Make sure you install the environment (instructions in the parent folder). Then, go to the directory after the dataset name you want to use and execute the run.sh script.

$ cd kptimes
$ bash run.sh

After the preprocessing script finishes, the will be three folders.

• processed: One json file containing a entry with key id, title, abstract, present_kps, and absent_kps for each document in the training, validation, and test dataset.
• fairseq: x.source and and x.target files for x in train, valid, and test. Mainly used for validation and finetuning with fairseq. x.source contains the inputs consisting of title and abstract, concatenated with a [sep] token. x.target contains the target keyphrases concatenated with ;.
• json: same content as fairseq in json format. Used for fine-tuning the sequence generation models.
• If you wish to run sequence labeling for keyphrase extraction, please follow the data preprocessing procedure in the sequence_tagging folder. After preprocessing, there will be a bioformat folder containing the data required for sequence tagging.

• Paper: https://www.aclweb.org/anthology/P17-1054/
• Download data from: https://drive.google.com/open?id=1DbXV1mZXm_o9bgfwPV9PV0ZPcNo1cnLp

• Paper: https://arxiv.org/abs/2211.12124
• Download data from: https://huggingface.co/datasets/taln-ls2n/kpbiomed

• Paper: https://www.aclweb.org/anthology/W19-8617/
• Download data from: https://github.com/ygorg/KPTimes

• Paper: https://www.aclweb.org/anthology/2020.acl-main.710/
• Download data from: https://github.com/memray/OpenNMT-kpg-release

• Paper: https://www.aclweb.org/anthology/D19-1521/
• Download data from: https://github.com/microsoft/OpenKP#download-the-dataset

• Paper: https://www.aclweb.org/anthology/N19-1070/
• Download data from: http:https://hdl.handle.net/11234/1-2943

• Paper: https://arxiv.org/pdf/2203.15349.pdf
• Download data from: https://huggingface.co/datasets/midas/ldkp3k and https://huggingface.co/datasets/midas/ldkp10k

• Paper: https://arxiv.org/abs/2006.05324
• Download data from: https://microsoft.github.io/TREC-2020-Deep-Learning/

• Paper: https://aclanthology.org/N18-2097/
• Download data from: https://github.com/armancohan/long-summarization