• Asia pacific journal of information systems
• /
• v.12 no.1
• /
• pp.139-158
• /
• 2002

As the telecommunication market becomes mature in Korea, severe competition has already begun on the market. While service providers struggled for the last couple of years to acquire as many new customers as possible, nowadays they are making more efforts on retaining the current customers. The churn management by analyzing customers' demographic and transactional data becomes one of the key customer retention strategies which most companies pursue. However, the customer data analysis has still remained at the basic level in the industry, even though it has considerable potential as a tool for understanding customer behavior. This paper develops several churn prediction models using data mining techniques such as logistic regression, decision trees, and neural networks. For model-building, real data were used which were collected from one of the major telecommunication companies in Korea. This paper explores various ways of comparing model performance, while the hit ratio was mainly focused in the previous research. The comparison criteria used in this study include gain ratio, Kolmogorov-Smirnov statistics, distribution of the predicted values, and explanation ability. This paper also suggest some guidance for model selection in applying data mining techniques.

• International Journal of Computer Science & Network Security
• /
• v.23 no.8
• /
• pp.190-198
• /
• 2023

To enhance customer satisfaction for higher profits, an e-commerce sector can establish a continuous relationship and acquire new customers. Utilize machine-learning models to analyse their customer's behavioural evidence to produce their competitive advantage to the e-commerce platform by helping to improve overall satisfaction. These models will forecast customers who will churn and churn causes. Forecasts are used to build unique business strategies and services offers. This work is intended to develop a machine-learning model that can accurately forecast retainable customers of the entire e-commerce customer data. Developing predictive models classifying different imbalanced data effectively is a major challenge in collected data and machine learning algorithms. Build a machine learning model for solving class imbalance and forecast customers. The satisfaction accuracy is used for this research as evaluation metrics. This paper aims to enable to evaluate the use of different machine learning models utilized to forecast satisfaction. For this research paper are selected three analytical methods come from various classifications of learning. Classifier Selection, the efficiency of various classifiers like Random Forest, Logistic Regression, SVM, and Gradient Boosting Algorithm. Models have been used for a dataset of 8000 records of e-commerce websites and apps. Results indicate the best accuracy in determining satisfaction class with both gradient-boosting algorithm classifications. The results showed maximum accuracy compared to other algorithms, including Gradient Boosting Algorithm, Support Vector Machine Algorithm, Random Forest Algorithm, and logistic regression Algorithm. The best model developed for this paper to forecast satisfaction customers and accuracy achieve 88 %.

• Journal of the Korea Safety Management & Science
• /
• v.9 no.3
• /
• pp.111-118
• /
• 2007

What is CRM(Customer Relationship Management) means that planning, executing, and re-accessing the marketing strategy based on the customer character by analyzing the material related to customers. That is CRM is a strategy of customer service on the base of data. In the case of the telecommunications and a newspaper, there are restricted application of CRM, because they are provided services by paying a given amount of money within a given period of time. This paper develops CRM model(chum prediction model) that can apply to a newspaper. For model-building, real data were used which were collected from one of the major a newspaper company in Korea. Also, this paper verifies the efficient result.

• Transactions of the Korean Society for Noise and Vibration Engineering
• /
• v.21 no.10
• /
• pp.916-923
• /
• 2011

The refrigerant after the expansion valve interchanges the heat at the evaporator. At this moment, the state of gas and liquid becomes two-phase flow and causes irregular noise. In order to avoid the noise, the two-phase flow pattern should be predicted. In this paper, the procedure to predict the two-phase flow patterns such as churn flow and annular flow was suggested using the CFD software. The experiments using refrigerant-supplying equipment was carried out and the noise levels according to the flow pattern were measured. The flow patterns predicted by this procedure showed good agreement with those by experiments. The churn flow is noisier than annular flow pattern.

• Journal of Intelligence and Information Systems
• /
• v.19 no.2
• /
• pp.125-140
• /
• 2013

We call a data set in which the number of records belonging to a certain class far outnumbers the number of records belonging to the other class, 'imbalanced data set'. Most of the classification techniques perform poorly on imbalanced data sets. When we evaluate the performance of a certain classification technique, we need to measure not only 'accuracy' but also 'sensitivity' and 'specificity'. In a customer churn prediction problem, 'retention' records account for the majority class, and 'churn' records account for the minority class. Sensitivity measures the proportion of actual retentions which are correctly identified as such. Specificity measures the proportion of churns which are correctly identified as such. The poor performance of the classification techniques on imbalanced data sets is due to the low value of specificity. Many previous researches on imbalanced data sets employed 'oversampling' technique where members of the minority class are sampled more than those of the majority class in order to make a relatively balanced data set. When a classification model is constructed using this oversampled balanced data set, specificity can be improved but sensitivity will be decreased. In this research, we developed a hybrid model of support vector machine (SVM), artificial neural network (ANN) and decision tree, that improves specificity while maintaining sensitivity. We named this hybrid model 'hybrid SVM model.' The process of construction and prediction of our hybrid SVM model is as follows. By oversampling from the original imbalanced data set, a balanced data set is prepared. SVM_I model and ANN_I model are constructed using the imbalanced data set, and SVM_B model is constructed using the balanced data set. SVM_I model is superior in sensitivity and SVM_B model is superior in specificity. For a record on which both SVM_I model and SVM_B model make the same prediction, that prediction becomes the final solution. If they make different prediction, the final solution is determined by the discrimination rules obtained by ANN and decision tree. For a record on which SVM_I model and SVM_B model make different predictions, a decision tree model is constructed using ANN_I output value as input and actual retention or churn as target. We obtained the following two discrimination rules: 'IF ANN_I output value <0.285, THEN Final Solution = Retention' and 'IF ANN_I output value ${\geq}0.285$, THEN Final Solution = Churn.' The threshold 0.285 is the value optimized for the data used in this research. The result we present in this research is the structure or framework of our hybrid SVM model, not a specific threshold value such as 0.285. Therefore, the threshold value in the above discrimination rules can be changed to any value depending on the data. In order to evaluate the performance of our hybrid SVM model, we used the 'churn data set' in UCI Machine Learning Repository, that consists of 85% retention customers and 15% churn customers. Accuracy of the hybrid SVM model is 91.08% that is better than that of SVM_I model or SVM_B model. The points worth noticing here are its sensitivity, 95.02%, and specificity, 69.24%. The sensitivity of SVM_I model is 94.65%, and the specificity of SVM_B model is 67.00%. Therefore the hybrid SVM model developed in this research improves the specificity of SVM_B model while maintaining the sensitivity of SVM_I model.

• Proceedings of the Korean Nuclear Society Conference
• /
• 1996.05b
• /
• pp.586-591
• /
• 1996

Through the analysis of many experimental post-dryout data, it is shown that the most probable flow regime near dryout or quench front is not annular flow but churn-turbulent flow when the mass flux is low. A correlation describing the initial droplet size just after the CHF position at low mass flux is suggested through regression analysis. In the post-dryout region at low pressure and low flow, it is found that the suggested one-dimensional mechanistic model is not applicable when the vapor superficial velocity is very low, i.e., when the flow is bubbly or slug flow regime. This is explained by the change of main entrainment mechanism with the change of flow regime. Therefore, the suggested correlation is valid only in the churn-turbulent flow regime ( $j_{g}$ $^{*}$＝0.5~4.5).).

• Proceedings of the Korea Technology Innovation Society Conference
• /
• 2003.11a
• /
• pp.115-133
• /
• 2003

본 연구에서는 번호이동성 제도의 시차 도입이 이동통신시장에 미치는 영향과 동 제도와 관련된 경제적 후생효과를 추정하였다. 전문조사기관이 조사한 churn rate를 이용하여 2004년-2007년간 사업자별 시장점유율 추정결과는 2004년에는 선발사업자의 점유율은 낮아지고 후발사업자의 점유율은 높아져 유효경쟁체제를 구축하는 방향으로 움직이지만 churn in/out 효과가 모두 반영되는 2005년 이후 장기적으로는 정부의 정책목표가 달성되지 않을 수도 있음을 보이고 있다. 한편 번호이동성이 가져올 경제적 후생효과로는 동 제도를 일시 도입하는 경우 약 3,323억원의 후생이 발생하고, 6개월 시차 도입하는 경우에는 경쟁활성화 효과를 무시하면 이보다 7.85% 감소한 약 3,062억원으로 추정되었다. 또한 요금인하 수준이 분당 0.3원 정도 되면 시차 도입으로 인한 경쟁 효과가 시차제 적용으로 인한 후생 손실을 보전할 수 있는 것으로 나타났다.

• IE interfaces
• /
• v.18 no.3
• /
• pp.277-287
• /
• 2005

In this paper, we deal with a new techno-economic modeling and analysis for optical access networks. In deploying the fiber-to-the-home (FTTH) architecture, network planner needs to consider the following techno-economic issues: when do we need to upgrade existing local access network to FTTH network? how much do we invest to maximize profit? In order to answer these techno-economic questions, we need to consider the impact of emerging technologies and business environment. Toward this end, we develop a new techno-economic modeling to deal with the inherent complexity of technology evolution and cost economics. In particular, the new modeling approach provides us with an techno-economic analysis of technology alternatives such as ethernet passive optical network (E-PON) and wavelength division multiplex passive optical network (WDM-PON). In this analysis, we focus on the impact of critical factors such as the cost characteristic of proposed architecture and digital subscriber line (DSL) subscriber's churn-in to FTTH service and churn-out. We develop mixed integer-programming models for finding the evolution path of local access networks to broadband network architectures.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2003.10a
• /
• pp.27-30
• /
• 2003

The introduction of reseller In Australian Fixed Telecommunication Market resulted in the reduction of Telstra's local call market share by 13 percent and average fall rate was reduced. Thus, Telstra increased basic rate at 14.5 percent to compensate loss revenue in the local call market. With the deployment of carrier pre-selection of long distance and international calls, it reduced long distance rate at 23.5 percent and international tall rate at 53 percent, and increased the Churn rate. Therefore, the deployment of service-based competition brought efficient results for long distance and international call market. However, LM market created 13.4 percent reduction in call rates, complications in charge system, technical barriers and the preference of one-bill by customers.

• Journal of Mechanical Science and Technology
• /
• v.15 no.7
• /
• pp.941-950
• /
• 2001

A study of counter-current two-phase flow in narrow rectangular channels has been performed. Two-phase flow regimes were experimentally investigated in a 760mm long and 100mm wide test section with 2.0 and 5.0mm gap widths. The resulting flow regime maps were compared with the existing transition criteria. The experimental data and the transition criteria of the models showed relatively good agreement. However, the discrepancies between the experimental data and the model predictions of the flow regime transition become pronounced as the gap width increased. As the gap width increased the transition gas superficial velocities increased. The critical void fraction for the bubbly-to-slug transition was observed to be about 0.25. The two-phase distribution parameter for the slug flow was larger for the narrower channel. The uncertainties in the distribution parameter could lead to a disagreement in slug-to-churn transition between the experimental findings and the transition criteria. For the transition from churn to annular flow the effect of liquid superficial velocity was found to be insignificant.