Detail publikačního výsledku

Query-Based Keyphrase Extraction from Long Documents

DOČEKAL, M.; SMRŽ, P.

Originální název

Query-Based Keyphrase Extraction from Long Documents

Anglický název

Query-Based Keyphrase Extraction from Long Documents

Druh

Stať ve sborníku v databázi WoS či Scopus

Originální abstrakt

Transformer-based architectures in natural language processing force input size limits that can be problematic when long documents need to be processed. This paper overcomes this issue for keyphrase extraction by chunking the long documents while keeping a global context as a query defining the topic for which relevant keyphrases should be extracted. The developed system employs a pre-trained BERT model and adapts it to estimate the probability that a given text span forms a keyphrase. We experimented using various context sizes on two popular datasets, Inspec and SemEval, and a large novel dataset. The presented results show that a shorter context with a query overcomes a longer one without the query on long documents.

Anglický abstrakt

Transformer-based architectures in natural language processing force input size limits that can be problematic when long documents need to be processed. This paper overcomes this issue for keyphrase extraction by chunking the long documents while keeping a global context as a query defining the topic for which relevant keyphrases should be extracted. The developed system employs a pre-trained BERT model and adapts it to estimate the probability that a given text span forms a keyphrase. We experimented using various context sizes on two popular datasets, Inspec and SemEval, and a large novel dataset. The presented results show that a shorter context with a query overcomes a longer one without the query on long documents.

Klíčová slova

keyphrase,keyword,long documents,query-based keyphrase extraction,BERT,transformer

Klíčová slova v angličtině

keyphrase,keyword,long documents,query-based keyphrase extraction,BERT,transformer

Autoři

DOČEKAL, M.; SMRŽ, P.

Rok RIV

2023

Vydáno

04.05.2022

Nakladatel

LibraryPress@UF

Místo

Jensen Beach

Kniha

The International FLAIRS Conference Proceedings

Edice

2022

ISSN

2334-0762

Periodikum

Proceedings of the ... International Florida Artificial Intelligence Research Society Conference

Svazek

2022

Číslo

35

Stát

Spojené státy americké

Strany od

1

Strany do

4

Strany počet

4

URL

https://journals.flvc.org/FLAIRS/article/view/130737

BibTex

@inproceedings{BUT179282,
  author="Martin {Dočekal} and Pavel {Smrž}",
  title="Query-Based Keyphrase Extraction from Long Documents",
  booktitle="The International FLAIRS Conference Proceedings",
  year="2022",
  series="2022",
  journal="Proceedings of the ... International Florida Artificial Intelligence Research Society Conference",
  volume="2022",
  number="35",
  pages="1--4",
  publisher="LibraryPress@UF",
  address="Jensen Beach",
  doi="10.32473/flairs.v35i.130737",
  issn="2334-0754",
  url="https://journals.flvc.org/FLAIRS/article/view/130737"
}

Dokumenty

Query-Based Keyphrase Extraction from Long Documents