تجزیه و تحلیل اعتماد کاربر در بانکداری الکترونیکی با استفاده از روش دادهکاوی
|کد مقاله||سال انتشار||مقاله انگلیسی||ترجمه فارسی|
|22299||2013||9 صفحه PDF||23 صفحه WORD|
Publisher : Elsevier - Science Direct (الزویر - ساینس دایرکت)
Journal : Expert Systems with Applications, Volume 40, Issue 14, 15 October 2013, Pages 5439–5447
اثر بحران اعتماد در بانکداری الکترونیکی و مسئله تقلب
اعتماد به دست آمده در بانکداری الکترونیکی
کار میدانیِ مطالعه و جمع آوری اطلاعات
معیارهای ارزیابی جوابها
الگوریتم ژنتیکِ انتخاب چند هدفه: MSGA
نتیجهگیری، محدودیتها و تحقیقات بیشتر
The potential fraud problems, international economic crisis and the crisis of trust in markets have affected financial institutions, which have tried to maintain customer trust in many different ways. To maintain these levels of trust they have been forced to make significant adjustments to economic structures, in efforts to recoup their investments and maintain the loyalty of their customers. To achieve these objectives, the implementation of electronic banking for customers has been considered a successful strategy. The use of electronic banking in Spain in the last decade has been fostered due to its many advantages, giving rise to real integration of channels in financial institutions. This paper reviews different methods and techniques to determine which variables could be the most important to financial institutions in order to predict the likely levels of trust among electronic banking users including socio-demographic, economic, financial and behavioural strategic variables that entities have in their databases. To do so, the most recent advances in machine learning and soft-computing have been used, including a new selection operator for multiobjective genetic algorithms. The results obtained by the algorithms were validated by an expert committee, ranking the quality of them. The new methodology proposed, obtained the best results in terms of optimisation as well as the highest punctuation given by the experts.
The behaviour of the financial system against the economic crisis has been different among the countries within the European Union. While many international institutions focused their interest on credit and risk transfer, neglecting customer service, the banking sector continued to have an extensive network of offices through which to distribute financial products and to foster close client relationships. This very competitive environment forced banks to strictly control costs, which has made the financial system one of the world’s most efficient (Álvarez, 2008). Despite these advantages, the Spanish financial system was also in a precarious position particularly due to its exposure in real estate. In the latter part of the 90’s and in the early part of the last decade there was an excess supply of real estate and therefore a large demand for financing. This situation forced financial institutions to go to wholesale markets since domestic markets did not have the resources to cover as much investment as was being generated. Due to this and the pressing international crisis, the government and the Central Bank had to intervene different economies, among them, the Spanish (Liébana-Cabanillas, Martínez-Fiestas, & Rejón-Guardia, 2011). The Spanish financial sector has already started to change as a result of this situation thanks to the Bank Restructuring Fund (FROB1), and new regulations which will be introduced in 2013 with the advent of Basilea III2 and more recently the Royal Decree for restructuring of the Spanish Savings Banks. According to the latest report on “Individual Financial Behaviour in Spain 2009” developed by Inmark (2009), 55.1% of the sample says their trust in the Spanish financial sector has worsened compared with 0.9% stating that it has improved and 40.1% who say there has been no change. In this complicated situation, the Spanish financial system has had to make technological improvements to reduce costs and optimize investments. Of all the available tools used to achieve these objectives, electronic banking has been the most widely implemented. Traditionally, financial products and services have been distributed through bank branches due to their proximity to customers, the large number of services they perform, the added value that the client receives at the branch, and the important role bank branches play in decisions made by customers. In spite of this, however, this conventional channel has begun to be replaced by a more agile and dynamic channel as reflected in the data of the World Retail Banking Report3 (2010) on the percentage of use of the main channels (see Fig. 1).From the 90s to the present, electronic banking has become the distribution channel with the greatest potential for financial institutions (Karjaluoto, Mattila, & Pento, 2002). Currently, the majority of companies offer their customers access to most of their services through this channel. Therefore, electronic banking has become a crucial service by which to gain customer satisfaction and loyalty and establish closer customer relationships, thereby meeting user expectations (Azcorra et al., 2001, Berrocal, 2009, Climent and Momparles, 2006 and Hsu, 2008). Thus, the primary alternative channel to the traditional bank branch is electronic banking as it has many advantages for customers including convenience, global access, availability, cost and time-savings, information transparency, choice and comparison, customization, and financial innovation (Delgado and Nieto, 2002 and Muñoz-Leiva, 2008). However, this service also has some drawbacks, mainly related to trust and security. But trust, together with satisfaction, is considered one of the key elements in building long-term relationships, a fundamental business strategy in the current economic situation (García et al., 2008b and Lam et al., 2004). In this context, this paper reviews different methods and techniques to determine which variables could be the most important to financial institutions in order to predict the likely levels of trust among electronic banking users including socio-demographic, economic, financial and behavioural strategic variables that entities have in their database. This paper is organized as follows. Section 2 describes the Electronic Banking at the European Union as well as the concept of costumer trust. Section 3 introduces the research methodology, the data description, and the models and algorithms used to analyse the data. Then, Section 4, presents the results of the analysis and its validation by a set of expert. Finally, Section 5, draws the conclusions from a business management perspective and further work.
نتیجه گیری انگلیسی
During the last decade the Spanish financial system has undergone a profound transformation with the aim of reducing the bank usage rate in order to save costs. One of the most important changes in the sector has been the revolution of electronic banking among customers although the reduction was not as high as expected. Nonetheless, electronic banking currently penetrates close to 50% of Internet users and 20% of the total population according to different sources, indicating significant societal acceptance of this service. This has forced banks to make major investments to maintain and enhance user satisfaction with this channel that is changing the traditional concept of financial institutions. The large number of advantages and few drawbacks electronic banking offers demonstrates its importance. Considering this, the work presented aims to shed some light in this area by using several machine learning methods over a data set that was collected by a financial entity. The data consisted on information about the clients, the offices and an evaluation of the electronic banking platform performed by the costumers. The goal was to identify which variables could influence more costumer’s trust. As the variable selection problem has been studied previously in the literature, multi-objective genetic algorithms were applied using several fitness functions: Test Delta and Mutual Information. Two new operators were defined in order to obtain solutions more significant and valuable from the expert’s point of view, which lead to better results. From the analysis it is interesting how Mutual Information obtains higher scores, indicating that the Test Delta might not be such a good indicator since it tends to select the less noisy variable instead of the most significant one. Therefore, to apply this new techniques and algorithms in the process of decision taken helps to identify which variables could be the most important ones and focus on improving those to increase costumer’s trust in electronic banking. The process is feasible due to the large amount of information that financial companies have about their clients. In recent years the financial sector has carried out customer loyalty campaigns at different levels. On one hand, management systems such as CRM, together with general communication campaigns about the activities of the entity have increased the involvement of customers through the branches. On the other hand, most entities link credit concessions and interest rates extra for customer deposits to other products favouring up-selling strategies (cross-selling, reducing product and service cancellations and improving client relationships). For these reasons it is logical that three bidirectional goals are accomplished with this strategy; increased cross sales, increased asset balance and therefore the customer’s gross profitability for the entity. For these reasons proper segmentation of customers is vital to optimal management of electronic banking services. When an organization knows different customer profiles and the variables that define them, it can anticipate needs and achieve increased profitability and improved levels of customer trust leading to greater brand loyalty. Finally, some future work goes through the analysis of regression models and how they behave using the solutions provided by the variable selection algorithms as well as the study of the expert’s opinion regrading the interpretability of those models