• Journal of Korean Institute of Industrial Engineers
• /
• v.41 no.1
• /
• pp.10-16
• /
• 2015

In this study, we suggest a method to predict probability distribution of a new customer's degree of loyalty using C-CRF that reflects the RFM score and similarity to the neighbors of the customer. An RFM score prediction model is introduced to construct the first feature function of C-CRF. Integrating demographical similarity, purchasing characteristic similarity and purchase history similarity, we make a unified similarity variable to configure the second feature function of C-CRF. Then parameters of each feature function are estimated and we train our C-CRF model by training data set and suggest a probabilistic distribution to estimate a new customer's degree of loyalty. An example is provided to illustrate our model.

• Journal of Distribution Science
• /
• v.20 no.6
• /
• pp.1-10
• /
• 2022

Purpose: This research proposes a new modified Recency-Frequency-Monetary (RFM) model by extending the model with spatial analysis for supporting decision-makers in discovering the promotional target market. Research design, data and methodology: This quantitative research utilizes data-mining techniques and the RFM model to cluster a university's provider schools. The RFM model was modified by adapting its variables to the university's marketing context and adding a district's potential (D) variable based on heatmap analysis using Geographic Information System (GIS) and K-means clustering. The K-prototype algorithm and the Elbow method were applied to find provider school clusters using the proposed RFM-D model. After profiling the clusters, the target segment was assigned. The model was validated using empirical data from an Indonesian university, and its performance was compared to the Customer Lifetime Value (CLV)-based RFM utilizing accuracy, precision, recall, and F1-score metrics. Results: This research identified five clusters. The target segment was chosen from the highest-value and high-value clusters that comprised 17.80% of provider schools but can contribute 75.77% of students. Conclusions: The proposed model recommended more targeted schools in higher-potential districts and predicted the target segment with 0.99 accuracies, outperforming the CLV-based model. The empirical findings help university management determine the promotion location and allocate resources for promotional information distribution and student recruitment.

• Journal of the Korea Society of Computer and Information
• /
• v.15 no.12
• /
• pp.227-235
• /
• 2010

This paper proposes the recommendation system which is advanced using RFM method and Association Rules in e-Commerce. Using a implicit method which is not used user's profile for rating, it is necessary for user to keep the RFM score and Association Rules about users and items based on the whole purchased data in order to recommend the items. This proposing system is possible to advance recommendation system using RFM method and Association Rules for cross-selling, and also this system can avoid the duplicated recommendation by the cross comparison with having recommended items before. And also, it's efficient for them to build the strategy for marketing and crm(customer relationship management). It can be improved and evaluated according to the criteria of logicality through the experiment with dataset collected in a cosmetic cyber shopping mall. Finally, it is able to realize the personalized recommendation system for one to one web marketing in e-Commerce.

• Journal of the Korea Society of Computer and Information
• /
• v.13 no.2
• /
• pp.41-50
• /
• 2008

This paper proposes the recommendation system which is a new method using RFM method in mobile internet environment. Using a implict method which is not used user's profile for rating, is not used complicated query processing of the request and the response for rating, it is necessary for user to keep the RFM score about users and items based on the whole purchased data in order to recommend the items. As there are some problems which didn't exactly recommend the items with high purchasablity for new customer and new item that do not have the purchase history data. in existing recommendation systems, this proposing system is possible to solve existing problems, and also this system can avoid the duplicated recommendation by the cross comparison with the purchase history data. It can be improved and evaluated according to the criteria of logicality through the experiment with dataset, collected in a cosmetic cyber shopping mall. Finally, it is able to realize the personalized recommendation system with high purchasablity for one to one web marketing through the mobile internet.

• Journal of Intelligence and Information Systems
• /
• v.9 no.3
• /
• pp.117-135
• /
• 2003

This study is to propose a procedure solving scale problem of traditional collaborative filtering (CF) approach. The CF approach generally uses some similarity measures like correlation coefficient. So, as the user of the Website increases, the complexity of computation increases exponentially. To solve the scale problem, this study suggests a clustering model-based approach using Self-Organizing Map (SOM) and RFM (Recency, Frequency, Momentary) method. SOM clusters users into some user groups. The preference score of each item in a group is computed using RFM method. The items are sorted and stored in their preference score order. If an active user logins in the system, SOM determines a user group according to the user's characteristics. And the system recommends items to the user using the stored information for the group. If the user evaluates the recommended items, the system determines whether it will be updated or not. Experimental results applied to MovieLens dataset show that the proposed method outperforms than the traditional CF method comparatively in the recommendation performance and the computation complexity.

• 한국데이터정보과학회:학술대회논문집
• /
• 2006.04a
• /
• pp.109-120
• /
• 2006

Huge information has been made due to the current computing environment and could not be acceptable. People want the information which they can understand and accept easily. They may want not only simple information but also knowledge. That is why data mining becomes a center of information. We use RFM analysis in order to create customer score. Customers are classified into five groups(most oxcellenrexcellenycommoflowerilowest) for a various marketing activities. We can found the significant patterns in each group, and classify customers from loyal customers to leaving customers in the near future by the indirect data mining(e.g. association analysis) and the direct data mining(e.g. decision tree, logistic regression analysis, etc.), which are named in this study. Our research focuses on the advanced models by applying the association rules in data mining. Our results indicate that the indirect data mining and the direct data mining seem to have same outputs, but the former shows more clear pattern then the latter one.

• Nutrition Research and Practice
• /
• v.10 no.4
• /
• pp.404-410
• /
• 2016

BACKGROUND/OBJECTIVE: Malnutrition and inflammation are reported as the most powerful predictors of mortality and morbidity in hemodialysis (HD) patients. Diet has a key role in modulating inflammation and dietary inflammatory index (DII) is a new tool for assessment of inflammatory potential of diet. The aim of this study was to evaluate the application of DII on dietary intake of HD patients and examine the associations between DII and malnutrition-inflammation markers. SUBJECTS/METHODS: A total of 105 subjects were recruited for this cross-sectional study. Anthropometric measurements, 3-day dietary recall, and pre-dialysis biochemical parameters were recorded for each subject. Subjective global assessment (SGA), which was previously validated for HD patients, and malnutrition inflammation score (MIS) were used for the diagnosis of protein energy wasting. DII was calculated according to average of 3-day dietary recall data. RESULTS: DII showed significant correlation with reliable malnutrition and inflammation indicators including SGA (r = 0.28, P < 0.01), MIS (r = 0.28, P < 0.01), and serum C-reactive protein (CRP) (r = 0.35, P < 0.001) in HD patients. When the study population was divided into three subgroups according to their DII score, significant increasing trends across the tertiles of DII were observed for SGA score (P = 0.035), serum CRP (P = 0.001), dietary energy (P < 0.001), total fat (P < 0.001), saturated fatty acids (P < 0.001), polyunsaturated fatty acids (P = 0.006), and omega-6 fatty acids (P = 0.01) intakes. CONCLUSION: This study shows that DII is a good tool for assessing the overall inflammatory potential of diet in HD patients.