This project is based on paper "Language understanding for text based games Using Deep Reinforcement Learning" by Karthik, Regina, Tejas. Here is the link for the paper- https://arxiv.org/pdf/1506.08941v2.pdf The Game is called "Home World" consisiting of 4 rooms. For playing the player recieves a text description about the rooms and a questgame. A quest is something like 'you are not sleepy but hungry'.Text description is also known as 'state'. To react according to a particular situation, the player has to use action + object combination. There are predefined 5 actions viz. eat, watch, go, sleep, do and 8 objects viz. north, east, west, south, tv, apple, exercise, now.