Claim Missing Document
Check
Articles

Found 27 Documents
Search

Pengukuran Kesiapan Transformasi Digital Smart City Menggunakan Aplikasi Rapid Miner Pascalina, Donna; Widhiastono, Raymondhus; Juliane, Christina
Technomedia Journal Vol 7 No 3 Februari (2023): TMJ (Technomedia Journal)
Publisher : Pandawan Incorporation, Alphabet Incubator Universitas Raharja

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (280.526 KB) | DOI: 10.33050/tmj.v7i3.1914

Abstract

Digital transformation of organizational change to be more effective and efficient in a city, digital transformation in the city is not yet ready, to determine readiness for change it is necessary to measure readiness in the Smart City Digital Transformation using quantitative data from Human Resources on readiness measurements carried out directly through surveys to all OPD Semarang City. Researchers use Data Mining and Decision Tree C4.5 Algorithm to examine the data, Research uses RapidMiner. The results of this study have an accuracy rate of 82.05% from 36 OPD agencies with 2 rules, namely ready and not ready in the city of Semarang, which are declared not ready with 3 enabler parameters, namely Understanding of Transformation, Cultural Transformation, Competence and Basic Knowledge.
Evaluasi Tingkat Kepuasan Mahasiswa Terhadap Pelayanan Akademik Menggunakan Metode Klasifikasi Algoritma C4.5 Widiastuti, Tri; Karsa, Koko; Juliane, Christina
Technomedia Journal Vol 7 No 3 Februari (2023): TMJ (Technomedia Journal)
Publisher : Pandawan Incorporation, Alphabet Incubator Universitas Raharja

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (400.773 KB) | DOI: 10.33050/tmj.v7i3.1932

Abstract

The purpose of this study was to determine the effect of academic services on student satisfaction so that students do not feel disappointed with academic services. This study measures the level of student satisfaction with the existing academic services at Jenderal Achmad Yani University, Cimahi. The data set from the survey results of student satisfaction with academic services at Unjani is used to generate models, rules and accuracy scores for student satisfaction using the Decision Tree C4.5 algorithm data mining classification method, to see the results of the accuracy values ​​of several attributes, namely tangible, empathetic, responsiveness. , reliability and assurance. The results of the tests carried out with the rapidminer application, the accuracy value of the 7 (Seven) Faculties testing at Unjani resulted in a value above 90%, which means that this value indicates that the service that has been running so far is considered very good. Testing student satisfaction surveys must of course be carried out continuously to be able to continue to improve academic services to students for the better.
Analisis Sentimen Terhadap Cryptocurrency Berbasis Python TextBlob Menggunakan Algoritma Naïve Bayes Azhar, Rizaldi; Surahman, Adi; Juliane, Christina
J-SAKTI (Jurnal Sains Komputer dan Informatika) Vol 6, No 1 (2022): EDISI MARET
Publisher : STIKOM Tunas Bangsa Pematangsiantar

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30645/j-sakti.v6i1.443

Abstract

Cryptocurrency users are now increasing as the market becomes more and more attractive. In 2019 recorded around 139 million account users verified id cryptocurrency. Recently, it was enlivened by the emergence of #crypto on Twitter and had become a world trending topic. This gives rise to many opinions and opinions from twitter users. With so many twitter users' opinions on the hashtag, it is very difficult to know whether positive, negative or neutral sentiments are manual. This requires machine learning to be able to automate labeling, be it positive, neutral or negative sentiments. Machine learning used is by utilizing Python TextBlob. The results of automatic labeling using Python TextBlob from a total of 1032 tweets obtained 632 tweets or 61.24% containing positive sentiments, 296 neutral sentiments or 28.68% tweets and 104 negative sentiments or 10.07%. The test results using the Naïve Bayes algorithm with each testing data and training data are 0.2 and 0.8. From this test, the accuracy value is 71.98%, precision is 83.04%, recall is 60.88% and f1_score is 65.07%.
Analisis Sentimen Terhadap Cryptocurrency Berbasis Python TextBlob Menggunakan Algoritma Naïve Bayes Azhar, Rizaldi; Surahman, Adi; Juliane, Christina
J-SAKTI (Jurnal Sains Komputer dan Informatika) Vol 6, No 1 (2022): EDISI MARET
Publisher : STIKOM Tunas Bangsa Pematangsiantar

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30645/j-sakti.v6i1.443

Abstract

Cryptocurrency users are now increasing as the market becomes more and more attractive. In 2019 recorded around 139 million account users verified id cryptocurrency. Recently, it was enlivened by the emergence of #crypto on Twitter and had become a world trending topic. This gives rise to many opinions and opinions from twitter users. With so many twitter users' opinions on the hashtag, it is very difficult to know whether positive, negative or neutral sentiments are manual. This requires machine learning to be able to automate labeling, be it positive, neutral or negative sentiments. Machine learning used is by utilizing Python TextBlob. The results of automatic labeling using Python TextBlob from a total of 1032 tweets obtained 632 tweets or 61.24% containing positive sentiments, 296 neutral sentiments or 28.68% tweets and 104 negative sentiments or 10.07%. The test results using the Naïve Bayes algorithm with each testing data and training data are 0.2 and 0.8. From this test, the accuracy value is 71.98%, precision is 83.04%, recall is 60.88% and f1_score is 65.07%.
Analisis Sentimen Putusan Mahkamah Konstitusi terhadap Batas Usia Capres dan Cawapres Menggunakan IndoBERT Septian, Luffi; Aljauza, Teguh; Juliane, Christina
The Indonesian Journal of Computer Science Vol. 12 No. 6 (2023): The Indonesian Journal of Computer Science (IJCS)
Publisher : AI Society & STMIK Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33022/ijcs.v12i6.3614

Abstract

Putusan Mahkamah Konstitusi nomor 90/PUU-XXI/2023 tentang batas usia calon presiden dan wakil presiden telah memicu perbincangan masyarakat. Hal ini ditandai dengan kata kunci ‘Putusan MK” pada media sosial Twitter/X menduduki peringkat tiga trending topik Nasional selama pertengahan bulan Oktober. Putusan tersebut dinilai kontroversial karena berkaitan dengan momentum Pemilihan Presiden 2024. Peneliti tertarik untuk memanfaatkan data dari media sosial twitter/X dalam menganalisis respon masyarakat terhadap Putusan Mahkamah Konstitusi dengan cara mengklasifikasikan respon tersebut ke dalam sentimen. Model yang digunakan dalam penelitian ini adalah IndoBERT, sebuah arsitektur transformer BERT yang dikembangkan oleh tim IndoNLU. Metode ini dipilih berdasarkan efektivitasnya dalam memproses teks berbahasa Indonesia untuk mengidentifikasi dan mengategorikan opini publik menjadi positif, negatif, atau netral terkait dengan keputusan Mahkamah Konstitusi. Hasil awal menunjukkan model IndoBERT tanpa augmentasi data mencapai akurasi 0.81 dan F1 skor 0.58. Selanjutnya, penggunaan teknik Synthetic Minority Over-sampling Technique (SMOTE) meningkatkan F1 skor namun tidak berdampak signifikan pada akurasi. Eksperimen selanjutnya dengan augmentasi random swap, menghasilkan peningkatan performa yang substansial, dimana model IndoBERT mencapai akurasi dan F1 skor sama-sama pada angka 0.90.
Optimalisasi Metode Naive Bayes Classifier Untuk Prediksi Persetujuan Kredit Syakur, Achmad; Purwandi Putra, Rendri; Juliane, Christina
The Indonesian Journal of Computer Science Vol. 13 No. 1 (2024): The Indonesian Journal of Computer Science (IJCS)
Publisher : AI Society & STMIK Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33022/ijcs.v13i1.3622

Abstract

Kredit adalah bentuk pembiayaan yang banyak orang ajukan ke bank atau perusahaan penyedia kredit. Dalam proses pengajuan kredit, dilakukan analisis untuk menentukan apakah kredit yang diajukan layak atau tidak. Penelitian ini bertujuan untuk membantu bank atau perusahaan penyedia kredit dalam melakukan persetujuan kredit dengan efektif dan akurat dalam menentukan status pengajuan. Penelitian ini menggunakan teknik data mining dan kumpulan dataset yang berasal dari kaggle.com. Terdapat 12 atribut dan 2 kelas yang digunakan dalam penelitian ini. Dalam penelitian ini, metode klasifikasi Naive Bayes dan optimasi kelompok partikel (PSO) digunakan. Prediksi persetujuan kredit dengan metode naïve bayes classifier menghasilkan nilai akurasi sebesar 80,00% dengan nilai AUC 0,884. Sebaliknya, prediksi persetujuan kredit dengan metode particle swarm optimization (PSO) menghasilkan nilai akurasi sebesar 96,67% dengan nilai AUC 0,69.
Kajian Data Mining untuk Klasifikasi Gender Menggunakan Data Wajah dengan Algoritma Naive Bayes dan K Nearest Neighbor (KNN) Fathah, Adittia; Juliane, Christina
Jurnal Teknologi Informasi dan Ilmu Komputer Vol 12 No 1: Februari 2025
Publisher : Fakultas Ilmu Komputer, Universitas Brawijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.25126/jtiik.2025128724

Abstract

Identifikasi gender saat ini lebih sulit dilakukan. Penyebabnya antara lain banyaknya klasifikasi gender, penggunaan identitas palsu di media sosial dan semakin maraknya foto palsu. Peristiwa nyata yang terjadi adalah banyaknya klasifikasi di negara Thailand yang memiliki 18 gender. Peristiwa lainnya adalah penambahan gender “X” pada aplikasi permohonan passport di Amerika dan beredarnya foto palsu yang diedit dengan aplikasi FaceApp. Kejadian tersebut menyebabkan perlunya membuat model yang bisa melakukan klasifikasi gender agar gender asli dari seseorang bisa diketahui. Penelitian dilakukan dengan mencari model yang bisa mengklasifikasikan gender. Caranya adalah dengan membandingkan hasil akurasi dua algoritma yaitu Naïve Bayes dan KNN (K Nearest Neighbor). Metode yang digunakan mengikuti tahapan dalam KDD (Knowledge Discovery in Database). Atribut yang dipakai adalah bagian-bagian pada wajah yaitu lebar dahi, lebar hidung, panjang hidung, bibir dan jarak hidung ke bibir. Akurasi kedua algoritma diuji dengan metode Cross Validation dan Confusion Matrix. Tujuan penelitian ini adalah memastikan apakah atribut wajah dapat digunakan untuk klasifikasi gender serta menentukan model yang lebih baik antara Naïve Bayes atau KNN. Hasil pengujian menunjukkan, kedua algoritma memiliki akurasi yang sangat baik. Namun algoritma Naïve Bayes memiliki nillai AUC yang lebih tinggi yaitu 0,996 dibanding algoritma KNN yang memiliki nilai AUC sebesar 0,992. Berdasarkan nilai tersebut, atribut bagian-bagian pada wajah yaitu lebar dahi, lebar hidung, panjang hidung, bibir dan jarak hidung ke bibir dapat digunakan untuk klasifiikasi gender, karena bisa menghasilkan akurasi yang baik. Namun, model Naïve Bayes lebih direkomendasikan karena nilai akurasinya lebih tinggi dan stabil. 
Global Network Cyberattack Classification Using Naive Bayes Method Time Range 2020 – 2023 Sandi Mutia, Acep; Irawan, Irawan; Juliane, Christina
ASTONJADRO Vol. 13 No. 2 (2024): ASTONJADRO
Publisher : Universitas Ibn Khaldun Bogor

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.32832/astonjadro.v13i2.15683

Abstract

This study focuses on developing a classification model for cyberattacks on global networks during the time span of 2020 to 2023 using the Naive Bayes method. The main objective of the study is to analyze and classify the frequent severity of cyber, which helps in improving network security and reducing vulnerabilities. The Naive Bayes method was chosen for its efficiency in handling large datasets and its ability to make predictions based on probabilities. Collecting cyberattack data from a variety of reliable and up-to-date sources, the study covers attacks such as ransomware, phishing, DDoS, and other malware. The classification process includes data pre-processing, feature extraction, and finally the application of Naive Bayes algorithms to identify patterns in such attacks. The classification results are then evaluated using the Apply Model and Performance validation methods to assess the effectiveness of the model. The results of this study show that Naive Bayes is able to accurately classify cyberattacks, providing a useful tool for cybersecurity professionals to understand attack trends and respond proactively. The study also suggests areas for further research, including the integration of the Naive Bayes model with other artificial intelligence systems for improved cyberattack detection. The study provides new insights into the application of the Naive Bayes method in cybersecurity and paves the way for improved data-driven cyber defense strategies.
Analysis and Design of Student Point Systems to Improve Student Achievement using The Clustering Method Bani Riyan, Ade; Fikri Rifai, Mochamad; Juliane, Christina
Journal of World Science Vol. 2 No. 3 (2023): Journal of World Science
Publisher : Riviera Publishing

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.58344/jws.v2i3.155

Abstract

The student points system is an application for recording students' achievement and offense points. The lack of recording and dissemination of information on achievement results makes students less motivated to improve achievement, and the distribution of scholarships for outstanding students is inappropriate. To improve student achievement, an application program is needed that can record and disseminate student achievement data in real-time, accurate, and effective. So, the purpose in this study is to know and analyze the design of the student point system to improve student achievement using the clustering method. Researchers use the Clustering Method in calculating data to determine the accuracy of scholarship distribution for outstanding students. Clustering with the most achievement points is clustering 2 with 25,254 Achievement Points. The total number in the level 2 cluster is 1,797 which indicates the number is close to 2,000 or 2 which is the result of data transformation from the junior high level. The implication of clustering research on student point data is to provide useful information for the Foundation as an institution that houses schools in allocating scholarships for outstanding students. In this case, clustering 2 with the highest number of Achievement Points indicates that there is a group of students with high achievement points. By using the clustering results, the Foundation can allocate scholarships more effectively and efficiently, because it can identify outstanding students from various school levels more easily.
Application of Data Mining to Measure the Effectiveness of the Islamic Boarding School’s Independent Curriculum based on Learning Achievement using the Clustering Method Imron Rosyadi, Iim; Nurhadits, Fitri; Juliane, Christina
Journal of World Science Vol. 3 No. 5 (2024): Journal of World Science
Publisher : Riviera Publishing

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.58344/jws.v3i5.595

Abstract

The evolution of educational curricula has been a focal point for institutions aiming to enhance learning outcomes and adapt to students' diverse needs. In this context, Islamic boarding schools, or pesantren, are increasingly exploring independent curricula to better serve their students. This research aims to measure the effectiveness of the independent curriculum at the Al Binaa Bekasi Islamic Boarding School, especially regarding learning achievements in general and Islamic subjects. The method used is data mining clustering to analyze student learning achievement data. In the initial stage, the data collected includes student scores in general subjects (such as Islamic Religious Education, Pancasila Education, Indonesian, English, Mathematics, Science, Social Sciences, Arts, Sports, ICT, Sundanese) and Syar'i (Quran tajwid, hadith, aqidah, fiqh, Hadassah, short). Then, data mining clustering techniques are used to group students based on their achievements in the two subjects. The results of the analysis show that the independent curriculum at Al Binaa Islamic Boarding School effectively increases student learning achievement. The groups formed from data mining clustering show patterns consistent with curriculum objectives, where students in the same group have similar levels of achievement in general and star subjects. This indicates that the independent curriculum has succeeded in leveling student learning achievement. This research contributes to understanding the effectiveness of the independent curriculum in Islamic boarding schools. It can be a basis for further development in designing Islamic boarding school education curricula that are more adaptive and responsive to student needs.