Approximate N-Gram Markov Model for Natural Language Generation

TL;DR

This paper introduces an Approximate n-gram Markov Model that leverages directed word associations to improve natural language generation, combining benefits of word association and Markov models for more effective bag generation.

Contribution

It presents a novel approximation method for n-gram Markov models using directed word associations, enhancing bag generation and potential applications in machine translation.

Findings

Effective approximation of n-gram tables using word associations

Improved bag generation quality demonstrated

Potential for enhanced lexical selection in translation

Abstract

This paper proposes an Approximate n-gram Markov Model for bag generation. Directed word association pairs with distances are used to approximate (n-1)-gram and n-gram training tables. This model has parameters of word association model, and merits of both word association model and Markov Model. The training knowledge for bag generation can be also applied to lexical selection in machine translation design.