Over the past few years, Knowledge Graphs have taken off at many organizations. Are knowledge graphs a new concept or is it the evolution of an older concept called Knowledge Maps? Let’s look more closely at both.
A Knowledge Graph (K-Graph) as it turns out is described in many ways. It has its roots as embedded intelligence within a Google Search. This intelligence exposes the “thing” you searched for and their relationship between other similar or related “things”. A more general description of a Knowledge Graph comes from “Towards a Definition of Knowledge Graphs”, which states “A knowledge graph acquires and integrates information into an ontology and applies a reasoner to derive new knowledge”. I would suggest reading the article to form a clearer definition of a knowledge graph (they provide several examples).
Constructing the Knowledge Graph
A K-Graph is essentially a data structure that allows you to contextualize entities and organize those correlations between entities or multiple types of entities. A sample process of constructing a knowledge graph consists of Collecting and analyzing your data, Data Extraction & Integration, Data Linking & Enrichment, Storage, Querying & Inferencing, Search and Visualization. An excellent tutorial on K-Graphs can be found at Metaphacts: Getting Started with Knowledge Graphs.
Knowledge Graph Applications: Some of the applications of Knowledge Graphs include: semantic search, automated fraud detection, intelligent chatbots, advanced drug discovery, dynamic risk analysis, content-based recommendation engines, and knowledge management systems.
At a time when organizations need to ‘know what they know’ and use that knowledge effectively, the size and geographic reach of many of them, along with the proliferation of data, information and knowledge make it especially difficult to locate existing knowledge and get it to where it is needed. Once you let employees leave, you run the risk of pertinent knowledge leaving with them. The focus of the knowledge mapping effort should center on the employees within the organizations. Mapping responsibilities, expertise and the products employees produce will help gain an understanding of the know-how within the organization and if that know-how is valuable to the organization going forward.
Constructing the Knowledge Map
Creating a knowledge map is an excellent tool to facilitate the identification of the key knowledge holders, knowledge gaps and identify areas where knowledge is eroding. However, performing a knowledge mapping exercise should focus on a department, functional area, or specific organization domain and gradually be built upon until an entire knowledge map of your organization exists.
Before starting to develop the knowledge, map ensure the intentions are clear. This includes understanding the purpose and scope of the knowledge map. The thing to keep in mind is that there is never a single map for every purpose. Knowledge mapping is about relationships (people-to-people, people-to-content, people-to-applications), and how these relationships interact within the organization.
To construct the knowledge map, you will go through a process that includes interviewing employees to understand how they work, the processes they utilize, the content they create and access, the applications they use and why, as well as the people they interact with. This is done because knowledge is found in processes, relationships, policies, people, documents, conversations, and through links to suppliers, competitors, and customers. This information is used to graphically represent these relationships onto a map. However, instead of creating a static map, in which several views of the knowledge would need to be created, I suggest using software to create an interactive map. An interactive knowledge map will provide different perspectives of the relationships and the knowledge each employee has.
Knowledge Map Applications: Knowledge Maps (K-Maps) are viewed as an ancillary but important tool in Knowledge Management. K-Maps are used to increase the efficiency and organization of knowledge within the enterprise. K-Maps are central to the creation of expertise locators; beneficial to KM processes used in knowledge capturing, sharing, and the visualization of knowledge; as well as being essential to understanding and identify where the knowledge is located in the enterprise, and determining what intellectual assets are essential to the enterprise.
K-Graphs and K-Maps
While K-Graphs and K-Maps are both focused on graphically representing relationships between entities; K-Graphs is focused on data (structured, semi-structured and unstructured) and K-Maps is focused on people and their relationships to not only data but also to other people and applications. They also served different purposes. K-Graphs is essential to improving search and adding intelligence to applications and systems (i.e., chatbots, content-based systems, and KM Systems), while K-Maps are focused on uncovering and representing knowledge and its related knowledge holders. Knowledge Graphs may have been influenced by Knowledge Maps but clearly, they are different. Due to their distinct differences and purposes an organization can make use of both of these powerful tools.