Quality Assessment of Text Data Using C-RNN

As the Internet came into existence, the new medium of communications was established, these include Emails, SMS, calls over the Internet, Social media, etc. In short duration, these mediums turned out to be quite popular among the masses. The world soon began to flood with textual data. This research is dedicated to performing an assessment on the widely available textual data. The necessity of assessing texts is since not all the data present in the world is beneficial for the common man. Textual data is unpredictable, unstructured, and noisy in nature. Thus, assessing texts is not an easy task. To further increase this complexity, this research uses SMS data for its purpose. SMS is a very popular medium of communication in the early 2000s and even today in the form of text-based chat applications. This generation is an age of smartphones, with the rapid increase in the number of smartphone users, the daily traffic of SMS data keeps increasing at an exponential rate. Quality assessment of SMS data has become extremely challenging due to the complexities and changing behavior of messages imposed by spammers. This research proposed a novel approach of quality assessment of textual data using a deep learning Convolutional Recurrent Neural Network (C-RNN) model, for quality assessment of textual data, it is generally used in image processing. It is a hybrid neural network; consists of Convolutional Neural Network (CNN) and Recurrent Neural Network (RNN) both. Thus, making it more powerful than the other two.