Claim Missing Document
Check
Articles

Found 2 Documents
Search
Journal : METIK JURNAL

Analisis Sentimen Menggunakan Metode IndoBERT Pada Ulasan Aplikasi Zoom Menggunakan Fitur Ekstrasi GloVe Andhika, Firza Rafiandi; Witanti, Wina; Nurul Sabrina, Puspita
METIK JURNAL (AKREDITASI SINTA 3) Vol. 9 No. 2 (2025): METIK Jurnal
Publisher : LPPM Universitas Mulia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47002/g38fxb78

Abstract

This research aims to conduct a sentiment analysis on Zoom application reviews by testing whether combining features from IndoBERT and GloVe can improve classification accuracy. The methodology begins with the collection of 5000 reviews from the Google Play Store, which then undergo a pre-processing stage. To address class imbalance, the Random Oversampling technique was applied. Features are extracted using IndoBERT for contextual meaning and GloVe for global semantic meaning, and their vectors are then combined through concatenation. The balanced dataset is divided into 80% training data and 20% testing data to train and evaluate the model. The test results show that the combined model achieved an overall accuracy of 91%, with a high precision value for the positive class (0.97) and a high recall value for the neutral class (0.95). For comparison, a model using only IndoBERT achieved 90% accuracy. Based on these results, it can be concluded that the combination of IndoBERT and GloVe is an effective and reliable approach for sentiment analysis, with its advantage lying in a richer feature representation due to the integration of global and contextual semantic information.
Analisis Tren dan Prediksi Penjualan Restoran Menggunakan Model Time Series Prophet Hidayat, Kiki; Witanti, Wina; Ramadhan, Edvin
METIK JURNAL (AKREDITASI SINTA 3) Vol. 9 No. 2 (2025): METIK Jurnal
Publisher : LPPM Universitas Mulia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47002/gd8y7q29

Abstract

Daily sales forecasting is a critical component of business planning that must adapt to the dynamics of market demand. While traditional approaches such as Single Moving Average and Trend Moment have been used in previous studies, their predictive accuracy on daily sales often remains suboptimal, with reported MAPE values up to 39.2%. Prophet, a time series model developed by Meta, offers enhanced flexibility in capturing non-linear trends, seasonality, and incorporating external regressors. This study proposes a hybrid forecasting model by combining Prophet with engineered features and external regressors, including calendar effects and recent sales statistics. The dataset consists of daily sales records that have undergone data cleaning, logarithmic transformation, and smoothing. Prophet is configured with additional monthly seasonality, national holiday indicators, and optimized parameters through grid search. Evaluation results demonstrate a substantial improvement, with the final model achieving an R² score of 0.9787 and a MAPE of 3.79%, outperforming conventional methods and aligning with the best results from recent Prophet-based studies. These findings confirm that the integration of external variables within Prophet significantly improves prediction accuracy, making it suitable for time series forecasting in various business domains with similar data patterns.