Yifan Hu is a Senior Director of Research. He leads the Research group in the Data and Common Services Organization. His research interests are machine learning, natural language processing, recommender systems, data visualization and security. In machine learning, he is interested in user profiling and proposed the idea of name embedding and co-authored a number of papers in applying this technique to to demographic classification and in security. He is also interested in graph embedding and worked on improving the DeepWalk optimization process via the multilevel approach. In natural language processing, he is interested in hate speech detection, and word embedding with application to advertising. In recommender systems, he worked on recommendation for TV/video/Tumblr, and is a co-author of the 2017 IEEE ICDM 10-Year Highest-Impact Paper on recommendation for implicit feedback datasets. In visualization, he is generally interested in very large graph embedding, and authored a number of codes for large scale graph visualization (e.g., sfdp) for Graphviz. He co-authored a number of best papers, including the EuroVis 2016 best paper on stream graph.
Yifan received his BSc/MSc in Applied Mathematics from Shanghai Jiao-Tong University, and a Ph.D. in Optimization from Loughborough University. He worked at Daresbury Laboratory and Wolfram Research and AT&T Labs. He joined Yahoo! Research in 2014.