Skip to content
Review Open access

Preprocessing Effects on Word2Vec and IndoRoBERTa for Shopee Sentiment Classification

Oct 2026 · JOURNAL OF APPLIED INFORMATICS AND COMPUTING · 0 citations · 21 references

TL;DR

The findings show that preprocessing effectiveness depends on model architecture, and minimal preprocessing better preserves contextual information for IndoRoBERTa, whereas aggressive preprocessing provides only a marginal benefit for the static baseline.

Abstract

This study evaluates the effects of heavy and minimal preprocessing on static and contextual text representations for sentiment classification of Indonesian Shopee reviews. The dataset comprised 29,618 Google Play Store reviews collected in January 2026 and weakly labeled from star ratings into negative, neutral, and positive classes. The data were divided into training, validation, and test sets using an 80:10:10 stratified split. Heavy preprocessing included case folding, text cleaning, slang normalization, stopword removal, and stemming, whereas minimal preprocessing applied case folding only. Four experimental configurations combined the two preprocessing strategies with Word2Vec–Logistic Regression and fine-tuned IndoRoBERTa. Model performance was evaluated using accuracy, weighted precision, weighted recall, weighted F1-score, macro F1-score, and confusion matrices. Heavy preprocessing produced a marginal weighted F1-score improvement for Word2Vec–Logistic Regression, from 86.0% to 86.2%. Minimal preprocessing improved IndoRoBERTa’s weighted F1-score from 88.4% to 89.7% and achieved the best accuracy of 90.7% and macro F1-score of 65.8%. The difference between weighted and macro F1-scores indicates that the neutral minority class remained difficult to identify. The findings show that preprocessing effectiveness depends on model architecture. Minimal preprocessing better preserves contextual information for IndoRoBERTa, whereas aggressive preprocessing provides only a marginal benefit for the static baseline.

Read PDF

Similar papers

Review Open access Oct 2026

Optimal Preprocessing and Feature Engineering Combination for DANA App Sentiment Classification: A Full Factorial Study

Purpose: Selecting optimal text-processing methods for sentiment analysis of digital financial application reviews, especially in Indonesian, is challenging. Previous studies have typically evaluated preprocessing techniques, feature engineering methods, and classification models in isolation, leaving the question of t...

Samsudin Samsudin, Linda Putriani · 0 citations
Conference Open access Sep 2026

Sentiment Analysis of TikTok Comments Using Fine-Tuned IndoBERT for Indonesian Brand Perception

The growth of user generated content on social media has increased the need for automatic sentiment analysis of unstructured text. This research fine tuned an IndoBERT based architecture for ternary sentiment classification of 3,857 TikTok comments on Point Coffee, collected between January 2020 and February 2025. The...

Romindo, K. Sirait, Triandes Sinaga et al. · 0 citations
Review Open access Oct 2026

Sentiment Analysis of Indonesian E-Commerce Product Reviews Using Indobert Fine-Tuning on A Multi-Source Combined Corpus

Consumer reviews on Indonesian e-commerce platforms now arrive faster than manual analysis can process them, and IndoBERT has become a common choice for classifying them automatically. Many studies, however, train on a single data source and report one training run, with no documented hyperparameter search and no test...

Aditya Rahmatullah, A. Rizaldy, Syahbudin et al. · 0 citations
Review Open access Oct 2026

Sentiment Analysis of ChatGPT Reviews on Play Store using Support Vector Machine

User reviews on Google Play Store contain opinions that can be used to evaluate digital applications, but large review volumes make manual analysis inefficient. This study analyzes 8,000 ChatGPT reviews collected from January to May 2026 using the CRISP-DM framework. Text preparation includes case folding, cleaning, to...

G. Handayani, Hotler Manurung, Melda Pita Uli Sitompul · 0 citations
Open access Oct 2026

Enhancing Multi-Domain Sentiment Classification Using Adaptive Lexicons and Machine Learning

Multi-domain sentiment classification remains challenging because lexical polarity and expression patterns vary across application contexts. This study evaluates an enhanced iVADER framework that combines domain-specific lexicon adaptation with text preprocessing and machine-learning-supported feature representations....

Saleh Yaakub, M. Deris, I. Windiarti · 0 citations
Review Open access Oct 2026

Handling Class Imbalance in Indonesian Mobile Service App Reviews Using IndoBERT and Hierarchical Data Augmentation

This study investigates sentiment classification on imbalanced Indonesian user-review data from the MyIM3 mobile application, focusing on the effectiveness of hierarchical data augmentation for improving minority-class detection. A total of 10,000 reviews were collected, of which 6,968 were retained after preprocessing...

Sang Ayu Putu Diah Setyawati, Roy Rudolf Huizen, Dadang Hermawan · 0 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.