Sequential Pattern Mining and Hybrid Sentiment-based Collaborative Architecture for Rating Prediction

This research presents a novel approach named "Sequential Pattern Mining and Hybrid Sentiment-based Collaborative Architecture for Rating Predictions". This approach overcomes the limitations of traditional techniques by considering multidimensional data, including users' past behaviour, buying patterns, and sentiments to enhance the rating predictions and recommendations. The proposed prediction approach incorporates users' past behaviour i.e. ratings through the Collaborative Filtering technique. The users' sentiments are included by implementing Hybrid Sentiment Analysis and the sequential buying patterns are considered through the Generalized Sequential Pattern Mining technique. The Hybrid Sentiment Analysis technique combines Lexicon-based and Deep Learning-based Sentiment Analysis methodologies for more comprehensive sentiment evaluation. The proposed Hybrid Rating Prediction System is evaluated using a standardized public dataset and standard evaluation metrics including Accuracy, Precision, Recall, and F1-Score. Therefore, the research study has three primary objectives. The first objective is to identify the existing recommendation techniques through a literature review. The second objective is to propose a hybrid approach that mitigates the limitations of traditional systems by incorporating multi-dimensional information about the user and items. The third objective is to evaluate, validate, and compare the proposed approach against existing state-of-the-art systems and possible hybrid systems. The results demonstrate that the proposed hybrid approach achieves an Accuracy of 79.79%, with significant improvements in Precision and F1-Score compared to existing systems.