One of the factors that improve businesses in business intelligence is summarization systems that can generate summaries based on sentiment from social media. However, these systems cannot produce such summaries automatically; they use annotated datasets. To support these systems with annotated datasets, we propose a novel framework that uses pattern rules. The framework has two procedures: 1) pre-processing, and 2) aspect knowledge-base generation. The first procedure is to check and correct any misspelled words (bigram and unigram) by a proposed method and tag the parts-of-speech of all of the words. The second procedure is to automatically generate an aspect knowledge base that is to be used to produce sentiment summaries by sentiment-summarization systems. Pattern rules and semantic similarity-based pruning are used to automatically generate an aspect knowledge base from social media. In the experiments, eight domains from benchmark datasets of reviews are used. The performance evaluation of our proposed approach shows the highest performance when compared to other unsupervised approaches.
JavaScript jest wyłączony w Twojej przeglądarce internetowej. Włącz go, a następnie odśwież stronę, aby móc w pełni z niej korzystać.