https://dl.ucsc.cmb.ac.lk/jspui/handle/123456789/4578 2026-04-24T17:17:28Z 2026-04-24T17:17:28Z Wijayarathne, W.M.S.M.M https://dl.ucsc.cmb.ac.lk/jspui/handle/123456789/4600 2022-06-17T17:27:59Z 2022-06-17T00:00:00Z

Title: Churn Prediction of Fiber to Home Users Authors: Wijayarathne, W.M.S.M.M Abstract: Due to the competitive strategies of service providers, Customers actively swap from one service provider to another to satisfies their needs which triggers them to ‘churn’. Hence, It’s time for the telecommunication industry to make necessary predictive decisions to increase the ‘happiness factor of customers’ and ‘Retaining the customers with stabilizing their market value.’This study discusses the strategies that can predict the Churn and the remedies of minimizing the churn factor of the fiber to home users. That could help relevant stakeholders to take a suitable decision at the right time and mitigate customers from churning, thereby reducing the churn rate. 14,461 instances related to a telecom sector are used in this study with fourteen network-related feature parameters for developing a churn prediction model for fiber to the home users, in which some of the features are numerical and some of the features are categorical. Among these features, eight significant features are selected using the Recursive Feature Elimination Technique. These variables are used as input variables for Logistic Regression classifier to build a churn prediction model. Model performance is measured using Recall, Precision, F1-score, Support, confusion Matrix, and ROC curve. It was able to achieve significant accuracy from the model with 0.86 area under the curve. Market research survey was carried out to observe why people are unhappy with fiber to home services. These results are used to develop business and technical strategies to retain the fiber to the home customers.

2022-06-17T00:00:00Z Weerasinghe https://dl.ucsc.cmb.ac.lk/jspui/handle/123456789/4599 2022-06-17T17:22:04Z 2022-06-17T00:00:00Z

Title: Early Prediction of Customer Abandonment in E-Business Authors: Weerasinghe Abstract: Customer churn, customer retention and attrition are topics which have been discussed and studied in so many researches. However, most of them have evaluated the churn prediction only based on the history data. Here in this research conduct, it is intended to derive results based on the customer segmentation. Considering the whole customer base might deliver the desirable output yet, when applying the customer segmentation based on the primary customer types, the final outcome will be more precise. Considering the customer type such as, loyalty and seasonal customers who really gives a better return on investment for the business conduced more customer churn and retention evaluation. By referring the E-Business dataset specifically with the access mode and channels of the customers, the tendency of leaving the system with the possible way is to be evaluated. Getting into the term ‘Churn’, this study has aware of the nature of the detachment considering the access channel or method of the customer as well. The suggested methodology would be approaching in two different paths in order to evaluate the best fit and the performance by measuring the accuracy and the reusability of the model. Aligning with two methods namely logistic regression and deep neural network address artificial neural network, the predictive model would be implemented. Specifically with the use of RFM Analysis, the study will be directed to customer segmentation. The segmentation would be involved in clustering for the customer segmentation and as for the predictive model, classification model would be used with class variable for identification of the user churn. Here in this study, main focus would be designing and building predictive models in different aspects under the similar criteria and mostly the evaluating the usefulness of segmentation of the customer over the total customer base that has been dedicated under specific clusters identified with the given set of criteria.

2022-06-17T00:00:00Z Wajirasena, W.A. https://dl.ucsc.cmb.ac.lk/jspui/handle/123456789/4598 2022-06-10T17:31:25Z 2022-06-10T00:00:00Z

Title: Crowd Behaviour Monitoring Using Aerial Surveillance Authors: Wajirasena, W.A. Abstract: With the rise of population and ever-growing megacities, it is essential for the authorities to monitor the crowd movements. Controlling crowd or mass gatherings at special events such as entertainment events or sport events and places that are essential to the daily lifestyle of people such as airports, hospitals is essential in the modern day. In this thesis we present two methods based on human action detection and crowd density prediction to monitor crowd behaviour.

2022-06-10T00:00:00Z Thilakaratne, P.H. https://dl.ucsc.cmb.ac.lk/jspui/handle/123456789/4597 2022-06-10T17:26:31Z 2022-06-10T00:00:00Z

Title: Student Result Predictor Using Machine Learning Techniques Authors: Thilakaratne, P.H. Abstract: Education is one of the most important aspect in human life. As human all of us spending twelve to thirteen years at schools and then passing through to higher education institutes such as universities. In universities student face exams and some of them get through it and others get stuck. This study will predict student academic performance with the help of machine learning algorithms. This study will discuss about four classification algorithms such as Decision tree classifier, Support Vector Machine Classifier, Naïve Bayes and Random Forest classifier. This study will address which would be the optimum algorithm can be used to predict student results under identified parameters.

2022-06-10T00:00:00Z