Factorization Machines - 國立臺灣大學

← Back to document page

Factorization MachinesSteffen RendleDepartment of Reasoning for IntelligenceThe Institute of Scientific and Industrial ResearchOsaka University, In this paper, we introduce Factorization Machines (FM) which are a new model class that combines the advantagesof Support Vector Machines (SVM) with Factorization SVMs, FMs are a general predictor working with anyreal valued feature vector. In contrast to SVMs, FMs model allinteractions between variables using factorized parameters. Thusthey are able to estimate interactions even in problems withhugesparsity (like recommender systems) where SVMs fail.

Alicehas rated Titanic, NottingHill and Star Wars. Additionally the example contains a variable (green) holding the time in months starting from January, 2009. And finally the vector contains information of the last movie (brown) the user has rated before (s)he rated the active one – e.g. for x(2), Alicerated Titanicbefore she rated Notting ...

  Titanic