{"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/news-session-based-recommendations-using-deep","title":"News Session-Based Recommendations using Deep Neural Networks","arxiv_id":"1808.00076","date":"2018-07-31","proceeding":null,"authors":["Gabriel de Souza P. Moreira","Felipe Ferreira","Adilson Marques da Cunha"],"abstract":"News recommender systems are aimed to personalize users experiences and help\nthem to discover relevant articles from a large and dynamic search space.\nTherefore, news domain is a challenging scenario for recommendations, due to\nits sparse user profiling, fast growing number of items, accelerated item's\nvalue decay, and users preferences dynamic shift. Some promising results have\nbeen recently achieved by the usage of Deep Learning techniques on Recommender\nSystems, specially for item's feature extraction and for session-based\nrecommendations with Recurrent Neural Networks. In this paper, it is proposed\nan instantiation of the CHAMELEON -- a Deep Learning Meta-Architecture for News\nRecommender Systems. This architecture is composed of two modules, the first\nresponsible to learn news articles representations, based on their text and\nmetadata, and the second module aimed to provide session-based recommendations\nusing Recurrent Neural Networks. The recommendation task addressed in this work\nis next-item prediction for users sessions: \"what is the next most likely\narticle a user might read in a session?\" Users sessions context is leveraged by\nthe architecture to provide additional information in such extreme cold-start\nscenario of news recommendation. Users' behavior and item features are both\nmerged in an hybrid recommendation approach. A temporal offline evaluation\nmethod is also proposed as a complementary contribution, for a more realistic\nevaluation of such task, considering dynamic factors that affect global\nreadership interests like popularity, recency, and seasonality. Experiments\nwith an extensive number of session-based recommendation methods were performed\nand the proposed instantiation of CHAMELEON meta-architecture obtained a\nsignificant relative improvement in top-n accuracy and ranking metrics (10% on\nHit Rate and 13% on MRR) over the best benchmark methods.","url_abs":"http://arxiv.org/abs/1808.00076v3","url_pdf":"http://arxiv.org/pdf/1808.00076v3.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":"news-session-based-recommendations-using-deep","repo_url":"https://github.com/gabrielspmoreira/chameleon_recsys","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"news-session-based-recommendations-using-deep","repo_url":"https://github.com/13520505/bigdataproj","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"news-session-based-recommendations-using-deep","repo_url":"https://github.com/Curlykonda/chameleon-recsys","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"articles","task_name":"Articles"},{"task_slug":"news-recommendation","task_name":"News Recommendation"},{"task_slug":"recommendation-systems","task_name":"Recommendation Systems"},{"task_slug":"session-based-recommendations","task_name":"Session-Based Recommendations"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1808.00076","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}