Automatic Keyphrase Extraction: A Survey of the State of the Art
By Kazi Saidul Hasan, Vincent Ng. Annual Meeting of the Association for Computational Linguistics, 2014
546 citations recorded by the source corpus.
While automatic keyphrase extraction has been examined extensively, state-of-theart performance on this task is still much lower than that on many core natural language processing tasks. We present a survey of the state of the art in automatic keyphrase extraction, examining the major sources of errors made by existing systems and discussing the challenges ahead.
Subfields: Artificial Intelligence
View Paper: https://www.semanticscholar.org/paper/9ea8f5ff2cbd3a65cf0db91b95430bae7d514359
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Sebastian Riedel
- Isabelle Augenstein
- Andrew McCallum
- Akiko Aizawa
- Mohit Bansal
- Hannaneh Hajishirzi
- Ygor Gallina
- Cornelia Caragea
Citations in the corpus
- SemEval 2017 Task 10: ScienceIE - Extracting Keyphrases and Relations from Scientific Publications (2017, background): “…the task to be interesting for researchers studying tasks aiming at understanding scientific content, such as keyphrase extraction (Kim et al., 2010b; Hasan and Ng, 2014; Sterckx et al., 2016; Augenstein and Søgaard, 2017), semantic relation extraction (Tateisi et al., 2014; Gupta and Manning,…”
- PositionRank: An Unsupervised Approach to Keyphrase Extraction from Scholarly Documents (2017, methodology): “The ranking based on tf-idf has been shown to work well in practice (Hasan and Ng, 2014, 2010), despite its simplicity.”
- Multi-Task Learning of Keyphrase Boundary Classification (2017, unknown): “Typical KBC approaches therefore rely on hand-crafted gazetteers (Hasan and Ng, 2014) or reduce the task to extracting a list of keyphrases for each document (Kim et al.”
- Keyphrase Generation for Scientific Document Retrieval (2020, background): “One reason for this lack of evidence may have been their relatively low performance discouraging attempts at using them for indexing (Liu et al., 2010; Hasan and Ng, 2014).”
- IdioLink: Retrieving Meaning Beyond Words Across Idiomatic and Literal Expressions (2026, unknown): “…metrics (more in Appendix D): (i) nDCG@10 (binary relevance), which emphasizes ranking quality at top positions (Lee et al., 2024; Günther et al., 2025); (ii) R– Precision, which measures precision at rank R, the number of relevant documents per query (Wu et al., 2024; Hasan and Ng, 2014).”