Identifying Relevant Text from Text Document Using Deep Learning

摘要

Now-a-days the amount of information available on the web is enormous and incrementing at an exponential rate. Thus identifying relevant text from text document has become very crucial. Text classification is the task of relegating a document under a predefined category. There are several methods to identify which words in text documents are important to explain the category it is associated with. The proposed approach uses convolution neural network with deep learning. And the deep learning is used to predict the categories accurately. Thus by calculating the test's accuracy by F1 Score, we get an accuracy value which is approximately equal to 1.