Text Normalization and Classification System for Internet Forum

Abstract

Internet forum is one of the most common modes of knowledge sharing through text. An internet forum is an online discussion site. From a technical point of view, forums are web applications managing user-generated text contents. Text normalization is converting „informally inputted‟ text into the canonical form, by eliminating „noises‟ in the text and detecting paragraph and sentence boundaries in the text and take case restoration and suggest valid words for each invalid word in the text by using dictionary. Text classification is the process of grouping text item into related predefined classes or categories to make it easier for the user to find it. The system intends to normalize and classify the internet forum. For text normalization, Cascaded approach is used and for classification Naïve Bayes (NB) method is used. In the system, hold out method is used to evaluate the system‟s performance.