{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/aspect-based-relational-sentiment-analysis","title":"Aspect-Based Relational Sentiment Analysis Using a Stacked Neural Network Architecture","arxiv_id":"1709.06309","date":"2017-09-19","proceeding":null,"authors":["Soufian Jebbara","Philipp Cimiano"],"abstract":"Sentiment analysis can be regarded as a relation extraction problem in which\nthe sentiment of some opinion holder towards a certain aspect of a product,\ntheme or event needs to be extracted. We present a novel neural architecture\nfor sentiment analysis as a relation extraction problem that addresses this\nproblem by dividing it into three subtasks: i) identification of aspect and\nopinion terms, ii) labeling of opinion terms with a sentiment, and iii)\nextraction of relations between opinion terms and aspect terms. For each\nsubtask, we propose a neural network based component and combine all of them\ninto a complete system for relational sentiment analysis. The component for\naspect and opinion term extraction is a hybrid architecture consisting of a\nrecurrent neural network stacked on top of a convolutional neural network. This\napproach outperforms a standard convolutional deep neural architecture as well\nas a recurrent network architecture and performs competitively compared to\nother methods on two datasets of annotated customer reviews. To extract\nsentiments for individual opinion terms, we propose a recurrent architecture in\ncombination with word distance features and achieve promising results,\noutperforming a majority baseline by 18% accuracy and providing the first\nresults for the USAGE dataset. Our relation extraction component outperforms\nthe current state-of-the-art in aspect-opinion relation extraction by 15%\nF-Measure.","url_abs":"http://arxiv.org/abs/1709.06309v1","url_pdf":"http://arxiv.org/pdf/1709.06309v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"aspect-based-relational-sentiment-analysis","repo_url":"https://github.com/santhoshmani888/Aspect-Based-sentiment-analysis","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":null,"task_name":"Relation"},{"task_slug":"relation-extraction","task_name":"Relation Extraction"},{"task_slug":"sentiment-analysis","task_name":"Sentiment Analysis"},{"task_slug":"term-extraction","task_name":"Term Extraction"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}