Graph Analytics to Improve Business Insights

This section will cover Why Knowledge Graph is important to business and How it improves customer engagement.

When discussions about any technological concept begin to trend in online media and among business stakeholders, it is only natural to wonder whether it is all hype or whether the technology actually solves any real business problems.

This section attempts to cut through the noise by informing you about few noteworthy use cases of graph technology and graph analytics so you can make your own mind up.

This section covers set of analytic techniques that allows for the exploration of relationships between entities of interest such as organizations, people and transactions in a structured graph

In the section will cover evolution of databases, AI, world-wide web and how graph technology came into play. We will cover need for a knowledge graph and basic characteristics of a Knowledge Graph.

In this section, we will be talking about Knowledge Graph concepts and will get an understanding on what knowledge graph does which is beyond structured graph and what those concepts are. How do they stack up to various levels of maturity so you can get to that as simple extension to your structured graph, After that you can start adding more levels that provide you more and more higher level constructs of knowledge graph. As you can easily appreciate higher level maturity requires more expertise as well as lot more preparation. So we have to balance against business objectives to make sure we are appropriately building the right level of maturity to respond to a particular use case.

Text search is different from tag search. In a tag, you can do a full or partial search and the algorithm will find a tag that best meets the tag. You can use wild cards to move across search.  Text is often harder to search this way. Some one may change the order, include or not include an article and spell “and” or &.

In the world of big data, as text analysis was introduced, Lucene provided mechanisms for searching in more flexible ways. In this section, we will explore how we can bring some of the elastic search concept to graph search.

In this section, we will cover the use of classification hierarchies to search and organize network graph vertices and edges. While structured elastic graphs provide good ways of searching individual vertices, they tend to lose forest from the trees. Classification provides us a way to organize the graph using abstract concepts. We will show how these classifications are represented in the graph, used for searches. We will also examine situations involving competing experts and how their information can be combined for a search.

In this section, we will add semantic and domain layer concepts to our knowledge representation, providing us with additional capabilities for traversal and searching. We will introduce qualitative algebra and how it can be used as a powerful graph construct for knowledge-based reasoning. This should provide you with a broadened understanding of how publicly available information can be added to any graph to enable queries we often use in our day-to-day language and how you can incorporate them in your graph.

In this section, we will introduce various solution components of a Graph system and how they are stacked together for an overall solution.

Irrespective of the complexity, the components more or less exist in all levels. The difference is in the degree of sophistication. Additionally, a graph tool may provide pre-built components which can be used for loading data or interfacing with source systems or developing vertices and edges for simple solutions. In the most complex situations, the system may be built by a team of developers with expertise in user interface design, data science, data engineering and backend query development. As time goes by, we can expect libraries of modules from application vendors or system integrators offering some of these components that can plug-and-play with popular applications, such as SalesForce.com and ServiceNow.

In this section, we will use a few slides to remind you of some of the key learnings in this class and let you test drive it.

In this lecture, we will provide details on courses in this area to aid in your future learning.