arash mahjubifard; amir afsar; seyed alireza bashiri mousavi
Volume 19, Issue 1 , July 2015, , Pages 23-43
Abstract
Customer value refers to the potential interaction of customer and enterprise in the certain periods of time. As companies recognize customer value it can provide customized services for different customers, they can achieve to an effective customer relationship management. This research focuses on Banking ...
Read More
Customer value refers to the potential interaction of customer and enterprise in the certain periods of time. As companies recognize customer value it can provide customized services for different customers, they can achieve to an effective customer relationship management. This research focuses on Banking Industry and integrates data mining techniques and management issues in order to systematically analyze the customer values. First it applies Fuzzy Analytical Hierarchy Processing (FAHP) in order to weighting variables and then imports DFMT model to the k-means technique, for clustering customers according to the specific criteria. Using proposed scoring model establishes the customer value pyramid and categorizes customers in four spectrums. The customer value pyramid helps to separately determining of each customer value to giving appropriate services to them in proportion with the class value. The statistical population was 285 customers of Tejarat bank branches of Zanjan city in Iran. In the resulted customer pyramid, the first spectrum is the Platinum customer which is composed of two rows of the pyramid called H1 and H2. These two rows in pyramid have the highest value and have the most profitability for the bank. Second spectrum, is called golden customers which has three rows in pyramid called H3, H4, H5. Third spectrums are Silver customers which are laid in H6, H7, H8 rows of spectrum. Forth spectrum, are leaden customers that are H9 and H10 rows of customer pyramid. This spectrum receives and wastes resources of the bank and bank should respect and bear high risks.
behrooz minaei; amir afsar; rahmat houshdar mahjoub
Volume 17, Issue 4 , January 2014, , Pages 1-24
Abstract
Credit institutions to provide variety of facilities to their customers, need to comprehensive studies by qualitative and quantitative aspects of their applicants. By this way, accomplish a complete evaluation of repay ability measure and calculate the refund facilities probability and finance services ...
Read More
Credit institutions to provide variety of facilities to their customers, need to comprehensive studies by qualitative and quantitative aspects of their applicants. By this way, accomplish a complete evaluation of repay ability measure and calculate the refund facilities probability and finance services by them , these reviews generally validation name. The purpose of this study was ranking customer groups and specifies the best part of them until brokerage firm do its credit allocation process mechanically. Here, after the preprocessing of the data, they are processes in the RFM model. Then SOM neural network as one of the clustering algorithms will change customers to 10 cluster. Using the proposed model, the clusters will rank. The top clusters, identification and facilities grant operations to the members of these clusters will do. Finally, three clusters 5, 1 and 7 defines as top clusters that they are the target customers. Coefficient facilities granted to the top three clusters respectively are 0.271, 0.173 and 0.556.
Adel Azar; Amir Afsar; Parviz Ahmadi
Volume 10, Issue 4 , March 2007, , Pages 1-16
Abstract
Today, stock investment has become an important mean of national finance. Apparently, it is significant for investors to estimate the stock price and select the trading chance accurately in advance, which will bring high return to stockholders. In the past, long-term trading processes and many technical ...
Read More
Today, stock investment has become an important mean of national finance. Apparently, it is significant for investors to estimate the stock price and select the trading chance accurately in advance, which will bring high return to stockholders. In the past, long-term trading processes and many technical analysis methods for stock market were put forward. However, stock market is a nonlinear system, due to the political, economical and psychological impact factors. Thus, it is difficult for us to use traditional analysis tools to make stock transaction decision. With the developmenting of nonlinear methods such as neural networks and fuzzy neural networks, we can now use these methods for stock price forecasting.
In this research, we presented three scenarios: 1) stock price forecasting with classical methods approach, 2) stock price forecasting with artificial intelligence methods approach, and 3) stock price forecasting with hybrid model. Therefore, first, we designed classical models such as exponential smoothing, trend analysis, and ARIMA, then artificial intelligent models such as neural networks and fuzzy neural networks were designed, next the third scenario i.e., hybrid model, was presented. Finally, the scenarios were measured. The experimental result showed that the hybrid model had more accuracy than the classical or artificial intelligent models.
The hybrid model enjoyed also such properties as fast convergence, high precision and strong function approximation ability, there fore, it is suitable for real stock price forecasting.
Mohammad Reza Sadeghi Moghadam; Amir Afsar; Babak Sohrabi
Volume 10, Issue 20 , June 2006, , Pages 212-226
Abstract
Among flows in every supply chain (finance, information, and material), material flow according to its part in product cost is very important. Many studies in the field of material flow showed that material flows are often single level. This study offers a comprehensive model for appropriate order assignment ...
Read More
Among flows in every supply chain (finance, information, and material), material flow according to its part in product cost is very important. Many studies in the field of material flow showed that material flows are often single level. This study offers a comprehensive model for appropriate order assignment in different levels of supply chain according to minimizing related cost to with genetic algorithm method. Results based on genetic algorithm is compared to other common pattern search methods such as Latin Hypercub and Mead-Nelder. The evidence shows that genetic algorithm method is the best method among other methods.
