SF-CNN: Deep Text Classification and Retrieval for Text Documents

Researchers and scientists need rapid access to text documents such as research papers, source code dissertations. Many are available on the Internet more time retrieve exact based keywords. An efficient classification algorithm for retrieving keyword words is required. The traditional performs less because it never considers words’ polysemy relationship between bag-of-words in To solve above problem, Semantic Featured Convolution Neural Networks (SF-CNN) proposed obtain key relationships among searching keywords build a structure matching correct documents. SF-CNN deep semantic-based bag-of-word representation document retrieval. Traditional learning methods Convolutional Network Recurrent use semantic bag-of-words. experiment performed with different datasets evaluating performance of method. classifies an accuracy 94% than algorithms.