Predicting ad click-through rates via feature-based fully coupled interaction tensor factorization

作者：

Highlights：

• We address CTR prediction by a novel tensor factorization model named FCTF.

• We incorporate all types of information into FCTF model to relieve data sparsity.

• We give the relationship between FCTF and PITF, Tucker, CD decomposition.

• We evaluate our model and algorithm on real-world bidding log data.

摘要

•We address CTR prediction by a novel tensor factorization model named FCTF.•We incorporate all types of information into FCTF model to relieve data sparsity.•We give the relationship between FCTF and PITF, Tucker, CD decomposition.•We evaluate our model and algorithm on real-world bidding log data.

论文关键词：Click through rate prediction,Tensor factorization,Feature-based,Fully coupled interaction,Real-time bidding,Demand-side platform

论文评审过程：Received 24 May 2015, Revised 23 January 2016, Accepted 23 January 2016, Available online 3 February 2016, Version of Record 8 February 2016.

论文官网地址：https://doi.org/10.1016/j.elerap.2016.01.004