: Vocabulario de PLN: stemmer

NLP methods and tools > NLP algorithm > stemmer

stemmer

An algorithm used in information retrieval, text mining, machine translation, summarization and classification to reduce the query words to a basic standardized form. (Adapted from Jabbar et al., A comparative review of Urdu stemmers: Approaches and challenges, in Computer Science Review, 2019)

Simple truncation may be a good option for other languages where stemmers are not readily available. (Bergsma, Lin & Goebel, 2008)
The Lithuanian stemmer eliminates all endings and also some suffixes (but do not touch prefixes). (Kapočiute-Dzikiene, N\\oklestad, Johannessen & Krupavičius, 2013)
The observations obtained from the experiments allow us to conclude that simple automatic prefix and suffix information capture enough of the Lithuanian language inflection information; therefore it can replace features generated by part-of-speech taggers lemmatizers and stemmers. (Kapočiute-Dzikiene, N\\oklestad, Johannessen & Krupavičius, 2013)
The stemmer is rule-based thus it can cope with proper names as well as with the other words. (Kapočiute-Dzikiene, N\\oklestad, Johannessen & Krupavičius, 2013)
We also assess whether it is possible to perform named entity classification effectively without resorting to external grammatical tools such as part-of-speech taggers lemmatizers or stemmers because they have limited availability and are slow and unreliable when identifying proper names in Lithuanian texts. (Kapočiute-Dzikiene, N\\oklestad, Johannessen & Krupavičius, 2013)

http://data.loterre.fr/ark:/67375/8LP-KSHXVMPD-M

RDF/XML TURTLE JSON-LD última modificación 3/5/24

Loterre