Coreference resolution is one of the most critical issues in various applications of natural language processing, such as machine translation, sentiment analysis, summarization, etc. In the process of coreference resolution, in this paper, a fully connected neural network approach has been adopted to enhance the performance of feature extraction whilst also facilitating the mention pair classification process for coreference resolution in the Persian language. For this purpose, first, we focus on the feature extraction phase by fusing some handcrafted features, word embedding features and semantic features. Then, a fully connected deep neural network is utilized to determine the probability of the validity of the mention pairs. After that, the numeric output of the last layer of the utilized neural network is considered as the feature vector of the valid mention pairs. Finally, the coreference mention pairs are specified by utilizing a hierarchical accumulative clustering method. The proposed method's evaluation on the Uppsala dataset demonstrates a meaningful improvement, as indicated by the F-score 64.54%, in comparison to state-of-the-art methods.
Introduction: With the widespread dissemination of user-generated content on different web sites, social networks, and online consumer systems such as Amazon, the quantity of opinionated information available on the Internet has been increased. Sentiment analysis of user-generated content is one of the main cognitive computing branches; hence, it has attracted the attention of many scholars in recent years. One of the main tasks of the sentiment analysis is to detect polarity within a text. The existing polarity detection methods mainly focus on keywords and their naïve frequency counts; however, they less regard the meanings and implicit dimensions of the natural concepts. Although background knowledge plays a critical role in determining the polarity of concepts, it has been disregarded in polarity detection methods.Method: This study presents a context-based model to solve ambiguous polarity concepts using commonsense knowledge. First, a model is presented to generate a source of ambiguous sentiment concepts based on SenticNet by computing the probability distribution. Then the model uses a bag-of-concepts approach to remove ambiguities and semantic augmentation with the ConceptNet handling to overcome lost knowledge. ConceptNet is a large-scale semantic network with a large number of commonsense concepts. In this paper, the point mutual information (PMI) measure is used to select the contextual concepts having strong relationships with ambiguous concepts. The polarity of the ambiguous concepts is precisely detected using positive/negative contextual concepts and the relationship of the concepts in the semantic knowledge base. The text representation scheme is semantically enriched using Numberbatch, which is a word embedding model based on the concepts from the ConceptNet semantic network. In this regard, the cosine similarity metric is used to measure similarity and select a concept from the ConceptNet network for semantic augmentation. Pre-trained concepts vectors facilitate the more effective computation of semantic similarity among the concerned concepts.Result: The proposed model is evaluated by applying a corpus of product reviews, called Semeval. The experimental results revealed an accuracy rate of 82.07%, representing the effectiveness of the proposed model.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.