Abstract
Hearing impairment is a partial or total inability to hear. Children cannot learn speech and develop vocabulary at a regular pace due to which process of language development in such children is slow. For hearing impaired children, reading and writing English language text is considered to be a complex task. A Text is always composed of sentences and the sentences are made by the combination of complex and simpler words. Extensive use of complex words in a text (stories, essays, study material, or any text available on web stores) makes it difficult to read, write, understand and interpret. This causes more language inconsistency and learning problems for hearing impaired children as compared to the normal hearing peers. Proposed methodology in this paper focuses on the identification of difficult words from the English text for assisting hearing impaired children in learning the English language in a less complex way. Developed system uses C45 decision tree learning model to classify words as difficult or not difficult from the given text after the extraction of lexical and semantic features of words based on linguistic rules specific to hearing impaired children. Proposed methodology, is a full fledge learning tool that not only helps a child to improve and learn reading but also assists the therapist in conducting academic exercises and conclude that which words are difficult or which are not. Results indicate improved accuracy as compared to previous techniques used for the respective research.