مشخصات پژوهش

عنوان	Persian Text Classification Based on K-NN Using Wordnet
نوع پژوهش	مقاله ارائه شده کنفرانسی
کلیدواژه‌ها	Text Classification, K-nearest Neighbors, Wordnet, Principle Component Analysis, Persian language, Information gain
چکیده	Abstract. K-NN is widely used for text classification purpose. Basic K-NN has poor accuracy; other methods should be applied to basic K-NN to improve accuracy and efficiency. In this paper we propose a method that uses wordnet to increase similarity of documents under the same category. Documents are represented by single words and their frequencies, by using wordnet, frequency of related words is changed to acquire higher accuracy. Information gained is used to eliminate terms that are not discriminated. Words like "and", "or" and "that" in English are not important in text classification and the best way to eliminate them is to calculate their information gain. PCA is used to reduce number of features and increase speed of the method. Applying this method, we designed a faster and much accurate classifier for Persian language. Experiments show that applying this preprocessing will increase accuracy and speed of K-NN. Accuracy of the proposed K-NN classifier on Hamshahri corpus is 88.18%.
پژوهشگران	مصطفی پرچمی (نفر اول)، بهاره اختر (نفر دوم)، میرحسین دزفولیان (نفر سوم)