Viewing morphology as an inference process

By Robert Krovetz. 1993

Subfields: Natural Language Processing, Computational Linguistics & Semantics

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Natural Language Processing

In Computational Linguistics & Semantics

Citations in the corpus

A Deep Relevance Matching Model for Ad-hoc Retrieval (2016, methodology): “During indexing and retrieval, both documents and query words are white-space tokenized, lowercased, and stemmed using the Krovetz stemmer [15].”
Learning to Extract Keyphrases from Text (2002, background): “ERB-1057, Learning to Extract Keyphrases from Text Some software tools for document management now support skimming by automatically highlighting keyphrases.”
Graph-based term weighting for information retrieval (2012, background): “This choice is motivated by findings showing that stemming in general is not consistently beneficial to IR (Harman 1991; Krovetz 2000), and that the overall performance of IR systems is not expected to benefit significantly from stopword removal or stemming (Baeza-Yates and Ribeiro-Neto 1999).”
Predicting query performance (2002, background): “Once queries are stopped and stemmed[ 11 ] there are quite a few duplicates, resulting in 36:1 queries per topic, on average, and 1804 queries for the 50 topics in total.”
Improving stemming for Arabic information retrieval: light stemming and co-occurrence analysis (2002, background): “Stemming appears to improve effectiveness more for highly inflected languages [38, 39] and when queries and/or documents are short [30].”
Entity query feature expansion using knowledge base links (2014, methodology): “During indexing and retrieval, both documents and query words are stemmed using the Krovetz stemmer [22].”
ReQ-ReC: high recall retrieval with query pooling and interactive classification (2014, methodology): “Documents are tokenized with Lucene’s StandardAnalyzer and stemmed by the Krovetz stemmer [10].”
Enhancing Expert Finding Using Organizational Hierarchies (2009, methodology): “We extracted free-text contents from each email and used the Krovetz stemmer [11] to stem the words in the text.”
Find-similar: similarity browsing as a search tool (2006, methodology): “We stemmed all words with the Krovetz stemmer [16].”
A Study of Stemming Effects on Information Retrieval in Bahasa Indonesia (2003, background): “Similar to what happened in English and Dutch [15, 17, 20], we found many examples where the rule-based stemmer, such as the Porter stemmer, produced non-comprehensible words.”
The Porter stemming algorithm: then and now (2006, methodology): “This algorithm spurred the development of many subsequent algorithms (Lennon et al., 1981; Porter, 2005) and, more generally, the use of stemming as a general tool in information retrieval (Frakes and Fox, 2003; Harman, 1991; Hull, 1996; Krovetz, 2000).”
Analysis of the Paragraph Vector Model for Information Retrieval (2016, methodology): “We use the Galago search engine(1) to index the corpus and stemmed terms with the Krovetz stemmer [12].”
Using Machine Translation Evaluation Techniques to Determine Sentence-level Semantic Equivalence (2005, background): “Stemming conﬂates morphologically related words to the same root and has been shown to have a beneﬁcial effect on IR tasks (Krovetz, 1993).”
A probabilistic approach to spatiotemporal theme pattern mining on weblogs (2006, methodology): “Krovetz stemmer [15] is used to stem the text.”
Query Expansion with Locally-Trained Word Embeddings (2016, methodology): “All corpora in Table 1 were stopped using the SMART stopword list5 and stemmed using the Krovetz algorithm (Krovetz, 1993).”
Deep Relevance Ranking Using Enhanced Document-Query Interactions (2018, methodology): “We removed stop words and applied Krovetz’s stemmer (Krovetz, 1993).”
Similarity Measures for Short Segments of Text (2007, background): “Many techniques have been proposed to overcome the vocabulary mismatch problem, including stemming [5] [9], LSI [3], translation models [1], and query expansion [6] [14].”
Overview of Touché 2022: Argument Retrieval (2022, unknown): “For the identification of claims and premises, the participants could use any existing argument tagging tool, such as the API 20 of TARGER [19] hosted on our own servers, or develop an own method if necessary.”
The Turn - Integration of Information Seeking and Retrieval in Context (2005, background): “Most IR studies on WSD have reported no or minor improvements in retrieval performance owing to WSD (Krovetz and Croft 1992; Sanderson 1994; Voorhees 1993). Pirkola (1998) found that in dictionary-based CLIR one may neglect WSD in request translation and database indexing, and incorporate all transl”