1 October 2018 in Journal of Big Data
MHMedria Kusuma Dewi Hardhienata
Medria Kusuma Dewi Hardhienata
One of the challenges in diagnosing stroke disease is the lack of useful analysis tool to identify critical stroke data that contains hidden relationships and trends from a vast amount of data. In order to address this problem, we proposed Intuitionistic Fuzzy Based Decision Tree in order to diagnosis the different types of stroke disease. The approach is implemented by mapping observation data into Intuitionistic Fuzzy Set. These results lead to a compound of a membership function, non-membership function, and a hesitation degree for each record. The result of Intuitionistic Fuzzy is calculated using Hamming Distance as main requirement for Intuitionistic Fuzzy Entropy. The Hamming Distance calculate the difference between values on the same variable. Main advantage of this approach is that we can find out variables effected on the stroke disease using information gain derived from Intutionistics Entropy. Furthermore, the Intuitionistic Fuzzy based Decision Tree are able to provide plenty of information to stakeholders regarding the hidden facts of established rules and utilize linguistic terms to accommodate unclearness, ambiguity, and hesitation in human perception. The results of Intuitionistic Fuzzy Entropy determine the root and node in the formation of the decision tree model based on the information gain of variables in the data. In this study, simulation results show that the approach successfully determine 20 variables that directly influence stroke. These variables are used to classify the types of stroke. Furthermore, results show that the approach has resulted in 90.59% in classifying stroke disease. Results of the study also demonstrates that the approach produces the best diagnosis performance compared to the other two models according to the accuracy of classification from the type of stroke disease.
7 December 2023 in International Journal of Management Research and Economics
The main objective of this research is to determine the influence of digital marketing on purchasing decisions. This research also aims to determine whether brand image mediates the relationship between digital marketing and purchasing decisions. The required data was obtained from a questionnaire via an online survey on Google Form. After going through responses from 45 respondents using Buttonscarves products who had made purchases on the Buttonscarves website, the data was analyzed using SEM PLS 3.0. The result is that there is a direct influence of digital marketing on brand image, but there is no significant influence of digital marketing on purchasing decisions. The results further state that brand image has a direct influence on purchasing decisions. Brand image is also able to mediate the indirect relationship between digital marketing and purchasing decisions. Theoretically, this research confirms that consumers do not immediately make purchasing decisions with digital marketing in a company, without knowledge about the brand image of the product they want to buy.
23 July 2024 in TIN: Terapan Informatika Nusantara
Cancer, the leading cause of global death, results from abnormal cell proliferation that spreads beyond the boundaries of normal tissue. Breast cancer is one of the most common types of cancer, with approximately 2.26 million cases reported in 2020. This research aims to develop a more effective Support Vector Machine (SVM) algorithm for breast cancer classification through efficient feature selection techniques. Previous research has used various algorithms such as K-Nearest Neighbor and Logistic Regression for breast cancer identification. This research focuses on improving accuracy by using alternative feature selection methods such as Recursive Feature Elimination (RFE) and Forward Selection. The dataset used consists of 569 instances with 32 features sourced from the UCI Machine Learning Repository, and classified into benign and malignant categories. Data pre-processing methods, including data cleaning, coding, and feature selection, were applied to the dataset. RFE and Forward Selection techniques were used to identify the most important features for model training. Evaluation of the improved SVM model shows a training accuracy of nearly 100% and a Cross Validation accuracy of 97%, demonstrating the effectiveness of the proposed approach in the context of breast cancer. In addition, the Learning Curve and testing showed the stability of the SVM model with no signs of overfitting or underfitting. Thus, this study developed an SVM algorithm with a feature selection method that produces better accuracy results in breast cancer classification.
30 June 2024 in Jurnal Teknik Informasi dan Komputer (Tekinkom)
Malnutrition is a disease whose incidence in Indonesia is increasing. Based on data from West Java Province, in 2019-2022 there were 1,986,890 toddlers who experienced nutritional problems. This high figure shows that the problem of malnutrition has not received adequate attention. This research will group districts and cities in West Java into three clusters. Cluster 1 (high), cluster 2 (medium), and cluster 3 (low) are based on the number of toddlers affected by malnutrition. It is hoped that this research can help the government in making decisions to overcome nutritional problems in West Java. The results of research on clustering cases of malnutrition in West Java using the K-Means algorithm include 3 clusters, Cluster 1 (high) which consists of 16 city districts. Cluster 2 (medium) consists of 10 city districts, Cluster 3 (low), while in the K-Medoids algorithm Cluster 1 (high) consists of 10 city districts. Cluster 2 (medium) consists of 4 city districts. Cluster 3 (low) consists of 13 districts and cities. The comparison results show that the K-Means algorithm is better using 3 clusters based on evaluation using the Silhouette Coefficient with a value of 0.617