{"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/multiple-instance-learning-networks-for-fine","title":"Multiple Instance Learning Networks for Fine-Grained Sentiment Analysis","arxiv_id":"1711.09645","date":"2017-11-27","proceeding":"TACL 2018 1","authors":["Stefanos Angelidis","Mirella Lapata"],"abstract":"We consider the task of fine-grained sentiment analysis from the perspective\nof multiple instance learning (MIL). Our neural model is trained on document\nsentiment labels, and learns to predict the sentiment of text segments, i.e.\nsentences or elementary discourse units (EDUs), without segment-level\nsupervision. We introduce an attention-based polarity scoring method for\nidentifying positive and negative text snippets and a new dataset which we call\nSPOT (as shorthand for Segment-level POlariTy annotations) for evaluating\nMIL-style sentiment models like ours. Experimental results demonstrate superior\nperformance against multiple baselines, whereas a judgement elicitation study\nshows that EDU-level opinion extraction produces more informative summaries\nthan sentence-based alternatives.","url_abs":"http://arxiv.org/abs/1711.09645v2","url_pdf":"http://arxiv.org/pdf/1711.09645v2.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":"multiple-instance-learning-networks-for-fine","repo_url":"https://github.com/stangelid/milnet-sent","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":{"status":"unanswered"}},{"paper_slug":"multiple-instance-learning-networks-for-fine","repo_url":"https://github.com/nlpat/MEGA-DT","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"multiple-instance-learning","task_name":"Multiple Instance Learning"},{"task_slug":"sentence","task_name":"Sentence"},{"task_slug":"sentiment-analysis","task_name":"Sentiment Analysis"}],"methods":[],"datasets_introduced":[{"slug":"spot","name":"SPOT","full_name":"Sentiment Polarity Annotations Dataset"}],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1711.09645","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}