{"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/stay-on-topic-generating-context-specific","title":"Stay On-Topic: Generating Context-specific Fake Restaurant Reviews","arxiv_id":"1805.02400","date":"2018-05-07","proceeding":null,"authors":["Mika Juuti","Bo Sun","Tatsuya Mori","N. Asokan"],"abstract":"Automatically generated fake restaurant reviews are a threat to online review\nsystems. Recent research has shown that users have difficulties in detecting\nmachine-generated fake reviews hiding among real restaurant reviews. The method\nused in this work (char-LSTM ) has one drawback: it has difficulties staying in\ncontext, i.e. when it generates a review for specific target entity, the\nresulting review may contain phrases that are unrelated to the target, thus\nincreasing its detectability. In this work, we present and evaluate a more\nsophisticated technique based on neural machine translation (NMT) with which we\ncan generate reviews that stay on-topic. We test multiple variants of our\ntechnique using native English speakers on Amazon Mechanical Turk. We\ndemonstrate that reviews generated by the best variant have almost optimal\nundetectability (class-averaged F-score 47%). We conduct a user study with\nskeptical users and show that our method evades detection more frequently\ncompared to the state-of-the-art (average evasion 3.2/4 vs 1.5/4) with\nstatistical significance, at level {\\alpha} = 1% (Section 4.3). We develop very\neffective detection tools and reach average F-score of 97% in classifying\nthese. Although fake reviews are very effective in fooling people, effective\nautomatic detection is still feasible.","url_abs":"http://arxiv.org/abs/1805.02400v4","url_pdf":"http://arxiv.org/pdf/1805.02400v4.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":"stay-on-topic-generating-context-specific","repo_url":"https://github.com/hokkaido/fake-reviews","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"machine-translation","task_name":"Machine Translation"},{"task_slug":"nmt","task_name":"NMT"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}