Lexical substitution with learning from user response

Anže Šuštar (2011) Lexical substitution with learning from user response. EngD thesis.

Preview

PDF Download (1002Kb)

Abstract

The thesis focuses on the implementation of an advanced lexical substitution system that was designed as a web application. Lexical substitution is the task of identifying synonyms for words and phrases used in the original text, which enables us to create a new text that is using different words as units of speech while conveying the exact same message. A brief description of the technologies that were used to build the application (namely PHP, MySQL, JavaScript, HTML and CSS) is presented. The problem of lexical substitution itself is presented from a theoretical standpoint, along with a theoretically sound design of a lexical substitution system. The actual implementation of a lexical substitution system and the accompanying technical challenges are presented as the reader progresses. Final results and ideas that could further improve the system are explained in the conclusion.

Actions (login required)