Claim Missing Document
Check
Articles

Perbandingan Algoritma Machine Learning dalam Analisis Penyebab Penyakit Gagal Jantung Kirono, Aryo Sasi; Nataliani, Yessica
JEPIN (Jurnal Edukasi dan Penelitian Informatika) Vol 10, No 2 (2024): Volume 10 No 2
Publisher : Program Studi Informatika

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26418/jp.v10i2.78369

Abstract

Penyakit gagal jantung meningkat seiring perkembangan jaman dikarenakan maraknya pola hidup yang tidak sehat, tingkat obesitas, dan angka perokok. Gagal jantung adalah kondisi medis yang abnormal pada struktur atau fungsi jantung. Gejala yang biasa dialami oleh penderita meliputi sesak nafas, kelelahan, dan penurunan tingkat aktivitas. Kemajuan teknologi yang sangat pesat dapat membantu dalam menganalisis penyebab penyakit gagal jantung, salah satunya teknologi machine learning yang mampu dalam memprediksi dan mengklasifikasi pasien yang beresiko gagal jantung dan normal. Penelitian ini menggunakan tiga model machine learning dalam analisis penyakit gagal jantung yaitu Decision Tree, Random Forest, dan XGBoost Ketiga model ini sama "“ sama memiliki fitur yang dapat diterapkan untuk mengetahui penyebab penyakit gagal jantung yaitu feature importance atau tingkat kepentingan karena termasuk dalam tree based model. Data yang digunakan dalam penelitian berjumlah 918 pasien gagal jantung. Dengan menerapkan fitur tersebut, ketiga model menghasilkan ST Slope yaitu kemiringan naik dan turun ST saat berolahraga menjadi variabel tertinggi terhadap resiko penyakit gagal jantung dengan 39% pada model Decision Tree, 22% pada model Random Forest, dan 46% pada XGBoost.
Analisis Penerimaan Masyarakat Kota Salatiga terhadap Fintech ShopeePay dengan Technology Acceptance Model Alwin Adi Putra; Yessica Nataliani
Indonesian Journal of Intelligence Data Science Vol 2 No 1 (2023): Volume 2 No. 1 2023
Publisher : Faculty of Mathematics and Natural Sciences Sam Ratulangi University

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Online shopping has become increasingly popular for many people in Indonesia since the Covid-19 pandemic took hold. Shopee is one of the most popular online buying and selling applications. Based on a survey conducted by App Annie, Shopee has been named the most popular marketplace in Indonesia regarding the number of active users and the volume of monthly visits. Shopee has a fintech called Shopeepay. Technology Acceptance Model (TAM) was used to analyze user acceptance during the Covid-19 pandemic. 100 ShopeePay users were taken as a sample using a purposive sampling technique, collected by questionnaire. Determination of the number of samples refers to the number of samples used in previous studies relevant to this research. The analytical techniques used in this study include regression analysis (simple and multiple) and path analysis. The analysis results from the Technology Acceptance Model show that the factors influencing the acceptance of ShopeePay fintech during the Covid-19 pandemic in Salatiga were Perceived Usefulness and Perceived Ease of Use. These two perceptions of TAM influence Attitude toward Using and Behavioral Intention to Use, which impacts Shopeepay's Actual System Usage during the Covid-19 pandemic.
ANALISIS SENTIMEN MASYARAKAT TERHADAP PENGGUNAAN VAKSIN COVID-19 DI INDONESIA MENGGUNAKAN METODE NAÏVE BAYES Leony Martiyana Putri; Yessica Nataliani
Indonesian Journal of Intelligence Data Science Vol 2 No 1 (2023): Volume 2 No. 1 2023
Publisher : Faculty of Mathematics and Natural Sciences Sam Ratulangi University

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

COVID-19 was officially declared a global pandemic by WHO on March 11, 2020. In Indonesia, cases of COVID-19 were first detected on March 2, 2020 and there were more and more positive confirmations every day. The government's strategy in fighting this pandemic is by carrying out vaccinations. The use of vaccination has received various responses from the public, both those who support it and those who oppose it. This study aims to analyze public opinion on vaccination, thereby helping the public to see whether vaccines are well received or not. The data used are 600 tweets for three keywords, namely "astra", "sinopharm", and "sinovac", with 200 tweets for each keyword. Each data is divided into 70% training data and 30% testing data. Classification is done using the Naïve Bayes method. Sentiment results with the keyword "astra" show 159 tweets giving neutral sentiment, 19 tweets giving positive sentiment, and 22 tweets giving negative sentiment, with an accuracy of 68,66%. The keyword "sinovac" shows 134 tweets giving neutral sentiment, 13 tweets giving positive sentiment, and 23 tweets giving negative sentiment, with an accuracy of 82,86%. The keyword "sinopharm" shows 77 tweets giving neutral sentiment, 22 tweets giving positive sentiment, and six tweets giving negative sentiment, with an accuracy of 28,77%. It can be concluded that the results of public sentiment towards vaccination received good support from the community.
Multi-Objective k-Nearest Neighbor for Breast Cancer Detection Nataliani, Yessica; Arthur, Christian; Wellem, Theophilus; Hartomo, Kristoko Dwi; Wahab, Nur Haliza Abdul
JOIV : International Journal on Informatics Visualization Vol 9, No 1 (2025)
Publisher : Society of Visual Informatics

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.62527/joiv.9.1.2669

Abstract

Early detection of cancer is crucial. This study aims to increase the efficiency of breast cancer detection using the modified k-nearest neighbor (k-NN) algorithm. Since k-NN faces challenges with sensitivity to k values and computational complexity, a modification of k-NN was proposed, namely a multi-objective k-NN model. It was developed to incorporate multi-objective optimization and local density to create a more robust and efficient classification algorithm. The model dynamically determines the k value based on the sample density, optimizing accuracy and efficiency. Breast cancer data were collected from the University of Wisconsin Hospitals, Madison. The experimental results showed that the multi-objective k-NN model outperformed traditional k-NN and k-NN with feedback support. The proposed model achieved an accuracy of 93.7%, with precision values of 93% for the negative cancer class and 94% for the positive cancer class. These results indicate that the multi-objective k-NN model provides superior accuracy and precision in breast cancer detection, demonstrating its potential for clinical applications.
Penentuan bidang unggulan akademik universitas melalui metode topik Linear Discriminant Analysis (LDA) Mramra , Welianus Yohanes Yehuda; Nataliani, Yessica
IT Explore: Jurnal Penerapan Teknologi Informasi dan Komunikasi Vol 4 No 1 (2025): IT-Explore Februari 2025
Publisher : Fakultas Teknologi Informasi, Universitas Kristen Satya Wacana

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24246/itexplore.v4i1.2025.pp93-105

Abstract

The identification of academic flagship areas is crucial for higher education institutions as it helps enhance academic reputation and produce graduates who are competent in the workforce. However, the determination of these flagship areas is often subjective. Therefore, this study aims to identify the academic flagship areas at XYZ University using the Topic Modeling Latent Dirichlet Allocation (LDA) method. In this study, the LDA method is specifically used in the research field to analyze research journals at XYZ University from 2017 to 2021. The research results reveal three main topics that dominate the academic publications at XYZ University: business risk analysis, application system implementation particularly related to the pandemic, and information management with a framework in information technology. Based on this analysis, the flagship areas of XYZ University can be concluded to be in the fields of Information Technology and Business. Consequently, the University can more efficiently direct research resources and academic programs, as well as strengthen collaborations and its reputation as a center of excellence in these fields.
Pemanfaatan k-Means Clustering dan Analytic Hierarchy Process terhadap Penilaian Prestasi Kerja Pegawai Advensius Natalis; Yessica Nataliani
Jurnal Teknik Informatika dan Sistem Informasi Vol 8 No 1 (2022): JuTISI
Publisher : Maranatha University Press

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.28932/jutisi.v8i1.4243

Abstract

 As a government agency related to education and culture, the Department of Education and Culture of Bengkayang Regency needs qualified employees in their performance. Clustering can be used to determine whether the employee is performing well, moderately, or poorly. The clustering method used in this study is the k-Means method. The research wasconducted by studying and understanding the k-Means method and knowing employee performance data at the Education and Culture Office of Bengkayang Regency. The results of calculations using the k-Means method and the Analytic Hierarchy Process (AHP) method are as many as six employees have good performance (where two of them got the highest score in the AHP calculation), nine employees have moderate performance, and five employees have poor performance. These results can be used as a benchmark for employees in the cluster either to be promoted or rank, while employees in the cluster less are able to begiven employee performance training, in order to be better in the future. With this, employees can become more competitive and superior in the face of increasingly rapid developments.
Perancangan sistem informasi manajemen gudang berbasis web pada Cozy Mart Prandiska, Kelvin; Nataliani, Yessica
IT Explore: Jurnal Penerapan Teknologi Informasi dan Komunikasi Vol 4 No 2 (2025): IT-Explore Juni 2025
Publisher : Fakultas Teknologi Informasi, Universitas Kristen Satya Wacana

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24246/itexplore.v4i2.2025.pp186-195

Abstract

PT. Surya Cahaya Kembar or known as Cozy Mart aims to improve the delivery process and inventory management by using a web-based warehouse management information system. Companies have difficulty monitoring stock and distribution accurately as their business grows rapidly. The system analysis and software development method used in this research is Agile, which allows quick adjustments to user needs. The system is designed to include delivery tracking, stock management features and reports to assist staff in making decisions. System design uses Unified Modeling Language (UML) which describes system functionality as a whole. Test results show significant improvements in user satisfaction, as well as the speed and accuracy of warehouse management. PT. Surya Cahaya Kembar (Cozy Mart) hopes to reduce operational errors and optimize business processes by implementing this system. This helps the overall growth of the company.
Clustering Performa Pemain Basket Berdasarkan Posisi dan Statistik Pemain Menggunakan Metode Fuzzy c-Means Gregorry, Febrianus; Nataliani, Yessica
Jurnal Transformatika Vol. 20 No. 1 (2022): July 2022
Publisher : Jurusan Teknologi Informasi Universitas Semarang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26623/transformatika.v20i1.5137

Abstract

Satya Wacana Saints Salatiga is one of the professional teams that compete in Indonesian Basketball League (IBL). Player evaluation has been done as an effort to maintain the quality and performance of all players which can be used as team records. They can also make a couple of improvements within the team for the next season. Classifying the players performance by using the fuzzy c-means algorithm is the aim of this research. Player performance is determined based on five kinds of player statistical criteria, namely points, assists, blocks, rebounds, and steals from each position. The assessment carried out in this study is using weighting criteria for each position. The grouping by weighting aims to get the highest to the lowest scores from each player so that they can be grouped into three performance groups; good, moderate, and poor performance. The results of the fuzzy c-means grouping of 15 players of the Satya Wacana Saints Salatiga team with weighting obtained three players with good performance, five players with moderate performance, and seven players with poor performance. Meanwhile, the results of the fuzzy c-means grouping without weighting obtained five players with good performance, three players with moderate performance, and seven players with poor performance. Both grouping results are compared with the actual performance data. The result of the comparison was found that the grouping with weighting resulted in an accuracy rate of 100% and the grouping without weighting resulted in an accuracy rate of 86.67%. Grouping with weighting on different statistical values for each player position has an effect on player performance. Each player position has different strengths in scoring points, assists, rebounds, steals, block shoots, and field goals.
Sentiment Analysis of Customer Review Using Classification Algorithms and SMOTE for Handling Imbalanced Class Sediatmoko, Nur Siradj; Nataliani, Yessica; Suryady, Irwan
Indonesian Journal of Information Systems Vol. 7 No. 1 (2024): August 2024
Publisher : Program Studi Sistem Informasi Universitas Atma Jaya Yogyakarta

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24002/ijis.v7i1.8879

Abstract

Ralali.com is a B2B e-commerce platform that offers various brands across categories ranging from automotive to building materials. The Play Store is a tool for downloading applications used by many people. This research aims to compare and find the best model among Naïve Bayes (NB), Support Vector Machine (SVM), and k-Nearest Neighbor (k-NN) in classifying the sentiment reviews of Ralali.com's application on the Play Store, and analyze the negative labels to provide recommendations for Ralali.com developers. Based on the research results, the NB Algorithm stands out as the best choice compared to SVM and k-NN in addressing class imbalance. The use of SMOTE generally improves the model performance on minority classes for Precision, Recall, and F-Measure, although there are still challenges related to the lower Accuracy compared to the use of non-SMOTE.
TEXT MINING WITH LATENT DIRICHLET ALLOCATION FOR ANALYZING PUBLIC COMMENTS ON THE M-PASSPORT APPLICATION Hapsari, Theresia Shinta; Nataliani, Yessica
Jurnal Teknik Informatika (Jutif) Vol. 5 No. 4 (2024): JUTIF Volume 5, Number 4, August 2024
Publisher : Informatika, Universitas Jenderal Soedirman

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.52436/1.jutif.2024.5.4.1929

Abstract

The M-Passport application is a service application developed by the Directorate General of Immigration of Indonesia to assist the public in applying for new passports and replacing passports online. However, in its implementation, this application has not been able to give satisfaction to its users. It is proven by the low rating of the application and the numerous negative comments on the Google Play Store. One way to identify the application's shortcomings is by analyzing user comments. In analyzing the abundance of comment data, this study utilizes the text mining method with Latent Dirichlet Allocation (LDA) topic modeling. The analysis with this method aims to find topics frequently discussed in comments so that the government can identify the shortcomings of the M-Passport application. The results of comment analysis with LDA topic modeling produced seven topics, from which three topics with the highest coherence values were selected. These three topics are then interpreted to obtain information about the public's concerns regarding the M-Passport application. The results of this interpretation include users frequently failing to log in or register to the M-Passport application, users feeling that the M-Passport application does not assist them in passport management due to constraints in the online queue feature, and some users still finding it difficult to use the M-Passport application.