DEEP-CLICK-MODEL is a small set of Python scripts for the implementation of different click models.

• RNN-MODEL
• DEEP-UBM-ALL
• DEEP-PSCM-ALL

This file.

• Theano
• Numba
• TensorFlow
• TFlearn
• Numpy
• Scipy

Directory with the scripts.

Directory with the sample dataset.

• train.qids.npy: each line is a query session with the query id and the results' id.
• train.questions.npy: each line is a query session with the query word list.
• train.answers.npy: each line is a query session with the results' title word list.
• train.additions.npy: each line is a query session with the prediction from UBM and PSCM model.
• emb_vectors.skip.1124.4m.10w.npy: each line is the word embedding vector for this word index.

in bin/ : for RNN-MODEL, run python click_model_AAAI14.py; for DEEP-UBM-ALL and DEEP-PSCM-ALL, run python click_model_DEEP.py .