Automatic Text Generation.
With increase in prominent users on social networking websites, it is interesting to be able to predict and generate text. We will be able to generate new content easily, without the intervention of editors and writers for blogs. This will reduce human effort and save valuable time for influential people.
We aim to generate new texts based on previous user text resources such as tweets, speeches and dialogues. To achieve this, we propose the use of Recurrent Neural Network for character level language models. The model will try to predict the next character in the sequence, resulting in generation of text.