Search results for: Churn%20Prediction.

Authors: John Hadden, Ashutosh Tiwari, Rajkumar Roy, Dymitr Ruta

Abstract:

The aim of this paper is to identify the most suitable model for churn prediction based on three different techniques. The paper identifies the variables that affect churn in reverence of customer complaints data and provides a comparative analysis of neural networks, regression trees and regression in their capabilities of predicting customer churn.

Keywords: Churn, Decision Trees, Neural Networks, Regression.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 3213

Authors: Damith Senanayake, Lakmal Muthugama, Laksheen Mendis, Tiroshan Madushanka

Abstract:

Customer churn prediction is one of the most useful areas of study in customer analytics. Due to the enormous amount of data available for such predictions, machine learning and data mining have been heavily used in this domain. There exist many machine learning algorithms directly applicable for the problem of customer churn prediction, and here, we attempt to experiment on a novel approach by using a cognitive learning based technique in an attempt to improve the results obtained by using a combination of supervised learning methods, with cognitive unsupervised learning methods.

Keywords: Growing Self Organizing Maps, Kernel Methods, Churn Prediction.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2504

Authors: Abbas Al-Refaie

Abstract:

The research examines the factors that affect customer churn (CC) in the Jordanian telecom industry. A total of 700 surveys were distributed. Cluster analysis revealed three main clusters. Results showed that CC and customer satisfaction (CS) were the key determinants in forming the three clusters. In two clusters, the center values of CC were high, indicating that the customers were loyal and SC was expensive and time- and energy-consuming. Still, the mobile service provider (MSP) should enhance its communication (COM), and value added services (VASs), as well as customer complaint management systems (CCMS). Finally, for the third cluster the center of the CC indicates a poor level of loyalty, which facilitates customers churn to another MSP. The results of this study provide valuable feedback for MSP decision makers regarding approaches to improving their performance and reducing CC.

Keywords: Cluster analysis, telecom industry, switching cost, customer churn.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2471

Authors: Emel Kızılkaya Aydogan, Mihrimah Ozmen, Yılmaz Delice

Abstract:

In recent days, there is a change and the ongoing development of the telecommunications sector in the global market. In this sector, churn analysis techniques are commonly used for analysing why some customers terminate their service subscriptions prematurely. In addition, customer churn is utmost significant in this sector since it causes to important business loss. Many companies make various researches in order to prevent losses while increasing customer loyalty. Although a large quantity of accumulated data is available in this sector, their usefulness is limited by data quality and relevance. In this paper, a cost-sensitive feature selection framework is developed aiming to obtain the feature reducts to predict customer churn. The framework is a cost based optional pre-processing stage to remove redundant features for churn management. In addition, this cost-based feature selection algorithm is applied in a telecommunication company in Turkey and the results obtained with this algorithm.

Keywords: Churn prediction, data mining, decision-theoretic rough set, feature selection.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1715

Authors: Alanoud Moraya Aldalan, Abdulaziz Almaleh

Abstract:

A crucial part of maintaining a customer-oriented business in the telecommunications industry is understanding the reasons and factors that lead to customer churn. Competition between telecom companies has greatly increased in recent years, which has made it more important to understand customers’ needs in this strong market. For those who are looking to turn over their service providers, understanding their needs is especially important. Predictive churn is now a mandatory requirement for retaining customers in the telecommunications industry. Machine learning can be used to accomplish this. Churn Prediction has become a very important topic in terms of machine learning classification in the telecommunications industry. Understanding the factors of customer churn and how they behave is very important to building an effective churn prediction model. This paper aims to predict churn and identify factors of customers’ churn based on their past service usage history. Aiming at this objective, the study makes use of feature selection, normalization, and feature engineering. Then, this study compared the performance of four different machine learning algorithms on the Orange dataset: Logistic Regression, Random Forest, Decision Tree, and Gradient Boosting. Evaluation of the performance was conducted by using the F1 score and ROC-AUC. Comparing the results of this study with existing models has proven to produce better results. The results showed the Gradients Boosting with feature selection technique outperformed in this study by achieving a 99% F1-score and 99% AUC, and all other experiments achieved good results as well.

Keywords: Machine Learning, Gradient Boosting, Logistic Regression, Churn, Random Forest, Decision Tree, ROC, AUC, F1-score.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 323

Authors: Ulas Vural, M. Ergun Okay, E. Mesut Yildiz

Abstract:

Telecommunication service providers demand accurate and precise prediction of customer churn probabilities to increase the effectiveness of their customer relation services. The large amount of customer data owned by the service providers is suitable for analysis by machine learning methods. In this study, expenditure data of customers are analyzed by using an artificial neural network (ANN). The ANN model is applied to the data of customers with different billing duration. The proposed model successfully predicts the churn probabilities at 83% accuracy for only three months expenditure data and the prediction accuracy increases up to 89% when the nine month data is used. The experiments also show that the accuracy of ANN model increases on an extended feature set with information of the changes on the bill amounts.

Keywords: Customer relationship management, churn prediction, telecom industry, deep learning, Artificial Neural Networks, ANN.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 688

Authors: F. Amin, S. Feizi

Abstract:

Big data has the potential to improve the quality of services; enable infrastructure that businesses depend on to adapt continually and efficiently; improve the performance of employees; help organizations better understand customers; and reduce liability risks. Analytics and marketing models of fixed and mobile operators are falling short in combating churn and declining revenue per user. Big Data presents new method to reverse the way and improve profitability. The benefits of Big Data and next-generation network, however, are more exorbitant than improved customer relationship management. Next generation of networks are in a prime position to monetize rich supplies of customer information—while being mindful of legal and privacy issues. As data assets are transformed into new revenue streams will become integral to high performance.

Keywords: Big Data, Next Generation Networks, Network Transformation.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2467

Authors: Stoyan Nedeltchev, Markus Schubert

Abstract:

The gas holdup fluctuations in a bubble column (0.15 m in ID) have been recorded by means of a conductivity wire-mesh sensor in order to extract information about the main transition velocities. These parameters are very important for bubble column design, operation and scale-up. For this purpose, the classical definition of the Shannon entropy was modified and used to identify both the onset (at U_G=0.034 m/s) of the transition flow regime and the beginning (at U_G=0.089 m/s) of the churn-turbulent flow regime. The results were compared with the Kolmogorov entropy (KE) results. A slight discrepancy was found, namely the transition velocities identified by means of the KE were shifted to somewhat higher (0.045 and 0.101 m/s) superficial gas velocities U_G.

Keywords: Bubble column, gas holdup fluctuations, Modified Shannon entropy, Kolmogorov entropy.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1648

Authors: Sadegh Papari, Mohammad Kazemeini, Moslem Fattahi

Abstract:

In the present study, a heterogeneous and homogeneous gas flow dispersion model for simulation and optimisation of a large-scale catalytic slurry reactor for the direct synthesis of dimethyl ether (DME) from syngas and CO2, using a churn-turbulent regime was developed. In the heterogeneous gas flow model the gas phase was distributed into two bubble phases: small and large, however in the homogeneous one, the gas phase was distributed into only one large bubble phase. The results indicated that the heterogeneous gas flow model was in more agreement with experimental pilot plant data than the homogeneous one.

Keywords: Modelling, Slurry bubble column, Dimethyl ether synthesis, Homogeneous gas flow, Heterogeneous gas flow

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2125

Authors: Usman Mahmood, Khawaja M. U. Suleman

Abstract:

A peer-to-peer storage system has challenges like; peer availability, data protection, churn rate. To address these challenges different redundancy, replacement and repair schemes are used. This paper presents a hybrid scheme of redundancy using replication and erasure coding. We calculate and compare the storage, access, and maintenance costs of our proposed scheme with existing redundancy schemes. For realistic behaviour of peers a trace of live peer-to-peer system is used. The effect of different replication, and repair schemes are also shown. The proposed hybrid scheme performs better than existing double coding hybrid scheme in all metrics and have an improved maintenance cost than hierarchical codes.

Keywords: Erasure Coding, P2P, Redundancy, Replication.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1666

Authors: Mujtaba Roshan, John A. Schormans

Abstract:

Quality of Experience (QoE) drives churn in the broadband networks industry, and good QoE plays a large part in the retention of customers. QoE is known to be affected by the Quality of Service (QoS) factors packet loss probability (PLP), delay and delay jitter caused by the network. Earlier results have shown that the relationship between these QoS factors and QoE is non-linear, and may vary from application to application. We use the network emulator Netem as the basis for experimentation, and evaluate how QoE varies as we change the emulated QoS metrics. Focusing on Video-on-Demand, we discovered that the reported QoE may differ widely for users of different age groups, and that the most demanding age group (the youngest) can require an order of magnitude lower PLP to achieve the same QoE than is required by the most widely studied age group of users. We then used a bottleneck TCP model to evaluate the capacity cost of achieving an order of magnitude decrease in PLP, and found it be (almost always) a 3-fold increase in link capacity that was required.

Keywords: Quality of experience, quality of service, packet loss probability, network capacity.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 893

Authors: Shyam Kumar, Nannuri Srinivasulu, Ashok Khanna

Abstract:

Bubble columns have a variety of applications in absorption, bio-reactions, catalytic slurry reactions, and coal liquefaction; because they are simple to operate, provide good heat and mass transfer, having less operational cost. The use of Computational Fluid Dynamics (CFD) for bubble column becomes important, since it can describe the fluid hydrodynamics on both local and global scale. Euler- Euler two-phase fluid model has been used to simulate two-phase (air and water) transient up-flow in bubble column (15cm diameter) using FLUENT6.3. These simulations and experiments were operated over a range of superficial gas velocities in the bubbly flow and churn turbulent regime (1 to16 cm/s) at ambient conditions. Liquid velocity was varied from 0 to 16cm/s. The turbulence in the liquid phase is described using the standard k-ε model. The interactions between the two phases are described through drag coefficient formulations (Schiller Neumann). The objectives are to validate CFD simulations with experimental data, and to obtain grid-independent numerical solutions. Quantitatively good agreements are obtained between experimental data for hold-up and simulation values. Axial liquid velocity profiles and gas holdup profiles were also obtained for the simulation.

Keywords: Bubble column, Computational fluid dynamics, Gas holdup profile, k-ε model.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2665

Authors: Priyadharshini Ahrumugam

Abstract:

Cooperative groups through much research have been recognized to churn remarkable achievements instead of solitary or individualistic efforts. Based on Johnson and Johnson’s model of cooperative learning, the five key components of cooperation are positive interdependence, face-to-face promotive interaction, individual accountability, social skills, and group processing. In 2011, the Malaysian Ministry of Higher Education (MOHE) introduced the Holistic Student Development policy with the aim to develop morally sound individuals equipped with lifelong learning skills. The Community Service project was included in the improvement initiative. The purpose of this study is to assess the relationship of team-based learning in facilitating particularly students’ positive interdependence and face-to-face promotive interaction. The research methods involve in-depth interviews with the team leaders and selected team members, and a content analysis of the undergraduate students’ reflective journals. A significant positive relationship was found between students’ progressive outlook towards teamwork and the highlighted two components. The key findings show that students have gained in their individual learning and work results through teamwork and interaction with other students. The inclusion of Community Service as a MOHE subject resonates with cooperative learning methods that enhances supportive relationships and develops students’ social skills together with their professional skills.

Keywords: Community service, cooperative learning, positive interdependence, teamwork.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2146