Automated Text Simplification Essay

Abstract – This paper tries to give a detailed explaination of the implementation of Automated Text Simplification system based on the paper `Text Simplification for Children` by Jan de Belder and Marie Francine Moens. We will give a detailed insight into the various components used in the automated system namely the Lexical Simplifier, the Syntactic Simplifier and the Wikfication module. We test the system on NCERT Science Textbooks from Standard 1st to 12th. Finally we will try to look, based on various parameters how effective the system actually is in simplifying text.

Keywords – Text Simplification, Readability, Syntactic Simplification, Lexical Simplification, Wikification
I. Introduction
Readability of a text refers to how well a reader is able to grasp or

The Lexical Simplifier
2. The Syntactic Simplifier
3. The Wikification module
A. The Lexical Simplifier
The word Lexical is derived from the Greek word Lexis which mean 'word'. Lexical simplification means simplifying difficult words so that they become easier to understand.

In the Lexical Simplification phase, we replace the difficult words occuring in the text with another simpler word. The replacement of the difficult word must be such that the meaning of the sentence is preserved. So the new word must be synonymous to the difficult word and also it must preserve the context in which the difficult word is used.

We use Wordnet to fetch the synonymous terms of a given word in the system. The system uses nltk's Wordnet wrapper to fetch the synonyms of the word. But replacing a word with its synonym does not always generate the correct replacement. It might happen that the synonym of the word might be out of context in which it was used.

For resolving the above situation we need to combine the Lexical Simplification system with a Language Model. We use the Latent Word Laguage Model for this purpose. The Latent word language model models language in terms of consecutive words, that is