{"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/dis-s2v-discourse-informed-sen2vec","title":"Dis-S2V: Discourse Informed Sen2Vec","arxiv_id":"1610.08078","date":"2016-10-25","proceeding":null,"authors":["Tanay Kumar Saha","Shafiq Joty","Naeemul Hassan","Mohammad Al Hasan"],"abstract":"Vector representation of sentences is important for many text processing\ntasks that involve clustering, classifying, or ranking sentences. Recently,\ndistributed representation of sentences learned by neural models from unlabeled\ndata has been shown to outperform the traditional bag-of-words representation.\nHowever, most of these learning methods consider only the content of a sentence\nand disregard the relations among sentences in a discourse by and large.\n  In this paper, we propose a series of novel models for learning latent\nrepresentations of sentences (Sen2Vec) that consider the content of a sentence\nas well as inter-sentence relations. We first represent the inter-sentence\nrelations with a language network and then use the network to induce contextual\ninformation into the content-based Sen2Vec models. Two different approaches are\nintroduced to exploit the information in the network. Our first approach\nretrofits (already trained) Sen2Vec vectors with respect to the network in two\ndifferent ways: (1) using the adjacency relations of a node, and (2) using a\nstochastic sampling method which is more flexible in sampling neighbors of a\nnode. The second approach uses a regularizer to encode the information in the\nnetwork into the existing Sen2Vec model. Experimental results show that our\nproposed models outperform existing methods in three fundamental information\nsystem tasks demonstrating the effectiveness of our approach. The models\nleverage the computational power of multi-core CPUs to achieve fine-grained\ncomputational efficiency. We make our code publicly available upon acceptance.","url_abs":"http://arxiv.org/abs/1610.08078v1","url_pdf":"http://arxiv.org/pdf/1610.08078v1.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":"dis-s2v-discourse-informed-sen2vec","repo_url":"https://github.com/tksaha/con-s2v","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"tf","reach":null}],"tasks":[{"task_slug":"clustering","task_name":"Clustering"},{"task_slug":"computational-efficiency","task_name":"Computational Efficiency"},{"task_slug":"sentence","task_name":"Sentence"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1610.08078","atlas_url":"https://app.syntology.ai/?focus=1610.08078","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}