- text: "In this work, we explore how to learn task specific language models aimed towards learning rich representation of keyphrases from text documents. We experiment with different masking strategies for pre-training transformer language models (LMs) in discriminative as well as generative settings. In the discriminative setting, we introduce a new pre-training objective - Keyphrase Boundary Infilling with Replacement (KBIR), showing large gains in performance (up to 9.26 points in F1) over SOTA, when LM pre-trained using KBIR is fine-tuned for the task of keyphrase extraction. In the generative setting, we introduce a new pre-training setup for BART - KeyBART, that reproduces the keyphrases related to the input text in the CatSeq format, instead of the denoised original input. This also led to gains in performance (up to 4.33 points inF1@M) over SOTA for keyphrase generation. Additionally, we also fine-tune the pre-trained language models on named entity recognition(NER), question answering (QA), relation extraction (RE), abstractive summarization and achieve comparable performance with that of the SOTA, showing that learning rich representation of keyphrases is indeed beneficial for many other fundamental NLP tasks."
Keyphrase extraction is a technique in text analysis where you extract the important keyphrases from a document. Thanks to these keyphrases humans can understand the content of a text very quickly and easily without reading it completely. Keyphrase extraction was first done primarily by human annotators, who read the text in detail and then wrote down the most important keyphrases. The disadvantage is that if you work with a lot of documents, this process can take a lot of time ⏳.
Here is where Artificial Intelligence 🤖 comes in. Currently, classical machine learning methods, that use statistical and linguistic features, are widely used for the extraction process. Now with deep learning, it is possible to capture the semantic meaning of a text even better than these classical methods. Classical methods look at the frequency, occurrence and order of words in the text, whereas these neural approaches can capture long-term semantic dependencies and context of words in a text.
This model uses [distilbert](https://huggingface.co/distilbert-base-uncased) as its base model and fine-tunes it on the [Inspec dataset](https://huggingface.co/datasets/midas/inspec).
Keyphrase extraction models are transformer models fine-tuned as a token classification problem where each word in the document is classified as being part of a keyphrase or not.
Kulkarni, Mayank, Debanjan Mahata, Ravneet Arora, and Rajarshi Bhowmik. "Learning Rich Representation of Keyphrases from Text." arXiv preprint arXiv:2112.08547 (2021).
Sahrawat, Dhruva, Debanjan Mahata, Haimin Zhang, Mayank Kulkarni, Agniv Sharma, Rakesh Gosangi, Amanda Stent, Yaman Kumar, Rajiv Ratn Shah, and Roger Zimmermann. "Keyphrase extraction as sequence labeling using contextualized embeddings." In European Conference on Information Retrieval, pp. 328-335. Springer, Cham, 2020.
* This keyphrase extraction model is very domain-specific and will perform very well on abstracts of scientific papers. It's not recommended to use this model for other domains, but you are free to test it out.
[Inspec](https://huggingface.co/datasets/midas/inspec) is a keyphrase extraction/generation dataset consisting of 2000 English scientific papers from the scientific domains of Computers and Control and Information Technology published between 1998 to 2002. The keyphrases are annotated by professional indexers or editors.
The documents in the dataset are already preprocessed into list of words with the corresponding labels. The only thing that must be done is tokenization and the realignment of the labels so that they correspond with the right subword tokens.
If you do not use the pipeline function, you must filter out the B and I labeled tokens. Each B and I will then be merged into a keyphrase. Finally, you need to strip the keyphrases to make sure all unnecessary spaces have been removed.
Traditional evaluation methods are the precision, recall and F1-score @k,m where k is the number that stands for the first k predicted keyphrases and m for the average amount of predicted keyphrases.