Text Normalization
Text normalization is an NLP preprocessing pipeline that converts noisy, abbreviated, or misspelled text — such as SMS messages, social-media posts, and OCR output — into a clean, standardised form. It is a prerequisite step for virtually every downstream NLP task, ensuring that inconsistent surface forms do not degrade tokenisation, parsing, or classification. The method gained systematic academic treatment through Baldwin and Li (2015) and Sproat and Jaitly (2017).
Izvorni zapis
Citati kopirani doslovno iz izvornog zapisa metode. Ne impliciraju nikakvu provjeru na razini tvrdnje.
- Baldwin, T. & Li, Y. (2015). An In-depth Analysis of the Effect of Text Normalization in Twitter. NAACL-HLT 2015. · URL
- Sproat, R. & Jaitly, N. (2017). RNN Approaches to Text Normalization: A Challenge. arXiv:1611.00068. · URL
Uređene tvrdnje
Tvrdnje pohranjene u knjigu dokaza, svaka s vlastitom procjenom.
Ovaj prikaz ne izmišlja procjenu tvrdnje kada knjiga dokaza nema nijednu.
Povezane metode
Generirano iz grafa metode i prikazano kao strojno predložene relacije — ne implicira se nikakva tvrdnja dokaza.