Construction of Knowledge Graphs for video lectures

Knowledge Graphs (KG) have become very important in representing both structured and unstructured data. Knowledge graphs are penetrating our daily lives, be it intelligent voice assistants or Facebook friend search. In this research paper, we are focusing on how Knowledge Graphs can be constructed for a video lecture and list down the various important steps that are involved in the process of construction of the graph. Knowledge Graphs are a way of modelling a knowledge domain programmatically with the aid of tools and techniques like machine-learning algorithms, packages like NLTK, subject experts etc. A knowledge graph representation combines data both in structured and unstructured format. Moreover, the knowledge graphs are commonly built on top of existing databases like Wikipedia, Yago[1] to name a few. Video lectures are the most sought-after form of learning in this current scenario. With the rise in demand for video lectures, people have started to make a lot of videos lectures and made them available in YouTube or as online courses. Knowledge graphs offer a way to streamline workflows, automate responses and scale intelligent decisions. By representing the video lecture as a graph, we will be able to represent the content and the knowledge of the video as a graph. Knowledge Graph thus obtained from the video lectures will become a knowledge cloud that can be used for developing various intelligent applications like domain specific chat-bots, recommender engines and so on.