This dataset is a list of important keyphrases for academic topics. The keyphrases file contains around 150,000 keyphrases that are given about 300-400 labels, such as algorithm, disease, theorem, lemma, chemial compound, research methods, fields, subfields, topics, ect.
We also have about 2 million additional keyphrases that have been sorted into unigrams, bigrams, trigrams and fourgrams.
The keyphrases were obtain fromed a mixture of webscraping academic databases such as pubmed, wikipedia.… See the full description on the dataset page:
https://huggingface.co/datasets/ClovenDoug/150k_keyphrases_labelled.
