this is an NLP course project at l3s
it recommends semantic similar queries
the semantic similarity is calculated by the word2vec
it was trained by wikicorpus in 2021 on the cluster system at uni-hannover
(the training process needs also to install environment in sigulairty)
result was evaluated with WordSim353, it outperforms the pre-trained models, and models from literatures in 2016
trained word2vec model could be found here:
https://drive.google.com/drive/folders/1Rg2j1oE0uwc_pZU_v1jDCH4zpQMWQ_eX?usp=share_link