eprintid: 10126932 rev_number: 12 eprint_status: archive userid: 608 dir: disk0/10/12/69/32 datestamp: 2021-04-30 14:29:18 lastmod: 2021-04-30 14:29:18 status_changed: 2021-04-30 14:29:18 type: proceedings_section metadata_visibility: show creators_name: Nazir, S creators_name: Cagali, T creators_name: Sadrzadeh, M creators_name: Newell, C title: Audiovisual, Genre, Neural and Topical Textual Embeddings for TV Programme Content Representation ispublished: pub divisions: UCL divisions: A01 divisions: B04 divisions: C05 divisions: F48 keywords: Multimedia systems; Information filtering; Recommender systems; Content-based retrieval note: This version is the author accepted manuscript. For information on re-use, please refer to the publisher’s terms and conditions. abstract: TV programmes have their contents described by multiple means: textual subtitles, audiovisual files, and metadata such as genres. In order to represent these contents, we develop vectorial representations for their low-level multimodal features, group them with simple clustering techniques, and combine them using middle and late fusion. For textual features, we use LSI and Doc2Vec neural embeddings; for audio, MFCC's and Bags of Audio Words; for visual, SIFT, and Bags of Visual Words. We apply our model to a dataset of BBC TV programmes and use a standard recommender and pairwise similarity matrices of content vectors to estimate viewers' behaviours. The late fusion of genre, audio and video vectors with both of the textual embeddings significantly increase the precision and diversity of the results. date: 2021-01-22 date_type: published publisher: IEEE official_url: https://doi.org/10.1109/ISM.2020.00041 oa_status: green full_text_type: other language: eng primo: open primo_central: open_green verified: verified_manual elements_id: 1854549 doi: 10.1109/ISM.2020.00041 isbn_13: 9781728186979 lyricists_name: Sadrzadeh, Mehrnoosh lyricists_id: MSADR73 actors_name: Sadrzadeh, Mehrnoosh actors_id: MSADR73 actors_role: owner full_text_status: public publication: Proceedings - 2020 IEEE International Symposium on Multimedia, ISM 2020 place_of_pub: Naples, Italy pagerange: 197-200 event_title: 2020 IEEE International Symposium on Multimedia (ISM) book_title: 020 IEEE International Symposium on Multimedia (ISM) citation: Nazir, S; Cagali, T; Sadrzadeh, M; Newell, C; (2021) Audiovisual, Genre, Neural and Topical Textual Embeddings for TV Programme Content Representation. In: 020 IEEE International Symposium on Multimedia (ISM). (pp. pp. 197-200). IEEE: Naples, Italy. Green open access document_url: https://discovery.ucl.ac.uk/id/eprint/10126932/1/SabaNazir.pdf