英和和英

名詞

PoS (plural PoSes or PoSs)

1. Initialism of part of speech.

   Alternative form: POS

   • 2010, Kees Hengeveld, Eva van Lier, “Parts of speech and dependent clauses in Functional Discourse Grammar”, in Umberto Ansaldo, Jan Don, Roland Pfau, editors, Parts of Speech: Empirical and Theoretical Advances (Benjamins Current Topics, →ISSN; volume 25), John Benjamins B.V., →ISBN, page 265:

     It should be noted that there is not always a perfect match between the logically possible systems displayed above and the PoS systems actually encountered in the sample languages. As the data in the next sub-section will show, languages may display an additional PoS class from a ‘neighboring’ system (i.e. the one represented above or below it), or they may display additional closed or derived classes of rigid lexemes.

   • 2011, Samuel Cruz-Lara, Tarik Osswald, Jordan Guinaud, Nadia Bellalem, Lotfi Bellalem, Jean-Pierre Camal, “A Chat Interface Using Standards for Communication and e-Learning in Virtual Worlds”, in Joaquim Filipe, José Cordeiro, editors, Enterprise Information Systems […] (Lecture Notes in Business Information Processing; 73), Springer-Verlag Berlin Heidelberg, →ISBN, →ISSN, part V (Human-Computer Interaction), page 544:

     Consequently, a good PoS tagger should be able to reach a degree of precision which goes beyond the superficial level of Noun/Verb/Adjective/Adverb/Grammar words.

   • 2012, Korbinian Riedhammer, “Key Phrase Extraction and Ranking”, in Interactive Approaches to Video Lecture Assessment (Studien zur Mustererkennung; volume 36), Logos Verlag Berlin GmbH, →ISBN, →ISSN, section 2 (Preprocessing), subsection 2 (Part-of-Speech Tagging), page 61:

     In a second step, each word is automatically attributed a certain part-of-speech category, such as noun, verb and alike, which is best illustrated by an example given in Fig. 4.1. […] A LCCRF can be used to predict a certain label sequence (the PoS tags) y* = {y₁, y₂,…,y_L} given a certain input word sequence w = {w₁, w₂,…,w_L} that maximizes the conditional probability p(y|w).

   • 2012, Nitendra Rajput, Amit Anil Nanavati, Speech in Mobile and Pervasive Environments (Wiley Series on Wireless Communications and Mobile Computing), Wiley, →ISBN:

     The PoS tag of the word ‘live’ leads to drastic differences in its pronunciation (/liv/ when used as a verb and /lahyv/ when used as an adverb or adjective). The PoS tagger can be a simple rule-based system Brill (1992), a data-driven technique (Brants 2000) or a hybrid approach. There are also a few standard corpora that contain word-PoS pairs.

   • 2017, A. Sheik Abdullah, S. Selvakumar, A. M. Abirami, “An Introduction to Data Analytics: Its Types and Its Applications”, in Shrawan Kumar Trivedi, Shubhamoy Dey, Anil Kumar, Tapan Kumar Panda, editors, Handbook of Research on Advanced Data Mining Techniques and Applications for Business Intelligence (Advances in Business Information Systems and Analytics, →ISSN), IGI Global, →ISBN, section 1 (Business Intelligence With Data Mining: Process and Applications), page 5:

     In PoS tagging, the sentences in the data set collection are tokenized using the PoS tagger. During this process, a part of speech such as noun, verb, adverb, adjective, conjunctions, negations and the like are assigned to every word in the sentences.

   • 2020 September 15, Ingrida Balčiūnienė, Aleksandr N. Kornev, “Quantitative Analysis of Language Competence vs. Performance in Russian- and Lithuanian-Speaking 6 Year-Olds”, in Andrius Utka, Jurgita Vaičenonienė, Jolanta Kovalevskaitė, Danguolė Kalinauskaitė, editors, Human Language Technologies – The Baltic Perspective […] (Frontiers in Artificial Intelligence and Applications; 328), Amsterdam: IOS Press, →DOI, →ISBN, page 225:

     Much less is still known about relationships between different parts of speech (PoSs) along the developmental course and their role in the acquisition of discourse skills. […] The acquisition of PoSs means “…knowing how to use the word in the language. […]”

   • 2020 December 23, Mattmann A. Chris, “Part-of-speech tagging and word-sense disambiguation”, in Machine Learning with TensorFlow, 2nd edition, Shelter Island, N.Y.: Manning Publications Co., →ISBN, part 2 (Core learning algorithms), pages 192 and 195–196:

     The good news is that language scholars have long studied PoS across the humanities, mainly by reading literature and producing helpful guidance and rules that convey the particular PoS class a word can take on. […] The Moby effort contains the file mobypos.txt, which is a database of words in English and the PoSes they can take.

   • 2022, Tadej Magajna, “Part-of-speech tagging in Flair”, in Natural Language Processing with Flair […], Packt Publishing, →ISBN, part 1 (Understanding and Solving NLP with Flair), section 4 (Sequence Tagging), chapter section “Part-of-speech tagging tag sets”, page 49:

     The PoS tagger in the preceding script uses the Penn Treebank Project PoS tag set. It is a standardized set of tags consisting of 36 different PoS tags.

   • 2022, Valentina Piunno, “Coordinated constructional intensifiers: patterns, function and productivity”, in Carmen Mellado Blanco, editor, Productive Patterns in Phraseology and Construction Grammar: A Multilingual Approach, Walter de Gruyter GmbH, →ISBN, part I (Productive patterns: monolingual), page 140:

     For the purposes of this analysis, coordinated CIs have been dectected through a PoS-based query which included the PoS-grams patterns and the coordinating conjunction e ‘and’: e.g. [Adjective + e + Adjective], [Past participle + e + Adjective/Past participle].

2. (cryptocurrencies) Initialism of proof of stake.

語源

From / + abbreviation of positive (compare pos), using the tone indicator format of slash-prefixing set by the earlier /s (“sarcasm”).