There are a vast amount of information stored in websites, wikis and social networks, but how do we process those information within a give time frame without being lost and disoriented.
Trying to visualize knowledge relations is one way to help organize what we know and we are by far faster in processing visual information than text information (comprehension of visual signals over text).
Getting to the bottom
Everyone need to comprehend knowledge by making notes about what he/she has heard, read and saw. Notes of various types help us to retrieve information much faster as we store individual characteristics (writing style, markings, colours,mind maps etc.) with those notes.
In the example below, we used a well known system called Mediawiki (Wikipedia is based on it), to scribble down our notes, combine them with additional information and at a latter point generate some lists and maps. The tool that will help us to do all that is called SMW (Semantic Mediawiki). In the first place it is used to mark semantic relevant items (words, numbers, sentences etc.) and to do some inference or create correlated semantic visualization without specialized IT or programming skills.
Our notes (notes on a Chapter 4 Segmentation from Drummond Strategic Marketing) below is some text, enhanced with some marking on semantic relevant words (symbolized through red circle) .
Using semantics to mark particulars
While with any ordinary text file (Word, Excel etc.) the retrieval system can not differentiate between important from unimportant information, our system on the other hand due to its annotation capability (mark particulars such as keywords and other properties) can turn simple notes into semantic information. Means selected characteristics can be used as filter and divided semantic relevant from irrelevant information (Just because I have a car doesn’t mean I will reach my target, only with the right directions on hand, will I be able to follow a route that lead to the target and so those semantic characteristics).
The screen dump below shows an example with all its marked property-value pairs (Keyword-Culture, Keyword-Market Segmentation,Type-Chapter etc.).
Those marked property-value pairs are now identified by the system as meaningful to the author as well as to the system.
Visualization and inference map
As the system is aware (meaning it recognizes it existence, characteristics) of those semantics (meanings), we now can ask the system to make inference maps or lists (or even combined list) of stored information.
Below, we can see a hypergraph map on our notes we just created, and without further need of additional programming the system generates a map of semantics and at the same time looks into relationships in earlier documents that have similar property-value pairs and put them into relationship with our current document.
The visualization of particulars (semantics, property-value pair) is a dynamic process and with every new information added (property-value) the relationships are shifting.
This process described can help us retrieve information in much quicker way than by just searching word by word connections or by trying to organize information through a more formalized approach, as id does not restrict us in how we make out notes as long as we identify semantic relevant particulars.
How it works …
With the notes made in our example, culture has been marked as semantic relevant. Using this keyword, the system finds other articles/chapters that have been created earlier that uses the same property-value pair (in our example Keyword::Culture). Relevancy is identified through the same property-value pair, and in terms of culture it finds others terms such as anthropology, EMIC, cultural brand etc. that are relevant for others documents as well, this visualization and their connections go beyond the notes we just made. The system finds those connections based on earlier information stored.
Of course the system can only act on objects that uses property-value pair, meaning it needs disciplined effort do mark those relevant information but with each information added, the system will increase its ability to find meaningful relationships.
Of course there are other ways of visualizations but in our example above we got a free lunch, made possible by people who are in involved in developing the Semantic Mediawiki or Hypergraph as a powerful tool to interact with information and knowledge.
For me personally, the time has been long gone where information are stored in plain Word or Excel files, nowadays I attachments those documents to a semantic entity.