Knowledge Base : Back-end interface and possible uses

Sammanfattning: This paper addresses two different aspects of the subject known as knowledge bases, or knowledge graphs. A knowledge base is defined as a comprehensive semantically organized machine-readable collection of universally relevant or domain-specific entities, classes, and facts. The objective of this paper is to explore how a knowledge base can be used to gain information about an entity. First we present one way to access information from the knowledge base using a back-end interface. This back-end interface takes simple parameters as input which are used to query the knowledge base. The main objective here is to be able to access the right entity, to be able to answers the questions correctly.  After that follows a discussion about the need for knowledge bases and possible uses. The discussions will partly be based on results from our implementation, but also consider other similar implementation, and interviews with possible users in the business and society.  We conclude that the back-end interface developed performs well enough, with a high precision, to be ran in an unsupervised system. Furthermore we realise that the interface can be improved in several ways by focusing on smaller domains of information. Several different possible uses have been identified. From these uses a market analysis has been done from which we conclude good market possibilities. Some of the key problems with implementing the interface regards the credibility of the information in the knowledge base. This is one of the main problems that needs to be solved to fully implement knowledge bases in business and society.