Natural language processing is a field within machine literacy that's still in its early stages. The need for easy communication with operating systems, new operations for artificial intelligence, and increased access to textual data sets all have an impact on how important natural language processing is to the development of artificial intelligence. The traditional textual representation ways like Bag- of- Words have failings, as they do not consider word dependences and semantics. To enable a more suggestive textual representation, we suggest a new system grounded on graph representations that considers both original environment and global relations between words.