Humans are storing information using a mental process called categorization. For each item you perceive there is a category and within each category you have the prototypical item who will represent all the items of this category.
A short example. Let’s take a chair. A chair is a piece of furniture with a raised surface supported by legs, commonly used to seat a single person. Chairs are supported most often by four legs and have a back. If you now take a pencil and a piece of paper and you draw a chair, what you will draw is the prototype of the chair. And no matter where you were or where you will be, when you will see an item with a raised surface, supported by legs, where you can sit on and this item has four legs, for sure you will recognize this item as being a chair.
The world is perceived by our senses. Relevant information (stimuli) is acquired and classified into a structure made of objects classes. This process is called categorization.
This structure is made of classes. You have 3 types of classes: the object, his prototype and his abstraction (example: Woodpecker, Bird, Animal). Each class contains information stored in properties: functional properties (■) and descriptive properties (▲). For a given class, his hyponym is a child class which does inherit of all the properties of the parent class. For a given class, a superordinate is a parent class which holds all the common properties. Once stored (in the long-term memory) in categories, knowledge can be represented, can be used when reasoning, making decisions and solving problems. The more we will interact with those objects, the more properties we will discover (descriptive and functional) and slowly develop an expertise. To make this mental process (categorization) become artificial, we added more properties, such as dynamical and mental properties. This last property is mainly used to learn or forget information and to enable deductive reasoning.
This categorized data structure is ready to support Artificial Intelligence processes
• Key data are kept and both functional and descriptive properties are stored
• Relations between all those data are established for smarter queries
• Superordinate levels are present so that bridges to access information are ensured
The “long term memory” of the system is now well structured and knowledge can be used in a human way.
We have called this categorized structure the VirtualBrain, a data file containing all the information, all the properties and all the needed relations between these objects classes defining the context.
As an example, we have illustrated a VirtualBrain made of one abstract object, three prototypical objects and for each three objects. We added for each of these objects their properties. Each circle can be considered as one information.
Our Workbench, called Alex, is dedicated to create VirtualBrains. From a given context, Alex will build a specific VirtualBrain which will then be used by other Artificial General Intelligent engines, such as Lisa or Hugo.
Alex core functions are SmartNeuron based for natural data processing, Semantic Network & Problem Space Modeling, Symbolic objects definition (classes & properties), Relationship definition (hypernyms and hyponyms), Learn & Forget alone from the data, Embedded Fuzzy Logic engine, Knowledge Representation, Support to create data interfaces, Virtual Brain reports. We also added a knowledge representation feature which will display all abstract and prototypical objects with their properties (example below).