Article Per Year (5 Year)
p-Index From 2019 - 2024
0.444
P-Index
This Author published in this journals
All Journal Inform : Jurnal Ilmiah Bidang Teknologi Informasi dan Komunikasi The Indonesian Journal of Computer Science
Verra Budhi Lestari
Universitas Amikom Yogyakarta
Computer Science & IT Control & Systems Engineering Electrical & Electronics Engineering Engineering Library & Information Science

Published : 2 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 2 Documents
Search

 1 
Sentiment Analysis for IMDb Movie Review Using Support Vector Machine (SVM) Method D. Diffran Nur Cahyo; Fidya Farasalsabila; Verra Budhi Lestari; Hanafi; Tutik Lestari; Fahmi Rusdi Al Islami; M. Akbar Maulana
Inform : Jurnal Ilmiah Bidang Teknologi Informasi dan Komunikasi Vol. 8 No. 2 (2023)
Publisher : Universitas Dr. Soetomo

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (493.014 KB) | DOI: 10.25139/inform.v8i2.5700

Abstract

Many researchers currently employ supervised, machine learning methods to study sentiment analysis. Analysis can be done on movie reviews, Twitter reviews, online product reviews, blogs, discussion forums, Myspace comments, and social networks. Support Vector Machines (SVM) classifiers are used to analyze the Twitter data set using different parameters. The analysis and discussion were undertaken to allow for the conclusion that SVM has been successfully implemented utilizing the IMDb data for this study (Support Vector Machine). To complete this study, the preprocessing phase, which consisted of filtering and classifying data using SVM with a total of 50.000 data points, was completed after collecting up to 40.000 reviews to use as training data and 10.000 reviews to use as testing data. 25.000 positive and 25.000 negative points make up the view. In this study, we adopted an evaluation matrix including accurate, precision, recall, and F1-score. According to the experiment report, our model achieved SVM with Bags of Word (BoW) used to get results for the highest accuracy test, which was 88,59% accurate. Then, using grid-search, optimize against the SVM parameters to find the best parameters that SVM models can use. Our model achieved Term Frequency–inverse Document Frequency (TF-IDF) was used to get results for the highest accuracy test, which was 91,27% accurate.
Combining Bi-LSTM And Word2vec Embedding For Sentiment Analysis Models Of Application User Reviews Verra Budhi Lestari; Ema Utami; Hanafi
The Indonesian Journal of Computer Science Vol. 13 No. 1 (2024): 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.v13i1.3647

Abstract

Tujuan analisis sentimen aplikasi dan produk adalah untuk menentukan polaritas sentimen dalam ulasan aplikasi berdasarkan umpan balik pengguna. Model LSTM dan turunannya, seperti GRU, menjadi semakin populer di antara berbagai desain Jaringan Neural yang digunakan dalam analisis sentimen. Namun, model LSTM masih memiliki masalah terkait lambatnya konvergensi dan hanya dapat mengumpulkan data dalam satu arah, sehingga untuk meningkatkan kinerja, teknik pembelajaran mendalam lainnya seperti Bi-LSTM harus digunakan. Selain itu, kombinasi penyematan kata dapat dipertimbangkan untuk menghasilkan variasi representasi kata yang semakin besar. Penelitian ini menyajikan perbandingan model LSTM dan BiLSTM untuk mengetahui apakah kinerja kombinasi model BiLSTM dengan penerapan word embedding mampu memiliki kinerja yang lebih baik dibandingkan model LSTM satu arah. Hasil pengujian akurasi menunjukkan model BiLSTM dengan Word2Vec mendapatkan hasil pengujian akurasi tertinggi, dengan akurasi sebesar 87%. Hal ini membuktikan bahwa model BiLSTM dengan Word2Vec mampu memiliki performa yang lebih baik dibandingkan LSTM.
Co-Authors D. Diffran Nur Cahyo Ema Utami Fahmi Rusdi Al Islami Fidya Farasalsabila Hanafi Hanafi Maulana, Muhammad Akbar Tutik Lestari