Jämför metoder
Granska de valda metoderna sida vid sida; rader som skiljer sig är markerade.
| Ordklassmärkning (POS-märkning)× | Textsegmentering× | |
|---|---|---|
| Ämnesområde | Textutvinning | Textutvinning |
| Familj | Process / pipeline | Process / pipeline |
| Ursprungsår≠ | — | 1997 |
| Upphovsperson≠ | — | Marti A. Hearst (TextTiling) |
| Typ≠ | NLP sequence-labelling task | NLP document-structure / topic-boundary detection |
| Ursprungskälla≠ | Ratnaparkhi, A. (1996). A Maximum Entropy Model for Part-Of-Speech Tagging. EMNLP. link ↗ | Hearst, M.A. (1997). TextTiling: Segmenting Text into Multi-Paragraph Subtopic Passages. Computational Linguistics, 23(1), 33-64. link ↗ |
| Alias≠ | part-of-speech tagging, grammatical tagging, Sözcük Türü Etiketleme (POS Tagging) | topic segmentation, discourse segmentation, linear text segmentation, Metin Bölümleme (Text Segmentation) |
| Närliggande≠ | 3 | 4 |
| Sammanfattning≠ | Part-of-speech tagging assigns a grammatical category label — noun, verb, adjective, and so on — to every word in a text. It is a foundational natural-language-processing task, formalised as a statistical model by Ratnaparkhi (1996) and packaged into widely used toolkits such as Stanford CoreNLP (Manning et al., 2014), and it serves as a preliminary step for syntactic analysis and information extraction. | Text segmentation divides a long document into meaningful sections (segments) along topic or discourse boundaries. Introduced for subtopic passages by Marti A. Hearst's TextTiling (1997), it supports document-structure analysis and the detection of topic transitions in continuous text. |
| ScholarGateDatamängd ↗ |
|
|