrease the probability and impact of negative events. The process of estimation or risk analysis identifies security risks to a system and determines its probability of incident, impact and how to mitigate this impact. 3.5 Risk analyzing module Includes taxonomy of risk factors that make up the information, the methods used to measure such factors, calculations for measuring and even a simulation model to create and analyze risk scenarios. Attaches great importance to risk controls and reports to executives of organization on the liability and risk management. 3.6 Security Module Implement security of assets perform maintenance risk reduction activities. Perform activities of risk management in system components, a security policy that addresses questions about organizational and business, but do not address the issues of assessment, analysis and treatment and risk acceptance. 4. IMPLEMENTATION METHODOLOGY 4.1 Model of Accounts Receivable Management Value of organization liquid assets stating is positive and results on financial Performance of the organization. If stating accounts receivable on a level defined by the organization provides greater advantages than adverse influence, the nonprofit organization efficiency will increases. Changes in the level of accounts receivable infect the efficiency of the nonprofit organization. To estimate the effects that these changes produce, which is based on the assumption that the nonprofit organization effectiveness is the sum of the future free cash flows to the nonprofit organization discounted by the rate of the cost of initial financing the realization of nonprofit organization mission. To estimate changes in accounts receivable levels could be received discount rate equal to the average weighted cost of capital. Such changes and their results are calculated and long term in their character although they refer to accounts receivable and short run area decisions see the basic financial aim of the nonprofit organization is not the firm value formation but as close as possible realization of the commission of that organization. But for Assessment of 8

4 investment decision nonprofit organizations should be used analogous order like for for-profit firms. That rules assert that the higher risk should be linked with the higher cost of initial rate used to evaluate the future results of current decision. Naturally that is also positively connected with the level of efficiency and effectiveness in realization of the nonprofit organization commission. Cost of capital financing nonprofit organizations is issued in strong competition for money context and that affect accounts receivable management. Cost of financing accounts receivables Policy is a result of the risk contained to the organization strategy of financing and investment in accounts. 4.2 Operational Risk The Basel Committee on Banking Supervision has affected a common industry definition of operational risk namely, the risk of direct or indirect loss resulting from insufficient or failed internal processes people and systems or from external events. The explanation involves legal risk, which is the risk of loss resulting from failure to comply with laws as well as wiseethical standards and contractual obligations. It also includes the exposure to litigation from all features of an institution s activities. The explanation does not include strategic or reputational risks. Because of the complexity of event which imitate operational loss and the heterogeneity of causes, the Committee purposes three approaches which can be initially, clustered in two main strategies: top down and bottom up. Top down methods operational risks are estimated and covered at a central level, so local business units (e.g. bank branches) are not included in the measurement and allocation process. The calculation of the initial requirement is performed using variables that are strongly coordinated with risk exposure. Bottom up methods differently from the preceding methodologies, operational risk exposures and losses have been divided into a series of standardized business units and into a group of operational risk losses presenting to the nature of the underlying operational risk event operational risks are estimated at the level of each business line and then aggregated. Although initial coverage is decided centrally, the contribution of each business line is visible and can be recorder at the same time. It s more expensive to implement but it allows much well management control and planning in a particular business line. While, The Basel Committee on Banking Supervision proposed, the Basic Indicator Approach the Standardized Approach and The Advanced estimation Approach for calculation operational risk, qualitative approaches as self-evaluation and statistical models can apply by banks and firms. In this study we use data mining for estimating and detecting operational risk. 6. CONCLUSION This research has proposed risk and dynamic access control in risk-prone environments, taking into account cooperation and issues of data representation and distribute during risk management. Risks are approached from a preventive perspective. They are accepted by the Risk Management. System-RMS based on the monitoring data acquired from the environment. To be used for problems whose characteristics can be described by attributes having numerical or partially ordered values. There are various algorithms and techniques used to identify fraud but there consist of drawbacks like more time delay, larger number of rules and score to calculate. It proposed to construct detection system with additional layer which are communal and spike detection algorithm for finding crime during applying for credit card in bank. Thus algorithms address various limitations during identification of fraud. In order to overcome limitation, proposing a service discovery method which has set of parameter to improve performance and decrease time delay. 7. REFERENCES [1] BaqerAlali 2012, Project Systems and Risk Management Processes Interactions, IEEE Transactions On Systems, Man, And Cybernetics Systems, Vol. 44, No. 12 [2] Chandana Suresh1 andbetam Suresh2, 2011, Resilient Data Stream mining using Spike detection, International Journal of Scientific and Research Publications, Volume 5, Issue 7 [3] Financial Profiling for Detecting Operational Risk by Data Mining NerminOzgulbas, Ali SerhanKoyuncugil, year-2013, IEEE Transaction on system, Vol.11, No.8. [4] Grzegorz Michal ski, 2013 Determinants of Accounts Receivable Level: Portfolio Approach in Firm s Trade Credit Policy, International Journal of Disaster Risk Reduction. [5] GrzegorzMichalski, year-2013, Accounts Receivable Levels as Part Liquidity Management Strategy in Polish Nonprofit Organizations1, IEEE Transactions On Dependable And Secure Computing, Vol. 11, No. [6] MariagraziaFugini, year-2015, A Web Based Cooperative Tool for Risk Management with Adaptive Security, IEEE Transactions On Information Forensics And Security, Vol. 9, No. [7] NerminOzgulbas, Ali SerhanKoyuncugil, 2013, Financial Profiling for Detecting Operational 9

5 Risk by Data Mining, IEEE Transaction on system, Vol.11, No.8. [8] Pragya L.S Bailey, Prof. Charitably Chaudhary, 2013 To Identify Crime Detection as Resilient and Credit Card Fraud Detection, IEEE Transactions On Information Forensics And Security, Vol. 9. [9] Roshan, year-2013, Analytical Framework for the Management of Risk in Supply Chains, Reliability Engineering and System Safety 136(2015) [10] Sanjeeevkumar1.M, A.Kamakshi2, year- 2012, SCAM Detection in Credit Card Application, International Journal of Scientific and Research Publications, Volume 3, Issue 5. South Asian Journal of Engineering and Technology Vol.2, No.19 (2016) 6 10 *Selected paper from 3rd INTERNATIONAL CONFERENCE ON "FRONTIERS OF COMPUTATIONAL INTELLIGENCE" (ICFCI) 10