DECKER, consists of three major modules: (i) Knowledge Retrieval Module which retrieves heterogeneous knowledge based on the input question; (ii) Double Check Module which merges information from structured and unstructured knowledge and makes a double check between them; (iii) Knowledge Fusion Module which combines heterogeneous knowledge together to obtain a final representation.
Install all required python dependencies:
pip install -r requirements.txt
Download the required knowledge bases:
sh download_rawdata_resource.shDownload the datasets from the following repository and put them under data/creak and data/csqa2:
https://github.com/yasumasaonoe/creak/tree/main/data/creak
https://github.com/allenai/csqa2/tree/master/dataset
Data preprocessing consists of three stages: (i) Ground concepts; (ii) Retrieve facts; (iii) Process and load graph information.
cd data_preprocess
sh data_preprocess.sh
sh confact_train.sh