Wikipedia Search Engine

The Wikipedia Search Engine takes the given heyword/phrase and finds top matching Wikipedia pages using Cosine Similarity.

Instead of finding similarity with the direct one-hot word representation, we used GloVe Word Embeddings, which maps words into more meaningful space.

In General, finding Cosine Similarity on large dataset will take a huge amount of time on CPU. With the Xilinx Cosine Similarity Acceleration, it will speedup the process by > ~80x.

This use case can be downloaded from Wikipedia Search Engine Jupyter notebook on GitHub.

Follow the setup process below before running the notebook:

• Setup

Run following commands in the Python virtual environment (from setup) to run the Notebook:

(fpga)$ cd cosinesim-examples/python
(fpga)$ ./run.sh jupyter notebook jupyter/wikipedia_demo.ipynb