LILA SETIYANI
@rosma.ac.id
Information System
STMIK ROSMA
EDUCATION
Bachelor Information Technology STT Pelita Bangsa, Magister Information System STMIK LIKMI,
RESEARCH INTERESTS
Information System
Scopus Publications
Scholar Citations
Scholar h-index
Scholar i10-index
Scopus Publications
Lila Setiyani, Yeny Rostiani, and Rahmat Gunawan
AIP Publishing
Lila Setiyani, Hanny Hikmayanti Handayani, and Wildan Adhitya Geraldine
IEEE
Modern application development needs to consider security. In addition, with the increasing threat of cybercrime, software developers must be able to improve the quality of the applications they build. Machine learning has been widely implemented as a model that can increase application sophistication in detecting threats. SQL Injection ranks in the top 10 vulnerabilities in the OWASP framework. The purpose of this research is to find the best machine learning and deep learning models in detecting SQL Injection attacks. In this research, 5 algorithms will be tested, namely CNN, Logistic Regression, Random Forest, Naive Bayes, and Decision Tree. The procedure of this study adopts the AI development life cycle process with stages including project planning, data collection, data preparation, model development and model deployment. In this study, the datasets collected were unbalanced, so the process of data acquisition and preparation became important. To balance the data, the SMOTE technique is used. The results of this study indicate that CNN is the best algorithm model in detecting SQL Injection with an accuracy value of 0.95, compared to Logistic Regression 0.93, Random Forest 0.91, Naive Bayes 0.81 and Decision Tree 0.91.
Lila Setiyani and Cutifa Safitri
IEEE
The development of modern applications to support company business processes requires management with high service availability. Most application development still implements traditional client server architecture, without considering the need for high availability. Karawang Regional Hospital currently still uses this traditional architecture, so data availability has the potential to experience problems. Data Center – Disaster Recovery Center (DC-DRC) is an opportunity to meet high availability needs, so the right method is needed to build a DC-DRC that has optimal performance. This research aims to increase data availability at Karawang Regional Hospital through the implementation of DC-DRC which has high availability capabilities. The load balancer method was chosen to divide the distribution of server performance loads. This research method adopts a network development process model, namely Network Development Life Cycle (NDLC) with stages including analysis, design, simulation, and implementation. Based on the test results, the applied load balancer can share the load on DC-DRC and support high availability at Karawang Regional Hospital. The results of this research can be applied to the development of DC-DRC which can be used to increase the availability of service quality.
Lila Setiyani, Nansy Herina, and Yeny Rostiani
IEEE
Udemy Course as an online course platform that has been recognized by both students and teachers as a medium that facilitates the transfer and receiving of knowledge, while also providing income to the Instructors. This study aims to explore datasets in Udemy Course and predict the income of the Instructors in the Udemy course. Datasets are taken from Kaggle. The research procedure begins with the collection of datasets from Kaggle, followed by datasets pre-processing, then exploratory data analysis is carried out and predicted, the implementation of machine learning random forest techniques is carried out. The results of this study show a description of the Udemy Course datasets for 2010 - 2022, course distribution prices, customer distribution, course content length distribution, number of courses for each category, number of subscribers for each category, and date of publication. In addition, this study shows that the percentage score n_estimator of machine learning random forest techniques in predicting the income of Udemy course instructors is 94%.
Lila Setiyani, Ayu Indahsari, Rosalina, and Tjong Wansen
IEEE
Knowing a lot about attributes, such as bank client data, marketing campaigns, social and economic context, and so on, is necessary to predict which bank customers are likely to sign up for time deposits. The direct marketing campaigns dataset for the bank institution was made available by the UCI machine learning repository. This dataset had previously been used in studies to forecast customers who would sign up for time deposits, but the outcomes of the two studies were different. The purpose of this study is to determine which machine learning methods are most effective for predicting customers who will sign up for time deposits. In this study, researchers used machine learning methods like random forests, K-nearest neighbors, decision trees, naive bayes, logistic regression, SVM, and XGBoost to uncover the truth. Results showed that XGBoost had the highest accuracy (91.73%), followed by the DT classifier (88.79%), LR classifier (91.2%), NB classifier (84.78%), KNN classifier (90.6%), SVM (89.6%), and XGBoost (91.44%).