Enhancing online video recommendation using social user interactions

Cross-Domain Recommendation for Cold-Start Users via Neighborhood Based Feature Mapping

“…On social networks [23], the cold start problem has been widely studied. Zhou [24] works on the cold-start batch video recommendation in shared community. Zhao [25] aims to recommend products from e-commerce websites to users at social networks in cold-start situations and they map users' social networking features to another feature representation for product recommendation.…”

[Paper] Personalized Recommendation of Tumblr Posts Using Graph Convolutional Networks with Preference-aware Multimodal Features

“…[5,[22][23][24][25]) have reported that the combined use of content features and network structure improved the performance of Web content analysis. For a typical task in this domain, i.e., recommendation, the effectiveness of using both content features and the network structure has been confirmed [5,23,24]. For example, Ying et al [23] proposed a method for recommending Pinterest * 6 contents by applying GCN to a network in which image features are embedded.…”

“…Matsumoto et al [5] proposed a method for recommending music videos on the basis of link prediction on a network that represented relationships between users and contents via sub-sampled CCA [25] based latent feature extraction. Inspired by these studies [22][23][24][25], we construct a network that represents relationships between users and posts with consideration of each user's preferences based on SMVAE-UP. Then we realize recommendation of each user's preferred posts by applying GCN [9] to the constructed network.…”

The 1st International Workshop on Context-Aware Recommendation Systems with Big Data Analytics (CARS-BDA)

“…It will align with the emphasis of WSDM on practical yet principled novel models of search and data mining, algorithm design and analysis, and relevant to the topics of WSDM, including multimodal data mining, web recommender systems and algorithms, social search, mining and other applications, social network dynamics, locationbased social networks, social network analysis, theories, models and applications. e areas of interest mainly include contexts and user behaviors in recommendation, such as behavioural categories [1], article popularity [3,12], social trust [4], user activities and interactions [2,[5][6][7][8][9][10][11], and the scalability of recommender systems such as the Apache Storm-based parallel processing [2,9,13] etc. In particular, topics of interest for this workshop include (but are not limited to):…”