Mehdi Elahi, Farshad Bakhshandegan Moghaddam, Reza Hosseini, Mohammad Hossein Rimaz, Nabil El Ioini, Marko Tkalcic, Christoph Trattner and Tammam Tillo


Problem: This paper addresses the so-called New Item problem in video Recommender Systems, as part of Cold Start. New item problem occurs when a new item is added to the system catalog, and the recommender system has no or little data describing that item. This could cause the system to fail to meaningfully recommend the new item to the users.
Main Contributions: 
  • we extracted a large dataset of (automatic) visual tags from 7,689 movie trailers, using Deep Learning models, capable of annotating movies with a wide range of automatic tags including celebrity tags (e.g., #TomHanks & #BradPitt), object tags (e.g., #sky & #children), and face tags (e.g., #happy & #withGlass); our dataset (called MVT9KD) is public and freely available on Figshare (see below)
  • we addressed the cold start problem by proposing a novel set of content features, extracted automatically, with no need for any human involvement and used for recommending new items with no rating and no tags;
  • we evaluated the proposed recommendation approach using a large dataset of thousands of movie trailers and compared our results with different baselines, including recommendation based on low-level visual features (e.g., colorfulness, and naturalness in movies);
  • our results have shown the superiority of recommendation based on our high-level visual tags, used all together or individually, in comparison to the manual tags and low-level visual features.
Dataset: MVT9KD Movie Visual Tags 9K Dataset Download Link
Evaluation & Results: In order to evaluate our proposed technique, we have performed a set of experiments using a large dataset of videos. The results have shown that the automatically extracted visual tags can be incorporated into the cold start recommendation process and achieve superior results compared to the recommendation based on human-annotated tags.
Image From UMAP2021 Elahi 2