cover
Contact Name
Mesran
Contact Email
mesran.skom.mkom@gmail.com
Phone
-
Journal Mail Official
jurnal.bits@gmail.com
Editorial Address
-
Location
Kota medan,
Sumatera utara
INDONESIA
Building of Informatics, Technology and Science
ISSN : 26848910     EISSN : 26853310     DOI : -
Core Subject : Science,
Building of Informatics, Technology and Science (BITS) is an open access media in publishing scientific articles that contain the results of research in information technology and computers. Paper that enters this journal will be checked for plagiarism and peer-rewiew first to maintain its quality. This journal is managed by Forum Kerjasama Pendidikan Tinggi (FKPT) published 2 times a year in Juni and Desember. The existence of this journal is expected to develop research and make a real contribution in improving research resources in the field of information technology and computers.
Arjuna Subject : -
Articles 969 Documents
Optimization of ID3 Structure for Academic Performance Analysis using Ant Colony Algorithm Fathudin, Dedin; Ambarsari, Erlin Windia; Paramita, Aulia
Building of Informatics, Technology and Science (BITS) Vol 6 No 1 (2024): June 2024
Publisher : Forum Kerjasama Pendidikan Tinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bits.v6i1.5353

Abstract

This study investigates the optimization of the ID3 algorithm for academic performance analysis using the Ant Colony Optimization (ACO) method. The primary research problem addressed is the inefficiency and overfitting of traditional ID3 in complex and noisy datasets. Therefore, the ACO method is integrated to enhance the ID3 structure, improving classification accuracy and computational efficiency. The research objectives include developing a decision tree model based on assignment, mid-term, and final exam scores for student performance evaluation. The method combines ID3's decision-making capabilities with ACO's optimization process, which uses pheromone trails to find optimal paths in constructing the decision tree. Temporary results show that the ACO-ID3 model achieves an accuracy of 85% with improved consistency and lower variability compared to the traditional ID3 model, which has an accuracy of 89% but higher variability; this indicates that while traditional ID3 may slightly outperform in accuracy, the ACO-ID3 model provides more stable and reliable performance across different data subsets. The study concludes that ACO-ID3 is a practical and effective tool for academic performance analysis, particularly in cases requiring consistent and reliable classification
Deteksi Penipuan Kartu Kredit Menggunakan Support Vector Machine dengan Optimasi Grid Search dan Genetic Algorithm Hasibuan, Lailan Sahrina; Jannah, Fatimah Alfiatul
Building of Informatics, Technology and Science (BITS) Vol 6 No 1 (2024): June 2024
Publisher : Forum Kerjasama Pendidikan Tinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bits.v6i1.5355

Abstract

Credit card transactions have increased significantly every year. Along with the increasing use of credit cards, the risk of fraud by irresponsible people also increases. Credit card fraud can be detected with the help of machine learning. The main problem that often encountered is the transaction data has very large dimensions, unbalanced classes, and requires a detection process with a short computation time. Therefore we need a model that can produce good performance with short computation time using the support vector machine (SVM) method with grid search and genetic algorithm optimization. From the three models built, it was found that the SVM model using an initial dataset which was balanced using ADASYN and searching for the best parameters using grid search as a hyperparameter optimization technique was able to carry out good detection and short computing time. This model is able to detect fraudulent transactions with 99% sensitivity and 99% specificity and the shortest model training time among the other two models.
Implementasi Sistem Rekomendasi dengan Content Based Filtering dan Teknologi Virtual Tour Untuk Strategi Pemasaran Pada Website Insany, Gina Purnama; Somantri, Somantri; Amalia, Phina Putri
Building of Informatics, Technology and Science (BITS) Vol 6 No 1 (2024): June 2024
Publisher : Forum Kerjasama Pendidikan Tinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bits.v6i1.5358

Abstract

The real estate business is rapidly developing, supported by a stable economy and increased purchasing power of the public, which triggers intense competition in the property industry. Consequently, many companies are striving to offer modern solutions. Prospective buyers often struggle to choose a house that suits their needs due to the plethora of options, while real estate companies face challenges in providing attractive and comprehensive information. Solutions like a recommendation system with Content-Based Filtering and Virtual Tour technology offer innovative approaches that make it easier for consumers to select a house and enhance marketing strategies on the Setiabudi Land website. This recommendation system provides suggestions according to user characteristics and preferences with inputs such as age, marital status, number of children, monthly income, desired facilities, and environmental preferences. The output includes recommendations on the name of the housing complex, type, and model of the house. On the other hand, Virtual Tour offers a realistic visual experience, allowing consumers to view properties virtually without having to visit the physical location, showcasing the living room, family room, 2 bedrooms, 1 bathroom, and 1 kitchen. Evaluation results of the recommendation system's performance show accuracy, precision, recall, and F1-score levels in the range of 92-93%, while functionality tests of the Virtual Tour run smoothly. User Acceptance testing reached 88.66%, indicating a high level of user satisfaction with the recommendation system and virtual tour features.
Implementasi Algoritma Branch and Bound Pada Aplikasi Mobile Pemandu Wisata Untuk Pengembangan UMKM Jawa Barat Sepriyadi, Adi; Sujjada, Alun; Somantri, Somantri
Building of Informatics, Technology and Science (BITS) Vol 6 No 1 (2024): June 2024
Publisher : Forum Kerjasama Pendidikan Tinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bits.v6i1.5359

Abstract

Tourism is a crucial economic sector in Indonesia, contributing significantly to economic growth and foreign exchange. In 2022, the number of domestic tourist visits reached 734.86 million, up 19.82% from the previous year. East Java recorded the most visits with 198.91 million trips, followed by West Java (128.66 million) and Central Java (103.99 million). However, tourist destinations in West Java are less well known due to a lack of information, hampering tourism potential and the development of local MSMEs. To overcome this problem, an Android-based guide application was developed that integrates information on tourist destinations and MSMEs in West Java. This application uses the Branch and Bound algorithm to facilitate the selection of the closest route from tourist sites to several MSMEs efficiently and accurately. This research identifies the lack of information as the main obstacle to tourism development in West Java. Application development and testing showed the effectiveness of the Branch and Bound algorithm in determining the closest route. The implementation of this application increases access to tourist information and MSMEs, as well as the income of local MSME players related to tourism. The application is expected to contribute positively to the development of tourism and MSMEs in West Java and become a model for the development of similar applications in other regions in Indonesia.
Sistem Pendukung Keputusan Pemilihan Pegawai Terbaik Menggunakan Kombinasi Metode Pembobotan MEREC dan Simple Additive Weighting Chandra, Iryanto; Hadad, Sitna Hajar
Building of Informatics, Technology and Science (BITS) Vol 6 No 1 (2024): June 2024
Publisher : Forum Kerjasama Pendidikan Tinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bits.v6i1.5360

Abstract

The selection of the best employees is an event that aims to appreciate and recognize the extraordinary performance of employees in a company. The implementation of the selection of the best employees is not without challenges and problems, one of the main problems is the risk of subjectivity in the assessment process, which can cause dissatisfaction among other employees if they feel that the assessment is unfair or transparent. The purpose of this study is to develop an SPK that can help in the selection of the best employees by using a combination of MEREC and SAW methods, a MEREC approach to manage and evaluate important criteria in employee selection, and integrate SAW as a mathematical method to provide weight and ranking candidates based on predetermined criteria. The recommendation for the results of the selection of the first best employee with a final SAW score of 0.8345 was obtained by Candidate 8, the second best employee with a final SAW score of 0.8253 was obtained by Candidate 6, and the third best employee with a final SAW score of 0.8068 was obtained by Candidate 3.
Penerapan Metode TOPSIS dalam Pemilihan Moda Transportasi Berkelanjutan untuk Pengurangan Emisi Gas Rumah Kaca Pramita, Galuh; Darwis, Dedi
Building of Informatics, Technology and Science (BITS) Vol 6 No 1 (2024): June 2024
Publisher : Forum Kerjasama Pendidikan Tinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bits.v6i1.5361

Abstract

The choice of sustainable modes of transportation for the reduction of greenhouse gas emissions is an important aspect in efforts to mitigate climate change. By choosing environmentally friendly modes of transportation, such as public transportation, electric vehicles, bicycles, or ride-sharing, individuals and organizations can contribute significantly to the reduction of greenhouse gas emissions produced by the transportation sector. The research objective of the application of decision support systems in the selection of sustainable modes of transportation for the reduction of greenhouse gas emissions is to provide a structured and effective framework in assisting individuals and organizations in choosing alternative modes of transportation that best suit their needs while also minimizing negative impacts on the environment. DSS can be a very useful tool in the selection of sustainable modes of transportation for the reduction of greenhouse gas emissions. DSS can integrate various factors such as energy efficiency, greenhouse gas emissions, infrastructure accessibility, and cost to determine the mode of transportation that best suits user needs and preferences. The ranking results are based on the respondents' assessment data for rank 1 with a final score of 0.92974 with an alternative name for Bicycles, rank 2 with a final score of 0.78159 with an alternative name for Hydrogen-Based Public Transportation, rank 3 with a final score of 0.76089 with an alternative name for Public Transportation, and rank 4 with a final score of 0.15703 with an alternative name for Electric Vehicles. The results of processing respondent response data based on 4 TRITAM Model criteria obtained Trust results of 76.25%, Risk of 75%, Perceived usefullness of 93.96%, Perceived easy of Use of 82.92%. Of the overall criteria of the TRITAM Model for technology acceptance, the result was very good at 84.17%
Implementation of the Naïve Bayes Algorithm to Predict New Student Admissions Salsabila, Aulia; Nasution, Marnis; Irmayanti, Irmayanti
Building of Informatics, Technology and Science (BITS) Vol 6 No 1 (2024): June 2024
Publisher : Forum Kerjasama Pendidikan Tinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bits.v6i1.5363

Abstract

New student admissions are critical to the success of an educational institution because they determine the existence and financial sustainability of that institution. The number of prospective students who register changes every year. The school cannot anticipate the number of students who will come. Additionally, data on prospective students who enroll is collected annually without being analyzed to extract valuable information. The school must make predictions to estimate the number of new students in the next school year. Predictions are essential for effective planning, both in the long and short term. This research aims to apply the Naïve Bayes algorithm with Gaussian type to predict new student admissions. To find out whether the Naïve Bayes algorithm works well, an evaluation matrix is used. The methods applied include the dataset collection process, data preprocessing, split data training and testing, feature engineering, the implementation of Naïve Bayes, and results evaluation. The dataset is divided into 70% training data and 30% testing data. The research results show an accuracy score of 86.11% during training and an accuracy score of 90.62% during model testing, with an increase of 4.51%. These results show that there is no indication of overfitting in the machine learning algorithm used. The evaluation matrix produces an accuracy score of 90.62%, precision of 100%, recall of 90.62%, and f1-score of 95.08%. From the results of the evaluation matrix scores, it can be concluded that the naive Bayes algorithm with Gaussian type succeeded in predicting new student admissions well.
Pengaplikasian Data Mining Dalam Mengelompokan Data Penerima Bantuan Subsidi Rumah dengan Menggunakan Metode K-Means Clustering Aranski, Alvendo Wahyu; Astiti, Sarah; Putra, Riko Andrian; Darmansah, Darmansah
Building of Informatics, Technology and Science (BITS) Vol 6 No 1 (2024): June 2024
Publisher : Forum Kerjasama Pendidikan Tinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bits.v6i1.5366

Abstract

From 2015 until now, the government has provided assistance for house renovations in the Tembesi area, Sagulung sub-district, Batam City. However, in determining the provision of assistance, sub-district governments sometimes face problems in determining which people will receive housing subsidies and there is no scheme or category for determining recipients of assistance. Therefore, the author will conduct this research by grouping or clustering the eligibility of recipients of housing subsidy assistance using the K-Means algorithm. The K-Means clustering algorithm can group each data into sets, so that data sets with the same characteristics will be grouped in the same set, or data sets with different characteristics will be grouped in different sets. The purpose of grouping is to determine that group 0 and group 1 are eligible to receive housing subsidy assistance, and group 1 is not. This research uses metrics such as number of family members, employment, housing conditions, and income. The results of this research obtained 91 data in cluster 0 and 79 data in cluster 1. Thus, from the 170 data, 91 people were eligible for housing subsidy assistance, and 79 people were not eligible
Implementasi Algoritma Random Forest Untuk Analisa Sentimen Data Ulasan Aplikasi Pinjaman Online Digoogle Play Store Wibisono, Yudistira Arya; Afdal, M.; Mustakim, Mustakim; Novita, Rice
Building of Informatics, Technology and Science (BITS) Vol 6 No 2 (2024): September 2024
Publisher : Forum Kerjasama Pendidikan Tinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bits.v6i2.5368

Abstract

Online lending programs are examples of financial service platforms offered directly by commercial fintech players. However, there are rampant cases of fraud and unethical actions by some online lenders such as threatening and harassing billing methods due to late payments. This research aims to classify sentiment from user reviews of online loan applications on the Google Play Store into positive, negative, or neutral categories. This research conducts sentiment analysis of user reviews of online loan applications such as AdaKami, AdaModal, Cairin, FinPlus and UangMe using a text mining approach. This approach can perform sentiment classification on user reviews quickly. Data was collected using the scrapping technique on the Google Play Store and obtained a total of 200 data on each online loan application. The modeling used in this research is the division of training data and test data as much as 80:20. The highest accuracy results using the Random Forest algorithm are Cairin and UangMe applications with 85% accuracy. While the application that gets the lowest accuracy result is the AdaModal application with 75% accuracy. A visualization analysis using word clouds was also conducted to understand the context of user reviews of the pinjol apps. The results show that users almost always discuss loan limits in every sentiment across the five apps.
Implementasi Sistem Pakar Untuk Diagnosis Penyakit Tomat: Pendekatan Backward Chaining Berbasis Web Wijaya, Taruma Leo; Fryonanda, Harfebi; Mardiah, Ainil; Defni, Defni; Ibrahim, Roy
Building of Informatics, Technology and Science (BITS) Vol 6 No 1 (2024): June 2024
Publisher : Forum Kerjasama Pendidikan Tinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bits.v6i1.5371

Abstract

This study aims to implement an expert system using a web-based backward chaining method approach to help farmers diagnose and manage tomato plant diseases. This study was conducted because diagnosing plant diseases requires the help of agricultural experts, which causes problems with consultation costs and farmers who have difficulty knowing the type of disease in tomato plants will cause losses for farmers due to crop failure. The methodology used in this study is the prototype method which begins with identifying problems through data collection from literature, interviews with agricultural experts, and field observations. The collected data is then analyzed using the backward chaining method to trace symptoms to the cause of the problem and provide recommendations for handling. This system is implemented in the form of a web application that facilitates access for farmers. The results of the study show that this expert system is able to provide accurate and reliable diagnoses and recommendations with an accuracy level of this study of 85%. Thus, this expert system is expected to improve farmers' knowledge and skills in managing tomato plants, as well as contribute to increasing yields and the quality of agricultural products