IJID (International Journal on Informatics for Development)

A Hybrid Approach of Pearson Correlation and PCA in Feature Selection for Opinion Mining

Nova Tri Romadloni, Wakhid Kurniawan, Muhammad Yusuf Ariyadi, Burhan Efendi — Tue, 18 Nov 2025 00:00:00 +0700

This study proposes a hybrid feature selection approach that combines Pearson Correlation and Principal Component Analysis (PCA) to improve classification performance in opinion mining tasks. The rapid growth of e-commerce on social media platforms, such as TikTok, has generated a significant volume of user-generated reviews, which are valuable sources of consumer sentiment. However, the high dimensionality of textual data poses challenges in achieving accurate sentiment classification. To address this issue, the proposed method first applies Pearson Correlation to remove irrelevant features with weak correlation to sentiment labels, followed by PCA to reduce dimensionality. The dataset consists of user reviews from the TikTok Seller platform. Experiments using SVM, Naive Bayes, and Random Forest show that the hybrid approach achieves the highest accuracy of 86.2% (SVM and RF), improving over PCA-only by +0.9% and recovering 13.8% accuracy loss for Naive Bayes (from 72.0% to 83.1%). The results demonstrate that integrating correlation- and projection-based methods yields a more compact and effective feature set. This approach is especially suited for opinion mining in noisy, high-dimensional e-commerce data.

Early Detection of Diabetic Retinopathy Through Explainable AI Models: A Systematic Review

Tinashe Ngwazi, Belinda Ndlovu, Kudakwashe Maguraushe — Sun, 28 Dec 2025 00:00:00 +0700

Diabetes, if not detected early, can lead to serious complications such as vision loss, known as diabetic retinopathy. Explainable Artificial Intelligence (XAI) can enhance traditional Machine Learning methods, which are not understandable and transparent in diagnostic tasks. This Systematic Literature Review explores data inputs that influence the performance of XAI models in detecting diabetic retinopathy, how XAI techniques can enhance early detection outcomes in diabetic retinopathy, the challenges in implementing these techniques and the ethical implications of using these models in clinical practice. The Preferred Reporting Items for Systematic Reviews and Meta-Analyses approach guided the search in 4 databases, Springer, Science Direct, PubMed and IEEE Xplore. The findings reveal that XAI techniques like Local Interpretable Model-agnostic Explanations (LIME), SHapley Additive exPlanations (SHAP) and Gradient-weighted Class Activation Mapping (GRAD-CAM) offer opportunities like early detection outcomes, integration with existing clinical processes, enhancing trust in AI systems, improving accuracy and personalised treatment. XAI can also facilitate collaboration among clinicians, maintaining fairness in AI systems and supporting adherence to ethical standards. However, research on clinical validation of these models, as well as standardised performance evaluation metrics, is lacking.

Twitter (X) Sentiment Analysis on Monkeypox: A Systematic Literature Review

Hazel Chamboko, Kudakwashe Maguraushe, Belinda Ndlovu — Mon, 29 Dec 2025 00:00:00 +0700

Monkeypox has a risk of growing into a global threat. Understanding public sentiments is crucial for effective emergency responses, as it helps counter misinformation, enhance communication, and improve the retention and application of public health information. This systematic review of literature aims to provide foundations for identifying existing algorithms, commonly used data collection methods, and pre-processing techniques applied to Twitter discussions on Mpox. The review followed the PRISMA guidelines. Relevant literature was retrieved from ScienceDirect, IEEE, PubMed, and Springer databases, resulting in 15 studies that met the inclusion criteria. Most preprocessing methods include stop word removal, lemmatisation, and tokenisation; commonly used data collection methods include Twitter API, Academic API V2, Snscrape, Twint, and Tweepy. Classification of sentiment tended to be hybrid models like CNN-LSTM or transformer-based models such as BERT, which also perform well in dealing with complex linguistic patterns. These recent models, additionally, addressed other very important issues like misinformation detection, irony, and bot-generated content, which earlier models would often fail to tackle. Despite these advancements, much work still needs to be done in improving the accuracy, generalizability, and interpretability of sentiment analysis models in live monitoring of public health.

Sentiment Analysis on Shopee Xpress Delivery Time Reviews Using Support Vector Machine and Logistic Regression

Thu, 08 Jan 2026 00:00:00 +0700

This study examines user sentiment towards Shopee Xpress delivery times using machine learning techniques. We collected 497 reviews from platforms like X and the Google Play Store, leveraging the valuable feedback despite its unstructured and informal nature. After labelling 398 reviews for model training and reserving 99 for sentiment prediction, we implemented two classification algorithms: Support Vector Machine (SVM) and Logistic Regression. These models categorised sentiments into negative, neutral, and positive classes. Despite class imbalance in the training data, SVM outperformed Logistic Regression with an accuracy of 93%, demonstrating a more balanced performance across sentiment categories compared to Logistic Regression's 90% accuracy. Both models showed consistent sentiment prediction on new data. Our findings highlight the potential of sentiment analysis as a valuable tool for Shopee Xpress to understand customer perceptions and improve delivery experiences. By providing actionable insights, this study can inform logistics improvements and enhance customer satisfaction. Future research could benefit from collaborating with Shopee to access internal data and integrating additional data sources for more comprehensive insights, ultimately driving business growth and customer loyalty. This study contributes to the growing body of research on sentiment analysis in logistics and e-commerce.

Implementation of Federated Learning for Alzheimer's Disease Classification Using FedAdagrad Algorithm

Arini Arini, Feri Fahrianto, Adil Ramadhan — Fri, 09 Jan 2026 00:00:00 +0700

Federated Learning (FL) offers a promising solution for training machine learning models on decentralized data while preserving privacy, making it particularly valuable for sensitive applications such as healthcare. This study implements FL for the classification of Alzheimer’s disease using MRI images, addressing two critical challenges: data heterogeneity and class imbalance. The research evaluates the performance of the FedAdagrad optimization algorithm against the standard FedAvg approach under varying data distribution scenarios. The methodology employs a CNN trained on a dataset of 6,400 MRI images across four severity classes, partitioned non-IID using Dirichlet distributions (α = 0.1, 0.5, 0.9) to simulate real-world heterogeneity. Experiments were conducted using the Flower framework with four clients over ten communication rounds. Results indicate that FedAdagrad achieves a superior F1-score of 50.33% compared to FedAvg’s 48.14%, though both fall short of centralized CNN performance (55%). High data heterogeneity (α = 0.1) leads to a 13.35% accuracy decline, underscoring FL’s sensitivity to uneven data distributions. Class imbalance emerges as the primary bottleneck, affecting all models. The findings contribute to the growing body of research on adaptive optimization in federated settings, offering insights for future improvements in decentralized healthcare AI.