Here are some projects I’ve been involved:
PhD: Probabilistic models for recommender systems: scalable inference and new designs, 2015-2019.
✕ Exploration of different designs for probabilistic models and approximate inference algorithms for recommender systems. Extending models to include complex side information source and social networks.
✕ Topics: Variational Inference, MCMC, INLA, Probabilistic Matrix Factorization, Poisson Matrix Factorization, Non-parametric statistical models, Collaborative Topic Regression, Dynamic probabilistic models.
– https://github.com/zehsilva/poissonmf_cs (Content-Based Social Recommendation with Poisson Matrix Factorization, ECML 2017)
EspiaCamara, 2013. (with Alan Godoy and Saullo Oliveira)
✕A project developed during the Brazillian Chamber of Deputies OpenData Hackathon. We gathered public data of deputies and made semantic textual analysis (automatic topic detection) and co-clustering of the voting pattern in order the find block votings.
✕ Code: https://bitbucket.org/djosacv/camara-mining
✕ Press Release: TV Camara documentary about the Hackathon and with interviews with our team.
✕Renderer of 3D images using raytracing technique. Some feature included are Phong Shader and Depth-Of-Field effect.
ZEL Language and ZehCompiler, 2010.
✕Simple programming language design and a compiler (targeted to JVM bytecodes) written in Java.