A Novel Approach to Multi-Document Summarization with Sentiment Analysis

This research aims to investigate various approaches and algorithms used in multi-document summarization (MDS), including early methods and deep learning techniques, as well as the associated evaluation metrics and challenges. To address the limitations of existing methods, a hybrid semantic-based MDS model is proposed, which combines the strengths of pre-trained models and deep neural networks. The model will be trained and evaluated on a dataset annotated with sentiment information. Performance assessment will include metrics such as ROUGE, BLEU, and METEOR, and the model will be compared with state-of-the-art MDS models. The project aims to improve the ability to extract relevant information with sentiment from large volumes of text data by introducing a new dataset and addressing the challenges in MDS. The outcomes of this research will contribute to the advancement of MDS and have practical applications in areas such as news summarization, document analysis, and information retrieval.