Aspect-based Sentiment Analysis (ABSA) using Machine Learning Algorithms

Aspect-Based Sentiment Analysis (ABSA) is a Natural Language Processing task that aims to identify and extract the sentiment of specific aspects or components of a product or service. ABSA typically involves a multi-step process that begins with identifying the aspects or features of the product or service that are being discussed in the text. This is followed by sentiment analysis, where the sentiment polarity (positive, negative, or neutral) is assigned to each aspect based on the context of the sentence or document. Finally, the results are aggregated to provide an overall sentiment for each aspect. The process involves training machine learning models to classify text sentiment (positive, negative, or neutral). First, we transform text data using Term Frequency-Inverse Document Frequency (TF-IDF), which assigns weights to words based on their importance within a document collection. This emphasizes informative terms. Then, these TF-IDF features are fed into both SVM and Logistic Regression models. SVM find a hyper plane that best separates sentiment classes, while Logistic Regression calculates the probability of a text belonging to a specific sentiment class. Extensive experiments have been conducted on datasets of covid vaccinations dataset and results show that the support vector machine model achieves excellent performance in terms of aspect extraction and sentiment classification. Sentiment on Twitter can be imbalanced, with more positive or negative tweets depending on the topic. This can affect the training process. Techniques like oversampling or undersampling the minority class might be necessary. This work investigates the performance of machine learning algorithms for a specific classification task. Support Vector Machine (SVM) and Logistic Regression (LR) were compared. The results indicate that Support Vector Machine achieved superior accuracy (87.34%) compared to Logistic Regression (84.64%), suggesting Support Vector Machine as a more suitable option for this classification task.