{"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/dialogue-act-recognition-via-crf-attentive","title":"Dialogue Act Recognition via CRF-Attentive Structured Network","arxiv_id":"1711.05568","date":"2017-11-15","proceeding":"SIGIR 2018 7","authors":["Zheqian Chen","Rongqin Yang","Zhou Zhao","Deng Cai","Xiaofei He"],"abstract":"Dialogue Act Recognition (DAR) is a challenging problem in dialogue\ninterpretation, which aims to attach semantic labels to utterances and\ncharacterize the speaker's intention. Currently, many existing approaches\nformulate the DAR problem ranging from multi-classification to structured\nprediction, which suffer from handcrafted feature extensions and attentive\ncontextual structural dependencies. In this paper, we consider the problem of\nDAR from the viewpoint of extending richer Conditional Random Field (CRF)\nstructural dependencies without abandoning end-to-end training. We incorporate\nhierarchical semantic inference with memory mechanism on the utterance\nmodeling. We then extend structured attention network to the linear-chain\nconditional random field layer which takes into account both contextual\nutterances and corresponding dialogue acts. The extensive experiments on two\nmajor benchmark datasets Switchboard Dialogue Act (SWDA) and Meeting Recorder\nDialogue Act (MRDA) datasets show that our method achieves better performance\nthan other state-of-the-art solutions to the problem. It is a remarkable fact\nthat our method is nearly close to the human annotator's performance on SWDA\nwithin 2% gap.","url_abs":"http://arxiv.org/abs/1711.05568v1","url_pdf":"http://arxiv.org/pdf/1711.05568v1.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":[],"tasks":[{"task_slug":"dialogue-act-classification","task_name":"Dialogue Act Classification"},{"task_slug":"dialogue-interpretation","task_name":"Dialogue Interpretation"},{"task_slug":"structured-prediction","task_name":"Structured Prediction"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/dialogue-act-classification-on-icsi-meeting","task":"Dialogue Act Classification","dataset":"ICSI Meeting Recorder Dialog Act (MRDA) corpus","model":"CRF-ASN","rank_in_archive_order":3,"of":8,"metrics":{"Accuracy":"91.7"},"uses_additional_data":false},{"leaderboard":"/sota/dialogue-act-classification-on-switchboard","task":"Dialogue Act Classification","dataset":"Switchboard corpus","model":"CRF-ASN","rank_in_archive_order":6,"of":11,"metrics":{"Accuracy":"81.3"},"uses_additional_data":false}],"syntology":{"syntology_url":null,"atlas_url":"https://app.syntology.ai/?focus=1711.05568","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}