WHAT WE'VE BUILT
We wish we could say we were the only people working to solve these problems. That's obviously not true, as it isn't true of any problems which hold widespread value.
The concept of "Semantic Processing" as you hear it today has been around for quite a while, and we think that Tim Berners-Lee does a pretty good job of explaining what it means, especially in the context of the World Wide Web.
We hold the belief that the principles we describe below, guide not only semantic relationships on the web, but also those relationships as they pertain to any set of information. This holds immense value and promise for any organizations who hold so much data, that it becomes impossible for them to see their big picture.
A rich framework to discover, report, and visualize these relationships meaningfully is critical, and is exactly what we are developing.
CONTEXT
Context is the relation of a set of data to a few keywords. This can be seen in many places, but perhaps the easiest example to describe is the modern search engine. When you search a term, it is matched against a set of sites the search engine has collected, and tags it has decided to associate with these sites.
The results emerge with large flaws, and the user is left to decide which results are garbage, and which results pertain to what they actual wanted to see.
The problem is that although it employs some pretty sophisticated techniques (or tricks), the search engine is undoubtedly unable to understand: (1) what the user meant (2) what the page was really about.
For now, this is acceptable for our non-critical searches and YouTube meanderings. Imagine though, if the user had no idea what the expected result should be. Trying to use these same techniques to pull the most relevant information from a set of data would be useless.
SEMANTICS
Semantics is what happens as a result of drawing relationships between the data itself, rather than only between the user's keywords and search engine's set of tags. Semantic information is meaningful, therefore allowing search engines, advertisements, summarizations, mashups, etc. to all function as a result of the meaning of a piece of information, and its importance to a given user.
Our semantic engine draws these relationships over time, building upon them a rich mapping which is searchable, meaningful, and maintainable.
WEB 3.0: THE SEMANTIC WEB
The last step in our process is to put the semantic relationships in a form that allows the user to capture the full size and scope of the original data, in edible portions. These "visualizations" form the scenic route around what otherwise would be a forest of unmanageable data.