{"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/an-attentive-neural-architecture-for-joint","title":"An attentive neural architecture for joint segmentation and parsing and its application to real estate ads","arxiv_id":"1709.09590","date":"2017-09-27","proceeding":null,"authors":["Giannis Bekoulis","Johannes Deleu","Thomas Demeester","Chris Develder"],"abstract":"In processing human produced text using natural language processing (NLP)\ntechniques, two fundamental subtasks that arise are (i) segmentation of the\nplain text into meaningful subunits (e.g., entities), and (ii) dependency\nparsing, to establish relations between subunits. In this paper, we develop a\nrelatively simple and effective neural joint model that performs both\nsegmentation and dependency parsing together, instead of one after the other as\nin most state-of-the-art works. We will focus in particular on the real estate\nad setting, aiming to convert an ad to a structured description, which we name\nproperty tree, comprising the tasks of (1) identifying important entities of a\nproperty (e.g., rooms) from classifieds and (2) structuring them into a tree\nformat. In this work, we propose a new joint model that is able to tackle the\ntwo tasks simultaneously and construct the property tree by (i) avoiding the\nerror propagation that would arise from the subtasks one after the other in a\npipelined fashion, and (ii) exploiting the interactions between the subtasks.\nFor this purpose, we perform an extensive comparative study of the pipeline\nmethods and the new proposed joint model, reporting an improvement of over\nthree percentage points in the overall edge F1 score of the property tree.\nAlso, we propose attention methods, to encourage our model to focus on salient\ntokens during the construction of the property tree. Thus we experimentally\ndemonstrate the usefulness of attentive neural architectures for the proposed\njoint model, showcasing a further improvement of two percentage points in edge\nF1 score for our application.","url_abs":"http://arxiv.org/abs/1709.09590v2","url_pdf":"http://arxiv.org/pdf/1709.09590v2.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":"an-attentive-neural-architecture-for-joint","repo_url":"https://github.com/bekou/ad_data","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":null}],"tasks":[{"task_slug":"dependency-parsing","task_name":"Dependency Parsing"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}