Paradigma - Jurnal Komputer dan Informatika

Predicting Graduation Outcomes: Decision Tree Model Enhanced with Genetic Algorithm

2024-03-20T14:29:46+07:00

This research aims to improve the accuracy of predicting student permit results in the digital era by utilizing machine learning techniques. The main focus is the use of a Decision Tree (DT) model optimized with a Genetic Algorithm (GA) to overcome the limitations of accuracy and testing of conventional methods. This research began with collecting student academic data, followed by preprocessing to eliminate incompleteness and organize the data format. The DT model is then built and optimized with GA, which is inspired by biological evolutionary processes to improve feature selection and parameter tuning. The results show a significant increase in prediction accuracy, from 86.19% to 87.68%, and an increase in the Area Under Curve (AUC) value from 0.755% to 0.788%. This research not only proves the effectiveness of GA integration in improving DT models, but also paves the way for the application of evolutionary techniques in educational data analysis and other fields. The main contributions of this research include the development of more accurate prediction models and practical applications in educational contexts, with the hope of assisting educational institutions in making more informed decisions for their students.

Analysis of IT Service Desk Applications Using the Servqual Method at The Republic of Indonesia National Library

2024-03-20T14:50:25+07:00

The National Library has implemented an Electronic-Based Government System (SPBE) so that library service operations and operations for the national library staff have used a computerized system which results in the PERPUSNAS requiring centralized management of IT problems, namely the IT Service Desk. As an IT problem service, IT Service Desk needs to be analyzed to measure the level of user satisfaction to determine the quality of its services. In this study, the measurement of the level of user satisfaction of IT service desk services was carried out using the Servqual method which has 5 dimensions of service quality, namely Tangible, Reliability, Responsiveness, Assurance, and Emphaty by looking for the value of the gap between the value of perception and the value of expectations from users. The results of the IT Service Desk measurement at PERPUSNAS RI obtained a gap of -0.78 with the category "Good Enough".

Web-Based Laundry Service Information System Using Rapid Application Development Method

2024-03-05T18:18:11+07:00

In the current era of digitalization, human needs for time make all household and community activities want to be practical, especially such as washing clothes. Especially for people who have busy demands and do not have time to wash clothes, dry and iron so they prefer to entrust their laundry to the services of a laundry or laundry, for the reason of saving time and to concentrate more on completing their work. This study aims to design a service information system that can help in conveying information about laundry and knowing the valid and efficient website-based laundry service information system. This form of research uses the Rapid Application Development method to produce a product and uses a prototype model to design the system. This Information System is the best solution for solving problems in the management of laundry service systems. With the use of computer data technology that is managed becomes faster, reducing time inefficiencies and reducing the occurrence of errors in processing data.

Comparison of Supervised Learning Classification Methods on Accreditation Data of Private Higher Education Institutions

2024-03-13T14:40:13+07:00

This research aims to analyze and compare supervised learning classification methods using a case study of accreditation data for private higher education institutions within the LLDikti Region III contained in BAN-PT. In addition, this research also uses Weka machine learning software in its calculations. The initial step taken is to prepare the software used for supervised learning analysis, then pre-processing the data, namely labeling data that has a categorical data type, after that determining data for testing data. The next step is to test each classification method. The methods used for comparison are logistic regression, K-nearest neighbor, naive bayes, super vector machine, and random forest. Based on the calculation results, the Kappa Statistic and Root mean squared error values obtained are 1 and 0 for the logistic regression method, 0.979 and 0.0061 for the K-nearest neighbor method, 1 and 0.2222 for the super vector machine method, 0.969 and 0.0341 for the naive bayes method, 1 and 0 for the decision tree method, and 0.5776 and 0.1949 for the random forest method, respectively. The logistic regression and decision tree methods in this study get Kappa Statistic and Root mean squared error values of 1 and 0 respectively so that they are said to be good and acceptable, thus the two classification methods are the most appropriate methods and are considered to have the highest accuracy.

Optimizing Heart Failure Detection: A Comparison between Naive Bayes and Particle Swarm Optimization

2024-03-05T18:25:03+07:00

This research focuses on the importance of early detection of heart failure which is a serious global health problem. Given the variety of symptoms of heart failure, accurate early detection methods are needed with the aim of reducing the impact of this disease. This study uses the Naïve Bayes (NB) method which has been proven effective in classifying heart failure with significant variations in accuracy by integrating Particle Swarm Optimization (PSO) to improve the model. The evaluation model involves a confusion matrix including accuracy, precision, recall, and Area Under the Curve. The research results show that the integration of PSO in NB results in an increase in accuracy of 7.73%, an increase in precision of 6.42%, and an increase in recall of 1.93%. Although there was a small decrease in AUC. This research shows that the success of NB with PSO can help improve the performance of early detection of heart failure. This indicates the importance of this research in developing more accurate and effective detection methods for critical health conditions such as heart failure.

Selection of Outstanding Students Using AHP and Profile Matching

2024-02-13T15:01:49+07:00

The determination of outstanding students is the giving of awards to those who excel in academic and non-academic fields, aimed at motivating increased achievement. However, this process is often hampered by various criteria that must be considered, such as English language skills, work results, awards, and so on. The solution offered to overcome this problem is the development of a decision support system for selecting outstanding students using the AHP and Profile Matching methods. So, the aim of this research is to develop a decision support system for selecting outstanding students using a combination of the AHP and Profile matching methods, where later the system developed can assist decision makers in determining outstanding students. The results obtained from this research are a decision support system that uses 8 criteria and 26 alternative sample data which shows that "Mahasiswa F" is an outstanding student with a score of 4.09. The results of manual calculations with the system show similarities, which shows that the system developed is in accordance with expectations.

User Centered Design in Analysis and Design of UI UX in the Simpeg Application of Dharmais Cancer Hospital

2024-03-13T14:49:07+07:00

Dharmais Cancer Hospital as a health service facility that is experienced in its field, of course, has human resources which are important, so to support employee performance in terms of employee rights such as employee leave, media is needed that really supports employee needs. Through the employee support system, namely the Employee Information System (SIMPEG) application, several problems were found in terms of appearance and the proposal process as well as the leave recapitulation process which did not exist in the previous SIMPEG application. From the problems found in the Personnel Information System (SIMPEG), a User Centered Design (UCD) method is offered which places the Personnel Information System (SIMPEG) users as the main consideration for building a new system. In the design process there are three steps that are carried out starting from the initial stage, the development stage and the final stage, where at the initial stage testing is carried out using a questionnaire, the average result is 53 which means OK but still low or not good, after the development of the system is carried out and the distribution of questionnaires was carried out again, the results were 71, which means Good. With the design of a Personnel Information System (SIMPEG) using the User Centered Design (UCD) method, it produces a new User Interface (UI) and User Experience (UX) design for the Personnel Information System (SIMPEG), which consists of a forgot password menu feature, leave recap and display easier to use for employees.

Phyton-Based Machine Learning Algorithm To Predict Obesity Risk Factors In Adult Populations

2024-03-18T15:24:52+07:00

Obesity is a serious health problem because it can lead to a variety of diseases. Adults are prone to obesity due to several factors such as age, physical activity, weight, diet, gender, lifestyle and so on. Machine Learning as one of the methods for predicting and classifying factors of obesity especially in the adult population. In machine learning, there are various types of algorithms that can be used to classify data. In this study, using the K-Nearest Neighbor, Decision Tree and Naïve Bayes algorithms, 2111 datasets were used and processed using the Phyton programming language. The results were obtained from the comparison of the three algoritms with the highest accuracy of 93.6%, the Decision Trees with 79.6% and the Naïv Bayes with 60%.

Building a Predictive Model for Chronic Kidney Disease: Integrating KNN and PSO

2024-05-02T08:40:54+07:00

This study examines the improvement of prediction accuracy for Chronic Kidney Disease (CKD) through the integration of the K-Nearest Neighbors (KNN) method with Particle Swarm Optimization (PSO). Amidst the rising prevalence of CKD, closely related to diabetes and hypertension, early detection of CKD becomes a significant challenge, especially in Indonesia where access to healthcare facilities and public awareness remain limited. This study utilizes the Chronic Kidney Disease dataset from the UCI Machine Learning repository, encompassing 400 patient records with 24 clinical, laboratory, and demographic variables. With the KNN method, this approach classifies data based on feature proximity, while PSO is used for feature selection and parameter optimization, enhancing the model's accuracy and efficiency in identifying CKD at early stages. The findings indicate a significant improvement in prediction accuracy, from 80.00% using KNN to 97.75% after integration with PSO. These results affirm that the combined approach of KNN and PSO holds great potential in improving early detection and management of CKD, paving the way for further research into practical applications in the healthcare field.