Journal of Student Research Exploration

Customer churn prediction in the case of telecommunication company using support vector machine (SVM) method and oversampling

2023-12-17T04:22:10+00:00

hurn is the act by which a customer withdraws from service, including service provider-initiated churn and customer-initiated churn. Churn is a big challenge for companies, especially churn-prone enterprise sectors such as telecommunications. Churn can affect both revenue and reputation if occurs for negative reasons. This study aims to predict customer churn in a telecommunication company dataset, investigating the impact of various variables and classes on churn occurrences to inform strategic decision-making for businesses. The Support Vector Machine (SVM) model is employed, and dataset imbalance is addressed through oversampling techniques, specifically Synthetic Minority Over-sampling Technique (SMOTE) and random oversampling (ROS). Three SVM models are created with different training datasets (normal, SMOTE, ROS), yielding varying results. The normal dataset achieves the highest accuracy at 92%, outperforming SVM with ROS (89%) and SVM with SMOTE (87%). However, the normal dataset exhibits lower sensitivity compared to both oversampling techniques. The study identifies the cause of decreased accuracy in oversampling and low sensitivity in the normal dataset. The novelty of this research lies in testing the SVM model's ability to surpass the accuracy of previous models on the same dataset and in exploring the unique impact of oversampling in churn prediction.

Classification of risk of death from heart disease or cigarette influence using the k-nearest neighbors (KNN) method

2024-06-26T07:09:35+00:00

Heart disease is one of the leading causes of death in Indonesia. In addition to coronary heart disease, smoking is the leading contributor to the death rate in Indonesia. This study aims to analyze the risk of death with the main variables of heart disease history and smoking history. This study classifies the risk of death of heart disease sufferers and smokers using the KNearest Neighbors (KNN) algorithm. The results showed that the KNN model had an accuracy of 52.38% in predicting the risk of death of smokers and heart disease patients. Confusion matrix analysis revealed that the model performed well in predicting classes 0 and 2, but had difficulty in predicting class 1. This study shows that KNN can be used to predict the risk of death of smokers and patients with heart disease with a satisfactory success rate.

Classification of travel class with k-nearest neighbors algorithm using rapidminer

2024-06-26T07:12:54+00:00

he tourism industry in Indonesia plays an important role in the national economy. The selection of travel class according to the needs and budget of tourists is an important aspect in the tourism industry. This research aims to develop a travel class classification model using dummy datasets and the K-Nearest Neighbors (KNN) algorithm with RapidMiner software. The travel class dummy data set was obtained from the internet and modified according to research needs. The KNN algorithm was used to classify new travel classes based on previously classified dummy data. These dummy data were preprocessed and analyzed using RapidMiner software. The performance of the KNN model was evaluated using accuracy, precision, recall and F1-score. The results showed that the KNN algorithm with the values k = 1-2, k = 3-6, k = 8-10, k = 11-14 and k = 15 resulted in accuracy of 35.71%, 39.29%, 48.26%, 46.43% and 50.00%, respectively. This shows that the KNN algorithm with a value of k=15 produces the highest accuracy that can be effectively used to classify new travel classes based on dummy data.

Analysis of k-means clustering algorithm in advanced country clustering using rapid miner

2024-07-03T13:15:51+00:00

In the era of globalization, the understanding of developed countries is no longer limited to the level of per capita income alone. As part of the analysis of developed countries based on aspects of government revenue, income balance, national savings, and domestic output based on sales. This research aims to cluster and to find out how these economic indicators are interrelated and affect the status of a country as a developed country. The K-means algorithm is used to identify patterns of countries with similar economic characteristics. From the research conducted, there are 4 clusters generated based on the characteristics of developed countries.

Online payment fraud prediction with machine learning approach using naive bayes algorithm

2024-06-11T02:16:25+00:00

The increase in e-commerce has provided easy access for the public, but it also opens up opportunities for fraud in online transactions. Payment fraud is also a problem that often arises in transactions through electronic media. This research aims to analyze payment fraud in e-commerce transactions. This research uses a machine learning approach using the Naive Bayes algorithm. This research uses online transaction datasets involving various attributes such as payment and shipping methods. The developed Naive Bayes model achieved an accuracy of 61.03% with K = 7. The evaluation shows a balance between precision (59.46%) and recall (62.05%), although this study is limited by data quality and basic assumptions of Naive Bayes. In future research, it is worth considering the use of additional features or more complex data processing to improve the performance of fraud detection in online transactions. This research provides important insights in the fight against financial crime in the context of electronic commerce.

Application of k-nearest neighbor algorithm in classification of engine performance in car companies using Rapidminer

2024-06-23T13:10:54+00:00

Implementation of the k-Nearest Neighbor (k-NN) algorithm in the classification of CAR Car company engine performance using RapidMiner software. The company's engine performance is a very important aspect in the automotive industry that greatly affects operational efficiency and customer satisfaction. As an effort to monitor and improve engine performance, classification is an important key to identify machines that are feasible and require repair. The dataset used is a generated dataset from the AI Chat GPT bot whose prompts have been adapted to the research needs. The k-NN algorithm was chosen due to its ability to produce accurate predictions. The k-NN classification method utilizes training and testing data and calculates the distance between the data to determine the appropriate class. The results of this study show excellent performance in terms of accuracy, precision, and recall. The highest accuracy is 90.62% at the value of k = 2. The highest precision and recall are 100% and 93.75% at the values of k = 2, k = 4, and k = 7.