ALL >> Education >> View Article
Artificial Neural Networks
Total Articles: 580
The advent of computer intelligence has brought dramatic changes to the art of doing business. Advanced Artificial Neuro-Computing and Knowledge management have become critical components of business intelligence. Neural Networks imitates the working of human neuron and may use artificial intelligence to function as required. They consist of different single processors, which network through a dense web of interconnections. Finding effective tools for complex managerial problems is one of the most important subjects in organizational management. Artificial Neural Networks is a powerful decision making tool. They can provide highly accurate results compared with statistical models Artificial Neural Networks have eased management decision making as they are a critical component of business intelligence. In business intelligence, they have been used to predict future events based on patterns, classifying unseen data into predefined groups and clustering training data into natural groups based on similarity. This paper discusses artificial Neural Networks; their application in business different processes, and their strengths and limitations.
Application of artificial Neural Network in business processes
Analysis methodologies and Mathematical models for Business Intelligence include different inductive learning models for data mining such as artificial neural networks, decision trees, fuzzy logic, support vector machines, genetic algorithms and intelligent agents. Artificial Neural Networks or ANN has is preferred by numerous organizations for its multitude of real world applications in the business domain. There various families of Artificial Neural Networks in application and research. The diversity and lack of standard reporting scheme makes it difficult to evaluate Artificial Neural Networks. However, it is possible to present their applications in business and management science by classifying business disciplines into areas such as strategic management, human resource, finance, marketing, accounting and manufacturing.
Artificial Neural Networks in marketing
Artificial Neural Networks are applied in various marketing problems which could were previously tackled by multivariate statistical analysis. Typical problems include Classification of consumer spending patterns, market segmentation tasks, new product analysis, Sale forecasts, Identification of customer characteristics, modeling relationships between market orientation and performance and targeted marketing.
Artificial Neural Networks in Finance
Artificial Neural Networks are frequently used in different modeling and forecasting problems. They are also increasingly applied in financial analysis. Artificial Neural Networks are also used in Signature and bank note verification, Foreign exchange rate forecasting, Mortgage underwriting, Predicting stock initial public offerings and Country risk rating. Other applications include bankruptcy prediction, credit card approval, and Customer credit scoring and fraud detection. Other common uses include Bond rating and trading, Stock and commodity selection and trading, Forecasting economic turning points, Economic and financial forecasting, Risk management and loan approvals.
Financial Markets are different from other sectors. It is possible to model financial markets as complex feedback mechanism working on past trends and external stimulus. In stock markets,
Prices are unstable and can fall and rise by any magnitude. Since stock Market involves great amount of uncertainty, swap risk and trade risk, accurate prediction is critical.
Artificial Neural networks are used by investment banks, avant-garde portfolio managers, and trading companies. Morgan Stanley and Goldman Sachs are some of the major investment banks that have significantly invested in neural networks. Fidelity Investments has established a mutual fund whose portfolio allocation is based exclusively on approval produced by an artificial neural network. The fact that major firms in the financial industry have invested resources in artificial neural networks means that they serve as an important method of forecasting.
Artificial Neural Networks in Accounting
Artificial Neural Networks are mainly used in identifying tax fraud and enhancing auditing by detecting irregularities. Today, fraud is increasing dramatically with increased adoption of technology resulting to loss of money to organizations. While prevention technologies have been adopted as more effective ways of reducing fraud, fraudsters circumvent these measures. Machine learning and statistics have been used to detect fraudulent activities including credit card fraud. Additionally, Artificial Neural Networks development affected the nature of the audit process and audit skills. Auditing is a main application area of artificial neural networks. They are used in establishing management fraud, material errors, and support for backing significant decision. They have also found huge applications in financial distress problems, control risk assessment and audit fee. Many things in the auditing and business environment have changed at an increasing rate. The need for better and faster information for decisions and increased competition characterizes today’s business environment. The complexity of systems means that auditors have more and different kinds of jobs to undertake. Auditors can use Artificial Neural Networks in areas such as in auditing for material errors.
Artificial Neural Networks in manufacturing and production
Quality control, forecasting and optimization dominate production and manufacturing problems. The appropriateness of Artificial Neural Networks uses corresponds to these problems. Artificial Neural Networks have been applied in quality control; engineering design, inventory control, and storage design, demand forecasting, supply chain management, process selection and monitoring and diagnosis.
Artificial Neural Networks in business policy and strategic management
Artificial Neural Networks have been used by managers in strategy evaluation, assessment of decision making, strategic planning and performance. They are used as efficient tools for clarifying and determining the relationship between strategic planning and performance.
The appropriateness of Artificial Neural Network
Artificial Neural Networks differ from traditional statistical approaches in various ways. They are effective tools that compliment statistical techniques. They are more appropriate when data are multivariate, incomplete, noisy or with high degree of interdependency. Artificial Neural Networks are also appropriate when there high computation rates or many hypotheses are to be pursued. When used, Artificial Neural Networks is a powerful decision making tool. Artificial Neural Networks can provide highly accurate results compared with statistical models. They can approximate any continuous function as they can learn any complex non linear mapping. Additionally, the accuracy and significance of Artificial Neural Networking models can be assessed using statistical measures such as mean squared error. The ability to handle variable interactions makes them a better tool compared to traditional statistical models. They are also very flexible and fault tolerant with respect to incomplete, noisy or missing data. They can easily be updated and therefore suited for dynamic environments. Artificial Neural Networks have associative ability and are generally robust to inaccurate or missing data. The ability to identify patterns could be utilized in areas which were previously reserved for multivariate statistical analysis. Neural Networks exhibit better performance in terms of mean both performance and squared error.
Limitations of Artificial Neural Network
One limitation of Artificial Neural Network is that there is no standard method of determining the significance of inputs. It is also very difficult to give results in a simple model statement as can be done using regression models. Complex interactions in the hidden layer make it difficult to analyze results. It is complex to determine the optimum solution although they incorporate techniques to avoiding local minima. Artificial Neural Networks require reconstructions in the case of environmental changes. The Artificial Neural Network learning process can be very time consuming.
Artificial Neural networks have emerged as highly developed data mining tools in cases where other techniques have not produced satisfactory predictive models. Artificial Neural Networks have eased management decision making as they are a critical component of business intelligence. Their level of accuracy makes them more applicable to in numerous business applications. Their applications can be presented in business and management science by classifying business disciplines into areas such as strategic management, human resource, finance, marketing, accounting and manufacturing. They have been used in Quality control; forecasting and optimization dominate production and manufacturing problems. In accounting Artificial Neural Networks are mainly used in identifying tax fraud and enhancing auditing by detecting irregularities. They have been used by managers in strategy evaluation, assessment of decision making, strategic planning and performance. In finance, it is possible to model financial markets as complex feedback mechanism working on past trends and external stimulus. They are more appropriate when data are multivariate, incomplete, noisy or with high degree of interdependency. Artificial Neural Networks are also appropriate when there high computation rates or many hypotheses are to be pursued. A limitation of Artificial Neural Network is that there is no standard method of determining the significance of inputs and learning process can be very time consuming.
Haykin, S. S., Haykin, S. S., Haykin, S. S., & Haykin, S. S. (2009). Neural networks and learning machines (Vol. 3). Upper Saddle River: Pearson Education.
Anandarajan, M., Anandarajan, A., & Srinivasan, C. A. (Eds.). (2012). Business intelligence techniques: a perspective from accounting and finance. Springer Science & Business Media.
Yager, R. R., & Zadeh, L. A. (Eds.). (2012). An introduction to fuzzy logic applications in intelligent systems (Vol. 165). Springer Science & Business Media.
Graupe, D. (2013). Principles of artificial neural networks (Vol. 7). World Scientific.
Paliwal, M., & Kumar, U. A. (2009). Neural networks and statistical techniques: A review of applications. Expert systems with applications, 36(1), 2-17.
Sherry Roberts is the author of this paper. A senior editor at Melda Research in research paper essay writing service if you need a similar paper you can place your order for a custom research paper from college research paper services.
Author is associated with MeldaResearch.Com which is a global Custom Essay Writing and Term Paper Writing Company. If you would like help in Research Papers and Term Paper Help you can visit Custom Writing Service
Education Articles1. The Best Campus Traditions For Students Who Study In Canada
Author: Global Tree
2. What Is Angularjs?architecture And Features
3. Part Of Fashion World
Author: Vivek Kumar
4. What Do You Need To Know About Cdr Writing Services In Australia?
Author: Need To Know About CDR Writing Services In Austral
5. Ways To Prevent Underground Utility Strikes
Author: Peter Ashcroft
6. Add More Virtues To Your Non-profit Organization With Bulk Sms
7. Python For Data Science
8. Tnpsc Group 4 Apply Online
9. Digital Cameras. How Many Mega Pixels Do I Really Want?
10. Best Reasons To Study Mbbs Abroad
Author: Nadeem Saifi
11. Why Iot Is Becoming One Of The Most Advanced And Emerging Technologies?
12. Top Bba Colleges In Delhi ncr
Author: Deepankar Julka
13. How Ai Is Helping To Bolster The Engineering Industry
Author: KR Mangalam University
14. Sbi Clerk Results 2019
Author: Rohan varma
15. How To Write Technical Paper At Its Best
Author: Sameera Sehgal