TnT - A Statistical Part-of-Speech Tagger

Abstract

Trigrams'n'Tags (TnT) is an efficient statistical part-of-speech tagger. Contrary to claims found elsewhere in the literature, we argue that a tagger based on Markov models performs at least as well as other current approaches, including the Maximum Entropy framework. A recent comparison has even shown that TnT performs significantly better for the tested corpora. We describe the basic model of TnT, the techniques used for smoothing and for handling unknown words. Furthermore, we present evaluations on two corpora.

8 pages

References in corpus (2)

• An Annotation Scheme for Free Word Order Languages
• Comparing a statistical and a rule-based tagger for German

Cited by in corpus (17)

• Creating emoji lexica from unsupervised sentiment analysis of their descriptions
• Natural Language Understanding with the Quora Question Pairs Dataset
• Detect Related Bugs from Source Code Using Bug Information
• Modeling of learning curves with applications to pos tagging
• A Hidden Markov Model Based System for Entity Extraction from Social Media English Text at FIRE 2015
• Highly Fast Text Segmentation With Pairwise Markov Chains
• A Logic-based Approach for Recognizing Textual Entailment Supported by Ontological Background Knowledge
• Adaptive scheduling for adaptive sampling in POS taggers construction
• Luganda Text-to-Speech Machine
• Reliable Part-of-Speech Tagging of Historical Corpora through Set-Valued Prediction
• A Review on Part-of-Speech Technologies
• Parsing Universal Dependencies without training
• External Lexical Information for Multilingual Part-of-Speech Tagging
• Formal models of Structure Building in Music, Language and Animal Songs
• A Hierarchical Deep Learning Natural Language Parser for Fashion
• Unsupervised Part-of-Speech Induction
• When silver glitters more than gold: Bootstrapping an Italian part-of-speech tagger for Twitter