This exercise demonstrates a movie Recommendation System based on K Nearest Neighbor algorithm. The used dataset is a subset of MovieLens. Students are going to use High Level Synthesis in order to accelerate the recommendation system using the SDSoC 2016.4 development environment. Finally, they will execute the accelerated application on a Zynq-7000 ARM/FPGA SoC development board.
- CPU directory contains a CPU only version for the movie recommendation system. Students can experiment with different distance metrics on the full dataset.
- FPGA directory contains the main code of the exercise.
- various directory contains the Jupyter notebook that was used in order to understand the input data and create the used dataset.
The MovieLens dataset subset can be found on: https://drive.google.com/drive/folders/1m_kzCO8PBifs6wIZnb-vLuubm76F1B8f?usp=sharing