Learning to Extract Keyphrases from Text

By Peter D. Turney. arXiv.org, 2002

364 citations recorded by the source corpus.

Many academic journals ask their authors to provide a list of about five to fifteen key words, to appear on the first page of each article. Since these key words are often phrases of two or more words, we prefer to call them keyphrases. There is a surprisingly wide variety of tasks for which keyphrases are useful, as we discuss in this paper. Recent commercial software, such as Microsoft?s Word 97 and Verity?s Search 97, includes algorithms that automatically extract keyphrases from documents. In this paper, we approach the problem of automatically extracting keyphrases from text as a supervised learning task. We treat a document as a set of phrases, which the learning algorithm must learn to classify as positive or negative examples of keyphrases. Our first set of experiments applies the C4.5 decision tree induction algorithm to this learning task. The second set of experiments applies the GenEx algorithm to the task. We developed the GenEx algorithm specifically for this task. The third set of experiments examines the performance of GenEx on the task of metadata generation, relative to the performance of Microsoft?s Word 97. The fourth and final set of experiments investigates the performance of GenEx on the task of highlighting, relative to Verity?s Search 97. The experimental results support the claim that a specialized learning algorithm (GenEx) can generate better keyphrases than a general-purpose learning algorithm (C4.5) and the non-learning algorithms that are used in commercial software (Word 97 and Search 97).

Subfields: Artificial Intelligence, Information Systems

View Paper: https://www.semanticscholar.org/paper/6f915f50f273c8b561978761f9ea2179e8d25296

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

In Information Systems

Citations in the corpus

Automatic Keyphrase Extraction: A Survey of the State of the Art (2014, methodology): “Early supervised approaches to keyphrase extraction recast this task as a binary classification problem (Frank et al., 1999; Turney, 1999; Witten et al., 1999; Turney, 2000).”
Learning Algorithms for Keyphrase Extraction (2000, methodology): “Extractor keyphrase extraction algorithm (Turney, 1997, 1999).”
SemEval-2010 Task 5 : Automatic Keyphrase Extraction from Scientific Articles (2010, methodology): “…in keyphrase extraction has led to the development of several new systems and techniques for the task (Frank et al., 1999; Witten et al., 1999; Turney, 1999; Hulth, 2003; Turney, 2003; Park et al., 2004; Barker and Corrnacchia, 2000; Hulth, 2004; Matsuo and Ishizuka, 2004; Mihalcea and Tarau,…”
Using Noun Phrase Heads to Extract Document Keyphrases (2000, methodology): “For keyphrase extraction, we are using Peter Turney’s Extractor [10, 11].”
Automatic Keyphrase Extraction via Topic Decomposition (2010, methodology): “A pioneering achievement in keyphrase extraction was carried out in (Turney, 1999) which regarded keyphrase extraction as a classification task.”
Main Core Retention on Graph-of-Words for Single-Document Keyword Extraction (2015, methodology): “supervised [11,16,24] depending on whether the extraction process involves a training part on some labeled inputs.”
Keyphrase Extraction in Scientific Publications (2007, methodology): “Turney’s GenEx [5] system computed a vector of nine features to represent candidates.”
Clustering to Find Exemplar Terms for Keyphrase Extraction (2009, methodology): “In methods like (Turney, 1999; Elbeltagy and Rafea, 2009), candidate keyphrases were ﬁrst found using n-gram.”
TextRank: Bringing Order into Text (2004, methodology): “Neither Turney nor Frank report on the recall of their systems, but only on precision: a 29.0% precision is achieved with GenEx (Turney, 1999) for five keyphrases extracted per document, and 18.3% precision achieved with Kea (Frank et al., 1999) for fifteen keyphrases per document.”