Natural language processing and knowledge graphs are interlinked concepts in the field of artificial intelligence and machine learning which are extremely useful in understanding the vast amount of information present in human language. This report seeks to outline the approach taken to use NLP in the Python programming language to create a graphical representation of data which is present in the form of long text. This has proven to be useful in interpreting the meaning and relationships hidden among real-world objects. Processing natural language and combining features of knowledge graphs with those of entity-relationship diagrams can result in an insightful visualization, ready for analysis and evaluation. Through this project, the capabilities of the Python language are explored in the field of NLP and graphical representations, to extract entities from not only simple sentences but also complex ones by processing complete PDF documents, which can contain more than two entities with multiple relations connecting them. Moreover, the report also points out ongoing developments and improvements made to get an even more powerful and worthwhile model to operate on data and illustrate the relationships between them, meanwhile also identifying attributes associated with each entity. The report also tries to provide an overview of Isabelle/HOL and its capabilities as a theorem prover, and how this can be used to understand entities and relations more deeply.