{"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/on-the-benefit-of-combining-neural","title":"On the Benefit of Combining Neural, Statistical and External Features for Fake News Identification","arxiv_id":"1712.03935","date":"2017-12-11","proceeding":null,"authors":["Gaurav Bhatt","Aman Sharma","Shivam Sharma","Ankush Nagpal","Balasubramanian Raman","Ankush Mittal"],"abstract":"Identifying the veracity of a news article is an interesting problem while\nautomating this process can be a challenging task. Detection of a news article\nas fake is still an open question as it is contingent on many factors which the\ncurrent state-of-the-art models fail to incorporate. In this paper, we explore\na subtask to fake news identification, and that is stance detection. Given a\nnews article, the task is to determine the relevance of the body and its claim.\nWe present a novel idea that combines the neural, statistical and external\nfeatures to provide an efficient solution to this problem. We compute the\nneural embedding from the deep recurrent model, statistical features from the\nweighted n-gram bag-of-words model and handcrafted external features with the\nhelp of feature engineering heuristics. Finally, using deep neural layer all\nthe features are combined, thereby classifying the headline-body news pair as\nagree, disagree, discuss, or unrelated. We compare our proposed technique with\nthe current state-of-the-art models on the fake news challenge dataset. Through\nextensive experiments, we find that the proposed model outperforms all the\nstate-of-the-art techniques including the submissions to the fake news\nchallenge.","url_abs":"http://arxiv.org/abs/1712.03935v1","url_pdf":"http://arxiv.org/pdf/1712.03935v1.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":"on-the-benefit-of-combining-neural","repo_url":"https://github.com/vineet2104/StanceDetection-CS626","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"fake-news-detection","task_name":"Fake News Detection"},{"task_slug":"feature-engineering","task_name":"Feature Engineering"},{"task_slug":"open-question","task_name":"Open-Ended Question Answering"},{"task_slug":"stance-detection","task_name":"Stance Detection"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/fake-news-detection-on-fnc-1","task":"Fake News Detection","dataset":"FNC-1","model":"Bhatt et al.","rank_in_archive_order":3,"of":10,"metrics":{"Per-class Accuracy (Agree)":"43.82","Per-class Accuracy (Disagree)":"6.31","Per-class Accuracy (Discuss)":"85.68","Per-class Accuracy (Unrelated)":"98.04","Weighted Accuracy":"83.08"},"uses_additional_data":false},{"leaderboard":"/sota/fake-news-detection-on-fnc-1","task":"Fake News Detection","dataset":"FNC-1","model":"Baseline based on skip-thought embeddings (Bhatt et al., 2017)","rank_in_archive_order":8,"of":10,"metrics":{"Per-class Accuracy (Agree)":"31.80","Per-class Accuracy (Disagree)":"0.00","Per-class Accuracy (Discuss)":"81.20","Per-class Accuracy (Unrelated)":"91.18","Weighted Accuracy":"76.18"},"uses_additional_data":false},{"leaderboard":"/sota/fake-news-detection-on-fnc-1","task":"Fake News Detection","dataset":"FNC-1","model":"Baseline based on word2vec + hand-crafted features (Bhatt et al., 2017)","rank_in_archive_order":9,"of":10,"metrics":{"Per-class Accuracy (Agree)":"50.70","Per-class Accuracy (Disagree)":"9.61","Per-class Accuracy (Discuss)":"53.38","Per-class Accuracy (Unrelated)":"96.05","Weighted Accuracy":"72.78"},"uses_additional_data":false},{"leaderboard":"/sota/fake-news-detection-on-fnc-1","task":"Fake News Detection","dataset":"FNC-1","model":"Neural baseline based on bi-directional LSTMs (Bhatt et al., 2017)","rank_in_archive_order":10,"of":10,"metrics":{"Per-class Accuracy (Agree)":"38.04","Per-class Accuracy (Disagree)":"4.59","Per-class Accuracy (Discuss)":"58.132","Per-class Accuracy (Unrelated)":"78.27","Weighted Accuracy":"63.11"},"uses_additional_data":true}],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}