Knowledge Bases
the basis of all deep inference processes
Knowledge bases (KB) are the basic fundamental on which Revelation Engine (RE) is built. So what is a KB? Simply put, it is the embodiment gained by transforming - by means of several phase transformations - raw data sources. Data sources, for example, can be defined in two general groups: structured (i.e., a database, a spreadsheet, a movie, a photo album, a song, a album, etc.) or unstructured (i.e., a plain text file, a non-specific sound recording, randomly taken images, etc.). KBs contain a very large number of small morsels of knowledge which are the building blocks used during deep inference.
Deep Inference
turns information into knowledge
So what is deep inference? This can best be explained by the simplest of example. Consider for a moment that 2 knowledge bases exist. Let's call them KB1 and KB2. KB1 contains the following piece of knowledge John knows Mary and KB2 contains Mary knows Peter. With Revelation Engine it is now possible to ask Who does John know?. And, as you may have already guessed, the response is: Mary and Peter. So why is Revelation Engine an example of deep inference? Simple! Imagine that KB1 and KB2 do not contain two data sources but thousands, or even many millions. Further consider, that there are not 2 KBs but millions or hundreds of millions. Finally, try to phantom the fact that the inference chain depth John -> Peter is not one, but can be many hundred or thousand levels deep. If you can imagine all this, then welcome to the World of Deep Inference.
Knowledge Bases currently contained in
Revelation Engine
What types of KBs does Revelation Engine have access to? Just about anything. Data sources can be either textual, imagery, or sound. Language support stands at the moment at 145 natural languages. KBs come from all corners of human knowledge foundation: companies, universities, government organizations, law enforcement, intelligence community, banks, social media (Facebook, Google, Twitter, Instagram, LinkedIn, Reddit, etc.), research labs, and many others. Yes, even your personal information residing on your mobile devices can be turned into a KB and made available via Revelation Engine's distributed P2P architecture.
Current uses of
Revelation Engine
At the moment, Revelation Engine is primarily used by software agents (i.e., bots). Humans are rarely allowed to directly interact with it. Instead several platforms have been build on top of it which are used both by humans as well as some bots. These include, for example, customized deep inference search engines (i.e., medical, person search, company/person scoring, etc.).
ARCHITECTURE
Highly flexible and tune-able P2P architecture. No hierarchies, no centralized repositories, fully automated/distributed and maintained by societies of software agents.
PROPERTIES
Fault tolerant, graceful degradation and resilient. Inspired by evolutionary processes.
Platforms & Devices
Just about any computing device. From high-end server farms all the way down to Edge devices. And, even some types of embedded systems can become part of Revelation Engine's infrastructure.
