Claim Missing Document
Check
Articles

Implementasi Extra Trees Classifier dengan Optimasi Grid Search CV pada Prediksi Tingkat Adaptasi AINA, LISTYA NUR; NASTITI, VINNA RAHMAYANTI SETYANING; ADITYA, CHRISTIAN SRI KUSUMA
MIND (Multimedia Artificial Intelligent Networking Database) Journal Vol 9, No 1 (2024): MIND Journal
Publisher : Institut Teknologi Nasional, Bandung

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26760/mindjournal.v9i1.78-88

Abstract

AbstrakTeknologi terus maju, terutama dalam komunikasi, pendidikan, dan informasi. Pendidikan online semakin diminati di banyak lembaga pendidikan, mendorong perlunya pemahaman sejauh mana peserta didik dapat beradaptasi dengan lingkungan online. Memprediksi tingkat adaptasi peserta didik menjadi penting untuk meningkatkan efektivitas dan kualitas pengalaman belajar. Dalam penelitian ini, menggunakan dataset dari Kaggle, metode Extra Trees Classifier dioptimalkan dengan Hyperparameter Tuning Grid Search CV. Sebelum optimalsi, akurasi mencapai 95.85%, setelahnya meningkat menjadi 96.26%, menunjukkan peningkatan sebesar 0.41%. Implementasi metode Extra Trees Classifier dengan optimasi Hyperparameter Tuning Grid Search CV lebih unggul dibandingkan penggunaan algoritma tanpa optimasi.Kata kunci: Prediksi, Extra Trees, Classifier, Hyperparameter, CVAbstractTechnology continues to advance, especially in communication, education and information. Online education is increasingly in demand in many educational institutions, prompting the need to understand the extent to which learners can adapt to the online environment. Predicting learners' adaptation level is important to improve the effectiveness and quality of the learning experience. In this study, using a dataset from Kaggle, the Extra Trees Classifier method was optimized with Hyperparameter Tuning Grid Search CV. Before optimization, the accuracy reached 95.85%, after which it increased to 96.26%, showing an improvement of 0.41%. The implementation of the Extra Trees Classifier method with Hyperparameter Tuning Grid Search CV optimization is superior to the use of the algorithm without optimization.Keywords: Prediction, Extra Trees, Classifier, Hyperparameter, CV
Perbandingan Kinerja Pre-Trained Word Embedding Terhadap Performa Klasifikasi Sentimen Ulasan Produk Tokopedia Dengan Long Short-Term Memory(LSTM) Dirfas, Naufal Angling; Nastiti, Vinna Rahmayanti Setyaning
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.v6i2.5634

Abstract

The product review dataset is a rapidly growing and interesting source of data to explore. The increase in the number of internet users and customer shopping habits through online stores has a significant impact on the growth of product review data, especially for online stores in Indonesia, such as Tokopedia. The sample data used amounted to 1079. This research aims to evaluate the performance of three types of pre-trained word embeddings, namely FastText, GloVe, and Word2Vec, in the Long Short-Term Memory (LSTM) model for sentiment classification of product reviews on Tokopedia. An automated sentiment classification system is needed to process many product reviews, making it easier for sellers to know what consumers think of their products. This research contributes by evaluating the impact of various pre-trained word embeddings on the performance of LSTM models in sentiment classification tasks. In addition, this research also aims to measure the effectiveness of LSTM models combined with multiple pre-trained word embeddings. By implementing a deep learning architecture, computers can learn and recognize contextual data stored in review sentences. The research was conducted in three stages: model selection, layer setup, and hyperparameter optimization, to feature in-depth testing of the deep learning architecture used and the appropriate combination of layers and parameters to obtain high sentiment classification performance. The experimental results show that FastText with LSTM provides the best performance with 85.08% accuracy, Word2Vec with 84.62% accuracy, and GloVe with 83.04% accuracy. The main contribution of this research is to present an in-depth test of the product review dataset and provide a deep learning architecture along with a combination of layers and parameters that has the best performance in recognizing sentiment on the product review dataset. This architecture achieves higher performance than the BERT method with CNN and BiLSTM layers.
Klasifikasi Tingkat Kemampuan Adaptasi Siswa dalam Pembelajaran Online Menggunakan Decision Tree Lailiyah, Asmaul; Nastiti, Vinna Rahmayanti Setyaning; Wahyuni, Evi Dwi; Aditya, Christian Sri Kusuma
Techno.Com Vol. 23 No. 1 (2024): Februari 2024
Publisher : LPPM Universitas Dian Nuswantoro

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.62411/tc.v23i1.9739

Abstract

Kemajuan dalam ilmu pengetahuan dan teknologi mendorong adaptasi terhadap pemanfaatan teknologi di berbagai sektor, seperti komunikasi, pendidikan, dan informasi. Terutama dalam konteks teknologi pendidikan, dapat diamati bahwa pembelajaran online sedang mendapatkan popularitas yang signifikan di berbagai lembaga pendidikan. Oleh karena itu, penting untuk mengeksplorasi seberapa baik peserta didik dapat beradaptasi dengan lingkungan pembelajaran online. Memprediksi tingkat adaptasi peserta didik memiliki signifikansi yang besar bagi pendidik dan pengembang platform pembelajaran online, dengan tujuan meningkatkan efisiensi dan kualitas pengalaman belajar. Penelitian ini menggunakan dataset “Students Adaptability Level in Online Education” dengan menerapkan pendekatan Algoritma Decision Tree. Hasil penelitian memperoleh akurasi sebesar 95%, meningkat 7,44% dari penelitian sebelumnya yang hanya memperoleh akurasi sebesar 87,56% dengan menggunakan algoritma yang sama tanpa Feature Engineering. Hal ini menunjukkan bahwa Feature Engineering memegang peranan penting dalam klasifikasi tingkat kemampuan adapatasi siswa untuk mendapatkan hasil yang baik dengan akurasi yang tinggi.
DANA App Sentiment Analysis: Comparison of XGBoost, SVM, and Extra Trees Setiawan, Muhamad Jodi; Nastiti, Vinna Rahmayanti Setyaning
Jurnal Sisfokom (Sistem Informasi dan Komputer) Vol. 13 No. 3 (2024): NOVEMBER
Publisher : ISB Atma Luhur

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.32736/sisfokom.v13i3.2239

Abstract

This research aims to analyze sentiment on DANA application reviews to find out user perceptions by comparing Extreme Gradient Boosting (XGBoost), Support Vector Machine (SVM), and Extra Trees Classifier classification methods. DANA application review data is obtained from the Kaggle site which consists of 50,000 Indonesian-language reviews labeled with positive and negative sentiments. The research stages include data preprocessing to clean and prepare the review text, applying word weighting using Word2Vec to give weight to words based on their context, balancing sentiment classes using SMOTE to address the imbalance of positive and negative review classes. It should be noted that the initial proportion of data before applying SMOTE may affect the results. The data is then divided into training and testing sets, then the models are trained and evaluated using Confusion Matrix and K-Fold Cross-Validation. The results of the three classification methods are measured by the accuracy matrix and F1-Score to assess model performance, the SVM and XGBoost methods obtained an accuracy of 93% and the ETC method achieved an F1-Score value of 96% at K=6, the three models proved to be very accurate in predicting the sentiment of DANA application reviews both positive and negative. The practical implications of this research can identify areas for application improvement, develop popular features, personalize services based on user preferences, and manage application reputation.
Analisis Sentimen dalam Opini Publik di Chanel Youtube Indonesia Lawyers Club Tentang Isu Populer dengan Menggunakan Metode LSTM dan Bi-LSTM Wicaksono, Bagus; Nastiti, Vinna Rahmayanti Setyaning
Jurnal Algoritma Vol 21 No 2 (2024): Jurnal Algoritma
Publisher : Institut Teknologi Garut

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33364/algoritma/v.21-2.1696

Abstract

One of the social media that often shares news and is popularly used in Indonesia is YouTube. Through the YouTube platform readers can leave comments to share their opinions under related news. This news and commentary has been a great source of information and research. This article presents a data set containing 368,299 public comments and replies from 73 video news published from 01 January 2024 to 18 May 2024 from the famous “YouTube Indonesia Lawyers Club” channel. To ensure commenter privacy, commenter names are coded in the dataset. This dataset is open for use by researchers with the access link https://data.mendeley.com/datasets/h9335fgsgr/1. This data can help researchers identify patterns in public opinion and analyze how public opinion changes over time. The news topic most liked by the public is related to the Jokowi government, resulting in an accuracy value for the LSTM method of 99.61% while with the LSTM method the accuracy is 98.11%.
Evaluasi Usability dan Rekomendasi Perbaikan Website SIP Bro Menggunakan Metode SUS Dan Think Aloud Naila, Tia Cahyani; Wahyuni, Evi Dwi; Rahmayanti, Vinna
POSITIF : Jurnal Sistem dan Teknologi Informasi Vol 9 No 2 (2023): Positif : Jurnal Sistem dan Teknologi Informasi
Publisher : P3M Politeknik Negeri Banjarmasin

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31961/positif.v9i2.1917

Abstract

In terms of governance, websites have a strategic function as tools that support government activities, where their existence plays a role in digital engagement with the public. One of the government agencies in Blora City, namely the Regional Development and Planning Agency (BAPPEDA), has utilized and implemented an administrative website in the field of research and development, called SIP Bro. Based on observations, after the implementation of the SIP Bro website, there are still issues or weaknesses identified. So far, there have been no efforts to conduct a minimum evaluation to assess its Usability and whether the intended goals are achieved. Therefore, this study aims to determine the Usability analysis scores of the SIP Bro website using the System usability scale method. To further enhance the effectiveness of evaluating and developing the SIP Bro website, this research also incorporates the Think Aloud approach. The research findings conclude that the Usability score of the SIP Bro website obtained a score of 61.071, which falls under the "Ok" category, supported by a grade scale value in the D range and acceptability ranges categorized as marginal low, indicating that the website is acceptable but with a relatively low level of acceptance. The final analysis of the Think Aloud method resulted in 23 recommendations for improvement. The recommendations made are to enhance and develop the SIP Bro website for better performance in the future.
Students Final Academic Score Prediction Using Boosting Regression Algorithms Dignifo Nauval Muhammady; Haidar Aldy Eka Nugraha; Vinna Rahmayanti Setyaning Nastiti; Christian Sri Kusuma Aditya
Jurnal Ilmiah Teknik Elektro Komputer dan Informatika Vol. 10 No. 1 (2024): March
Publisher : Universitas Ahmad Dahlan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26555/jiteki.v10i1.28352

Abstract

Academic grades are crucial in education because they assist students in acquiring the knowledge and skills necessary to succeed in school and their future. Accurately predicting students' final academic performance grade score is important for educational decision-makers. However, creating precise prediction models based on students' historical data can be challenging due to the complex nature of academic data. This research analyzes student academic data totaling 649 Portuguese language course student data that has been processed according to data requirements which are then predicted using XGBoost Regressor, Light Gradient Boosting Machine (LGBM), and CatBoost. This research aims to develop a robust prediction model that can effectively predict students' final academic performance. This research offers valuable insights into the factors that influence academic success and provides practical implications for educational institutions looking to improve their decision-making processes. The prediction requires identifying key predictors of academic performance, such as previous grades, attendance records, and socio-economic background. The research makes a contribution by improving the matrix MAE in this research is less than the previous research from 2.2 average each algorithm to 0.22 average, this less MAE means the better model. The research achieved MAE score of 0.22 average. In conclusion, this research is expected to address the challenge of predicting student academic performance through the application of advanced machine learning techniques. The results provide valuable insights for decision-makers in education and highlight the importance of a data-driven approach to improving academic performance. By utilizing machine learning algorithms, educational institutions can effectively support student learning and success.
PERBANDINGAN KINERJA PRE-TRAINED INDOBERT-BASE DAN INDOBERT-LITE PADA KLASIFIKASI SENTIMEN ULASAN TIKTOK TOKOPEDIA SELLER CENTER DENGAN MODEL INDOBERT Wildan Amru Hidayat; Nastiti, Vinna Rahmayanti Setyaning
Jurnal Sistem Informasi Vol 11 No 2 (2024)
Publisher : Universitas Serang Raya

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30656/jsii.v11i2.9168

Abstract

Era digital telah membawa revolusi dalam dunia e-commerce dengan mengintegrasikan platform media sosial dan platform e-commerce, yang menghasilkan inovasi seperti aplikasi TikTok Tokopedia Seller Center. Aplikasi ini menggabungkan platform e-commerce dengan fitur media sosial, memungkinkan pengguna untuk mengelola penjualan sekaligus memperluas jangkauan pasar dan mempromosikan produk melalui video pendek yang interaktif pada platform media sosial TikTok. Dengan adanya inovasi fitur baru dalam aplikasi ini, penelitian ini melakukan analisis sentimen untuk memahami persepsi dan ulasan berbahasa Indonesia dari para pengguna aplikasi TikTok Tokopedia Seller Center menggunakan model deep learning IndoBERT. Data ulasan dikumpulkan menggunakan teknik scraping pada Google Play Store sebanyak 3.145 ulasan yang dilabeli secara manual menjadi 1.755 klasifikasi sentimen negatif dan 1390 klasifikasi sentimen positif. Tahapan preprocessing seperti teks cleaning, case folding, normalisasi teks, dan stopword removal dilakukan untuk memberihkan data teks sebelum digunakan untuk pelatihan model. Data yang sudah dibersihkan terbagi menjadi 64% data training sebesar 2.012 data, 16% data validation sebesar 504 data, dan 20% data testing sebesar 629 data. Dua varian pre-trained model IndoBERT, yaitu Indobert-base-p2 versi besar dan Indobert-lite-base-p2 versi lebih ringan digunakan dalam penelitian ini untuk pemrosesan bahasa alami khusus bahasa Indonesia. Hasil penelitian menunjukkan bahwa komparasi model IndoBERT dengan kedua pre-trained menunjukkan bahwa pre-trained Indobert-base-p2 mendapatkan hasil akurasi yang lebih unggul dibandingkan Indobert-lite-base-p2, dengan akurasi sebesar 97%, presisi sebesar 97%, recall sebesar 97%, dan f1-score sebesar 97%, sedangkan pre-trained Indobert-lite-base-p2 dengan akurasi sebesar 94%, presisi sebesar 94%, recall sebesar 94%, dan f1-score sebesar 94%.
Mapping The Relationship Between Virtual Reality and Bullying Prevention: An Analysis Of Bibliographic Coupling Maryaeni, Maryaeni; Nastiti, Vinna Rahmayanti Setyaning; Oktaviani, Chintya Tria Diana; Reikisyifa, Clarissa Sanindita; Kenanga, Larynt Sawfa; Kusuma, Wahyu Andhyka; Wahyuni, Evi Dwi
JOIV : International Journal on Informatics Visualization Vol 9, No 2 (2025)
Publisher : Society of Visual Informatics

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.62527/joiv.9.2.2183

Abstract

Bullying prevention has become a significant topic in contemporary society. The bibliographic analysis method employed is bibliographic coupling, which allows for identifying relationships among relevant scientific articles to understand the role of technology in combating bullying. The research methodology involves identifying previous publications on virtual reality and bullying prevention from Scopus. Bibliographic analysis quantifies relationships between publications based on shared references. The bibliographic data collected focuses on VR-related literature and bullying prevention from various academic sources. This article discusses the interconnections among the reviewed articles regarding the impact of these findings, such as the development of VR applications that can enhance social skills and empathy, which are crucial factors in preventing bullying. Research results indicate a significant correlation between virtual reality and bullying prevention. Relevant prior studies include topics such as using virtual reality to avoid bullying through bystanders and victims, as well as simulating bullying to enhance bystander empathy. These studies provide information on the role of virtual reality technology in effectively combating bullying. Research findings are presented as descriptive and visual analyses using VOS viewer software. Additionally, the article underscores the importance of interdisciplinary collaboration in integrating VR into effective bullying prevention strategies, making us all part of a larger community working towards a common goal. Hopefully, this article will provide a foundation for future research and the development of technology applications with the potential to combat bullying.
PENGEMBANGAN SISTEM INFORMASI DAN MAJALAH DIGITAL DI PIMPINAN CABANG MUHAMMADIYAH LAWANG Vinna Rahmayanti Setyaning Nastiti; Denar Regata Akbi
Community Development Journal : Jurnal Pengabdian Masyarakat Vol. 5 No. 1 (2024): Volume 5 No 1 Tahun 2024
Publisher : Universitas Pahlawan Tuanku Tambusai

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31004/cdj.v5i1.24743

Abstract

Pimpinan Cabang Muhmmadiyah (PCM) Lawang merupakan salah satu organisasi keagamaan yang memiliki peran penting dalam meningkatkan kualitas dakwah di Lawang. Salah satu upaya yang dilakukan oleh PCM Lawang adalah memanfaatkan teknologi informasi dan komunikasi (TIK). Namun, PCM Lawang belum memiliki website untuk dokumentasi kegiatan-kegiatan dakwah. Tim pengabdi melakukan pengabdian kepada masyarakat di PCM Lawang dengan tujuan untuk mengembangkan system informasi dan memberikan pelatihan pembuatan majalah digital. Sistem informasi berbasis website tersebut berfungsi untuk website resmi kepengurusan PCM Lawang dan majalah digital berfungsi untuk dokumentasi kegiatan dari kegiatan dakwah PCM Lawang. Hasil dari pengabdian ini adalah prototype system informasi PCM Lawang dan pelatihan konten majalah digital yang diikuti oleh perwakilan dari PCM Lawang.