cover
Contact Name
Mesran
Contact Email
jurnal.josh@gmail.com
Phone
+6282161108110
Journal Mail Official
jurnal.josh@gmail.com
Editorial Address
Sekretariat Forum Kerjasama Pendidikan Tinggi (FKPT) Jalan Sisingamangaraja No. 338, Medan, Sumatera Utara
Location
Kota medan,
Sumatera utara
INDONESIA
JOURNAL OF INFORMATION SYSTEM RESEARCH (JOSH)
ISSN : -     EISSN : 2686228X     DOI : -
Core Subject : Science,
Artikel yang dimuat melalui proses Blind Review oleh Jurnal JOSH, dengan mempertimbangkan antara lain: terpenuhinya persyaratan baku publikasi jurnal, metodologi riset yang digunakan, dan signifikansi kontribusi hasil riset terhadap pengembangan keilmuan bidang teknologi dan informasi. Fokus Journal of Information System Research (JOSH)
Articles 795 Documents
Analisis Sentimen Pengguna Media Sosial Terhadap Identitas Kependudukan Digital Menggunakan Metode Support Vector Machine (SVM) Sulistiani, Vina Alipah; Hamka, Muhammad
Journal of Information System Research (JOSH) Vol 5 No 4 (2024): Juli 2024
Publisher : Forum Kerjasama Pendidikan Tinggi (FKPT)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/josh.v5i4.5614

Abstract

The Government of the Republic of Indonesia has implemented a policy of using Digital Population Identity (IKD) as one of the priority services of the Electronic Based Government System (SPBE) to accelerate digital transformation in the context of integrase national digital service. The implementation of IKD raises pros and cons perspectives in society. The counter perspective that is a problem in using IKD is that people are still worried about the security and privacy of personal data listed on digital systems. People can easily voice opinions, aspirations and participation through social media. However, given the large amount of social media data available, the process of obtaining the right information requires a lot of time and special skills. So a sentiment analysis of IKD on social media was carried out using the method Support Vector Machine (SVM) and improved with techniques Synthetic Minority Oversampling Technique (SMOTE) which aims to find out views or opinions regarding IKD which are positive, negative and neutral based on other people's points of view. The opinion data used in this research was 6697 with the keywords "digital population identity", "electronic ID card", and "digital ID card" taken from 2021 to 2024. Then, the data was classified based on polarity values ​​using a dictionary lexicon that is senticnet7 with sentiment results positive by 57.1% and sentiment negative amounting to 42.9%. The results of the classification process using the SVM method and enhanced with the SMOTE technique have results accuracy of 89%, negative precision of 88%, positive precision of 90%, recall negative by 85%, recall positive by 92%, f1-score negative by 86%, and f1-score positive by 91%.
Analisis Sentimen Pengguna Terhadap Layanan Aplikasi Seabank Indonesia di Instagram Menggunakan Metode Support Vector Machine Arrafiq, Muhammad Sunni; Kurniawan, Rakhmat
Journal of Information System Research (JOSH) Vol 5 No 4 (2024): Juli 2024
Publisher : Forum Kerjasama Pendidikan Tinggi (FKPT)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/josh.v5i4.5620

Abstract

Various aspects of life have been significantly changed by rapid technological advances; this includes the banking industry, which has developed digital banking services. SeaBank is a digital banking application that allows us to do many things with our money, from saving to making online transactions with our mobile phones anytime and anywhere. By using the Support Vector Machine (SVM) method to classify user comments on the SeaBank Indonesia Instagram account into positive and negative comments, this research aims to find analytical ways to improve service quality and customer satisfaction through this sentiment. Data is processed in several stages, such as cleaning, normalization, tokenization, stopword removal, and stemming. Then the SVM algorithm is used to classify sentiment. These performance algorithms are measured by metrics such as accuracy, precision, recall, and F1 score. The results of the analysis of 1201 comment data show that 536 data are positive and 665 data are negative. The Support Vector Machine method shows an accuracy of 89%, precision of 93%, recall of 83%, and fl-score of 88%.
Analisis Kesuksesan Aplikasi Presensi Mobile QR Code Universitas Muhammadiyah Purwokerto Menggunakan Model DeLone dan McLean Putri, Eno Sukarno; Harjono, Harjono
Journal of Information System Research (JOSH) Vol 5 No 4 (2024): Juli 2024
Publisher : Forum Kerjasama Pendidikan Tinggi (FKPT)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/josh.v5i4.5624

Abstract

Universitas Muhammadiyah Purwokerto (UMP) has adopted QR Code technology for its student attendance application. However, it is important to evaluate the success and performance of the application. This study aims to determine the level of success and factors that drive the success of the UMP Attendance application. The DeLone and McLean (2003) model uses 6 variables as a research evaluation framework. The research method used is quantitative with an associative approach and uses the SEM-PLS model assisted by SmartPLS 4.1.0.6 software in analyzing data. This research was conducted at Muhammadiyah Purwokerto University with 99 respondents who were active undergraduate students of UMP. The results showed that the implementation of the UMP Presence Application was considered quite successful and successful based on the DeLone and McLean (2003) SI success model. Judging from the aspects of system quality and information quality, it is strong enough to encourage use, user satisfaction and net benefits. However, it is necessary to improve the quality of service to users. So that later it will also affect the increase in use and provide benefits for users of the UMP Presence Application itself.
Implementation of Perspective, Vader, and TextBlob in Toxicity and Sentiment Analysis of Food and Tourism Singgalen, Yerik Afrianto
Journal of Information System Research (JOSH) Vol 5 No 4 (2024): Juli 2024
Publisher : Forum Kerjasama Pendidikan Tinggi (FKPT)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/josh.v5i4.5632

Abstract

This research investigates the sentiment and toxicity of viewer responses to digital content on food and tourism using the Digital Content Reviews and Analysis Framework. Employing advanced text processing and sentiment analysis models such as Perspective, TextBlob, and Vader, the study analyzed 4,166 comments. The findings reveal a predominantly positive viewer sentiment, with VADER identifying 18.98% negative, 21.33% neutral, and 59.69% positive sentiments, while TextBlob identified 14.42% negative, 33.86% neutral, and 51.72% positive sentiments. The toxicity analysis highlighted various levels, with an average toxicity score of 0.15783 and notable scores for severe toxicity, identity attack, insult, profanity, and threat. The research underscores the importance of comprehensive sentiment analysis in understanding viewer engagement, providing valuable insights for content creators and marketers in the tourism industry. The study concludes with recommendations for further exploration and refinement of sentiment analysis methodologies to enhance the understanding and management of digital content interactions.
Implementasi Metode MABAC Untuk Menentukan Siswa Yang Layak Menerima Beasiswa PIP SDN Pasaribu, Haryati; Putri, Raissa Amanda
Journal of Information System Research (JOSH) Vol 5 No 4 (2024): Juli 2024
Publisher : Forum Kerjasama Pendidikan Tinggi (FKPT)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/josh.v5i4.5633

Abstract

The MABAC (Multi-Attributive Border Approximation area Comparison) method is a multi-criteria decision making technique that evaluates alternatives based on distance from the border approximate area. This method allows a more objective assessment by considering various criteria and their importance. This research aims to implement the MABAC method in determining students who are worthy of receiving scholarships at SDN 15001 Kolang Nauli 2. Using a Research and Development (R&D) approach with a waterfall model, this research develops a web-based system to process data for 10 sample students based on five assessment criteria. The MABAC method is applied through a series of stages, including forming a decision matrix, normalization, weighting criteria, determining limit values, and calculating alternative distances. The research results show that the system succeeded in implementing the MABAC method accurately, producing objective rankings of scholarship recipient students. Gavin Manuel Hutabarat was selected as the strongest candidate with the highest score of 14,567. This implementation proves the effectiveness of the MABAC method in improving the decision-making process for scholarship allocation, offering a more transparent and measurable solution for schools.
Penerapan Algoritma C4.5 Dalam Sistem Prediksi Kelulusan Santri Tahfidz Qur’an Siregar, Putri Aprilia; Putri, Raissa Amanda
Journal of Information System Research (JOSH) Vol 5 No 4 (2024): Juli 2024
Publisher : Forum Kerjasama Pendidikan Tinggi (FKPT)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/josh.v5i4.5635

Abstract

Tahfidz Ulumul Qur'an Islamic Boarding School Medan is an Islamic religious educational institution that focuses on educating memorizers of the Al-Qur'an. However, not all students can graduate on time according to the study period taken, resulting in a buildup of students who graduate not on time. The determination of student graduation is based on several criteria that students must pass while studying at the Tahfizd Ulumul Qur'an Islamic Boarding School. Therefore, the C4.5 algorithm is a machine learning method used to build a decision tree based on 160 students' data. The data used in this research includes various factors that influence students' graduation, such as Makhorijul Letters, Memorizing, Discipline, and Relevant lafadz. The results of this research show that the C4.5 algorithm is able to provide accurate predictions regarding the graduation of Tahfidz Qur'an students with an accuracy rate of 98.21%. Where, based on the results of processing student data, it was found that 107 data of students graduated on time and 53 did not graduate on time.
Perbandingan Metode Decision Tree Dan K-Nearest Neighbor Terhadap Ulasan Pengguna Aplikasi Mypertamina Menggunakan Confusion Matrix Syahril, Ade; Cahyana, Yana; Kusumaningrum, Dwi Sulistya; Rohana, Tatang
Journal of Information System Research (JOSH) Vol 5 No 4 (2024): Juli 2024
Publisher : Forum Kerjasama Pendidikan Tinggi (FKPT)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/josh.v5i4.5639

Abstract

The large number of vehicles in Indonesia makes fuel oil (BBM) very important, especially for cars and motorbikes. The Indonesian government works closely with PT Pertamina Persero and requires transactions using the MyPertamina application to ensure that fuel subsidies are properly targeted. However, the MyPertamina app has received mixed feedback and criticism from users, such as complaints about frequent bugs, instability of the app during use and difficulties in the registration or login process. User feedback on the app has been both positive and negative. Users also provided their ratings and reviews on the Google Play Store. The purpose of this research is to analyse the opinions of MyPertamina application user comments and compare the accuracy of the Decision Tree and K-Nearest Neighbor algorithms. This research includes scraping, text preprocessing, weighting, algorithm implementation and evaluation. The data used was obtained from Google Play Store as much as 10,000 data based on the latest reviews, after data cleaning such as removing duplicate data and missing values obtained 8,072 reviews. The data is then grouped into positive classes (2,506 reviews) and negative classes (5,566 reviews), with more negative data. The classification results using the Decision Tree and K-NN methods, it is known that the Decision Tree method has a higher accuracy of 83%, while K-NN method is 58%. This finding indicates that the Decision Tree method is more effective in analysing user reviews of the MyPertamina application compared to the K-NN method.
Analisa Perbandingan Algoritma Support Vector Machine dan K-Nearest Neighbors Terhadap Ulasan Aplikasi Vidio Gumilar, Rizki Bintang; Cahyana, Yana; Sukmawati, Cici Emilia; Siregar, Amril Mutoi
Journal of Information System Research (JOSH) Vol 5 No 4 (2024): Juli 2024
Publisher : Forum Kerjasama Pendidikan Tinggi (FKPT)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/josh.v5i4.5640

Abstract

Internet usage in Indonesia reached 77% of the total population in January 2023, with Over The Top (OTT) services showing user growth of 25% every year. The Vidio application, one of the popular OTT platforms with downloads exceeding 50 million, has a 3.5 star rating based on 649 thousand reviews on the Google Play Store. Despite its popularity, Vidio faces complaints regarding limited film selection, payment errors, and excessive advertising, which affects user satisfaction. This research aims to analyze the opinions of Vidio application user comments by applying the SVM (Support Vector Machine) method and the KNN (K-Nearest Neighbors) method to determine the model with the best accuracy. 15,000 review data were collected through scraping, then processed using text preprocessing and TF-IDF vectorization techniques. Model evaluation shows that SVM has an accuracy value of 82%, a precision value of 82%, a recall value of 83%, and an F1-score value of 82%, while KNN has an accuracy of 69%, precision 74%, recall 73%, and F1-score 69% . The research results show that SVM is superior to KNN in classifying the sentiment of Vidio application reviews. It is hoped that these findings can be used by application developers in an effort to improve service and satisfaction of Vidio application users.
Analisis Sentimen Terhadap Kesehatan Mental Remaja Menggunakan Metode Naive Bayes Syahputra, Pii; Kurniawan, Rakhmat
Journal of Information System Research (JOSH) Vol 5 No 4 (2024): Juli 2024
Publisher : Forum Kerjasama Pendidikan Tinggi (FKPT)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/josh.v5i4.5644

Abstract

Everyday life, especially for teenagers, now involves internet technology. Twitter (now known as X) is one of the most popular social media platforms. Sentiment analysis of social media data can improve people's understanding of mental health problems. This research uses the Naive Bayes algorithm to analyze the sentiments of X social media users regarding adolescent mental health. Another goal of this research is to measure how effective and accurate the technique is in identifying sentiment and presenting analysis results in the form of word clouds and graphs. Data was collected from the beginning of 2024 to the present from tweets with the hashtag Mental Health. The research results show that the Naive Bayes algorithm has an effective level of accuracy in classifying sentiment towards health using the InSetLexicon dictionary. The data preprocessing process also includes cleaning, tokenizing, normalization, stockwords, and stemming. In addition, performance evaluation is carried out using confusion matirx to calculate precision, recall, F-1 Score, and accuracy. The classification results obtained obtained an accuracy of 0.8049792531120332 or around 80%, precision of 83%, recall of 68% and F-1 Score of 74.9%.
Implemetasi Data Mining Menggunakan Algoritma Apriori Dalam Menentukan Pola Penjualan Carton Box Hardi, Nila; Putra, Jordy Lasmana; M, Tika Adilah
Journal of Information System Research (JOSH) Vol 5 No 4 (2024): Juli 2024
Publisher : Forum Kerjasama Pendidikan Tinggi (FKPT)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/josh.v5i4.5646

Abstract

Good company managers must be able to examine the sales patterns that exist in the company. Some companies have shortcomings, including the problem of stock of goods that do not match the number of goods sold. This certainly affects the level of sales. The existence of sales activities every day, sales transaction data will continue to grow, causing greater data storage. Sales transaction data is only used as an archive without being put to good use. Basically the data set has very useful information. In data mining there are several algorithms or methods that can be done, one of which is the a priori algorithm which is included in the association rules in data mining. A priori algorithm which aims to find frequent item sets in a set of data. A priori algorithm is defined a process to find a priori rules that meet the minimum requirements for support and the minimum requirements for confidence. Test results with a priori algorithm and the system built shows the results which has fulfilled the need to determine sales patterns based on the number of transactions of goods sold. This shows the effectiveness of information from the system about determining the pattern of selling carton boxes to manage stock properly in accordance with the goods with the highest number of transactions seen from 2 caton box sets.