استفاده از شبکه های عصبی برای تشخیص مشتریان سودآور برای بازاریابی خدمات دندانپزشکی ـــ موردی از درمانگاه دندانپزشکی در تایوان
|کد مقاله||سال انتشار||مقاله انگلیسی||ترجمه فارسی||تعداد کلمات|
|2830||2009||10 صفحه PDF||سفارش دهید||6360 کلمه|
Publisher : Elsevier - Science Direct (الزویر - ساینس دایرکت)
Journal : Expert Systems with Applications, Volume 36, Issue 1, January 2009, Pages 199–208
The purpose of the research was the development of a neural networks model to recognize profitable customers for dental services marketing. Data set was built up from proprietary customer databases and survey of seven dental clinics in Taiwan. Multi-layer feed-forward neural networks with sigmoid function trained by back-propagation training algorithm were utilized to build the recognition model. The result reveals that the recognition accuracy of the test on the model is greater than that expected by chance. Meanwhile, a set of contribution weights representing the general importance of each independent variable was produced and their marketing implications were illustrated. This research confirms that the neural network model is useful in recognizing existing patterns of customers’ data. The advantages of using the model are highlighted and marketing implications are demonstrated. Authors believe that the model is useful and suitable as an analyzing tool for dental marketers on market strategy planning.
Since the 1970s, the medical care cost has been rapidly raised. The cost-containment mechanism and global budgeting system are used to arrest the untamed growth of medical care expenditures in many countries (Chu, 1992, Detsky et al., 1990, Henke et al., 1994, Lave et al., 1992, Redmon and Yakoboski, 1995 and Wiley, 1992). Nearly 98% of all care is reimbursed under Taiwanese National Health Insurance (TNHI) since its inception in 1995. The experience of cost-containment and global budgeting for dental care in Taiwan is different from other countries. Dental care is included as part of the benefit package in TNHI. Although the dental care expenditures are growing, they are capped by the allocated budget in different regions (Six Health Insurance Regions in Taiwan). Except for the co-payment and registration fee per visit, there is no out-of-pocket cost for the insured that receive care. Hence, a dentist’s income is limited. Therefore, providing dental cares which are not covered by TNHI and finding profitable customers have been urgent topics for increasing a dentist’s income. Customer centricity is to grow enduring relationships with profitable customers, and retaining important customers is crucial to business success (Biong and Selnes, 1996, Sollner, 1998 and Webster, 1992). How to retain profitable customers is the critical issue of relationship marketing. Segmentation is one of the most useful methods to discover profitable quotient. Michalski (1983) mentioned that the widely used traditional mathematical and statistical data analysis techniques, such as regression analysis, numerical taxonomy or factor analysis are not sufficiently powerful for the task of detecting interesting conceptual patterns or in revealing structure in a collection of observations (Michalski, 1983). The purpose of the research is to develop a neural networks model to recognize profitable customers for dental services marketing.
نتیجه گیری انگلیسی
A method for recognizing profitable customers by one class of neural networks was developed successfully and applied to recognize profitable customers properly for the operation of dental clinics. A questionnaire survey was conducted to collect data among consumers of seven dental clinics in Taiwan. Then, input variable generated from survey and customers’ database were used to recognize profitable customers. The correct recognition rate of the model is 90.18%. Meanwhile, the result was evaluated by the discriminant analysis. Press’s Q statistic of 65.63 is greater than the critical value of 6.63 (χ2 value with one degree of freedom is statistically significant at α of 0.01). That is, the recognition accuracy of the test on the model is greater than that expected by chance and it has higher accuracy than the method of discriminant analysis. A set of contribution weights representing the general importance of each independent variable was produced at the same time. Their marketing implications were discussed and competitive marketing strategies, which can fit better in the competitive business environment, could be made. Based on the results of the research, the evidence is enough to suggest that the neural network model is useful in recognizing existing patterns of the data. The advantages of using the model are highlighted and marketing implications are demonstrated. Authors believe that the model is useful and suitable as an analyzing tool for dental marketers on the market strategy planning. Meanwhile, the powerful recognition mechanism will be helped to understand what goes on between input and output. The result of the current study reveals that the proposed method could provide a heuristic model, because there is no model fit for any cases. (Lensberg, Eilifsen, & McKee, 2006).