Showing posts with label knowledge_representation. Show all posts
Showing posts with label knowledge_representation. Show all posts

Thursday, September 27, 2012

Life is Full of Tough Choices: Mindmap or Concept Map?

In my previous two posts, I introduced Mindmaps and Concept Maps. Today, I'll present some rules of thumb regarding when to use each type of map.
This Concept Map illustrates the similarities and differences among Concept Maps, Common Mindmaps, and Pure Mindmaps. As Roy Grubb stated in WikIT, "Concept Maps have rigor, Mindmaps have vigor." Although this is a generalization, it really rings true regarding the differences between the uses of the two types of map. Since, to the best of my knowledge, all software packages except one create Common Mindmaps, that is what I'll focus on.

To cut to the chase, Mindmaps are best used:
  • When you have ideas in your mind and want to quickly record them,
  • For categorizing, sub-categorizing, and re-categorizing information,
  • For brainstorming, since ideas can be recorded quickly and then be reorganized, as needed,
  • For writing projects, to aid in the generation of ideas, and then to organize those ideas into a consistent, usable form,
  • For notetaking, to either take notes on a book or article you are reading, or to take notes live in a meeting or a lecture,
  • For meetings, to plan the meeting, take notes, and publish minutes,
  • For presentations, the map's central topic is the topic of the presentation, the main (first-level) branches represent the main topics to be covered and the order in which to do so, and each successive level of the map uncovers greater levels of detail of the topics specified by the main branches,
  • For planning or organizational tasks, the Mindmap allows you to quickly record all of the information that needs to be organized without worrying whether you are placing the information in the right spot in the map, and then moving branches around as necessary, after further reflection,
  • For managing the execution of tasks specified in a planning map,
  • For studying or memorizing, the categorization properties of Mindmaps illustrates which ideas are part of of the various higher-level ideas, and the visual design aids in retention of this information, and
  • For problem solving and generally stimulating the flow of creative juices, the Mindmap allows you to see the big picture and details at the same time, presenting you with the opportunity to see associations and new ideas that are not apparent in other representations.
Concept Maps, on the other hand, are best used:
  • When there is a primary focus question to be answered,
  • For breaking down and representing complex knowledge, when the knowledge is complex enough that the representation requires types of relationships other than simple categorization and sub-categorization,
  • To teach a concept as described above, through the use of relationships and their types,
  • As a Human-Computer Interface between you and an ontology and/or a computational reasoning system,
  • For the representation of knowledge in an historical or archival context, where new relationship and concepts are added as they are created or discovered,
  • For the deep understanding of a concept, creating a Concept Map is effective, because if you must have a complete understanding of a concept, the building of a map points out areas where your understanding still has gaps, which you may then fill by consulting external sources, and
  • For the creation of a representation sufficiently rigorous for a computer to reason over, a properly constructed Concept Map is ideal.
To summarize, use Mindmaps for problem solving, stimulation of creativity, planning and organization, and use Concept Maps for rigor, detailed modeling of knowledge, and as a way of representing knowledge in a form that a computer can use to reason with.

These are just examples of uses for Mindmaps and Concept Maps. There are many that I have not mentioned. However, this comparison should provide you enough of the flavor for you to decide which type of map to use for which situations that you encounter.

Once again, thanks go out to WikiT for the image, and next week the Mindmap Shootout actually begins!

Monday, March 9, 2009

Reality, What a Concept (Map)!

Outwardly, Mindmaps and Concept Maps are quite similar. They both have:

  • Nodes, representing concepts, ideas, or topics,
  • Links between the nodes, and
  • They both begin with a central idea.
If we dig into the ideas behind them, however, differences begin to emerge. Last time, I briefly presented this information for Mindmaps. Today, I'll do the same for Concept Maps. Again, keep in mind that this post just skims the surface of the topic of Concept Maps. It is a large field of study, with plenty of literature to study on your own if you are interested.

In the late 1970s and early 1980s, Joseph D. Novak first developed Concept Maps as a tool for education. Since that time, it has developed into a powerful tool used in a wide array of fields. I stumbled upon a great Concept Map, created by Novak himself, for a simple case of a Concept Map explaining Concept Maps. It does such a wonderful job that there is really very little chance that I can add much of value to it, but I'll try to do it anyway.

In the simple case where the Concept Map is a stand-alone tool, it is often used to understand complex ideas. The user starts out with a Focus Question, and goes on to map out relationships to and within the Central Concept. There are several properties that describe a Concept Map:

  • It starts with a Central Concepts at the top and is read top-to-bottom, so it starts out looking like a tree,
  • The Map helps to answer a Focus Question,
  • A Concept Map represents organized knowledge which allows for better understanding,
  • Unlike in Mindmaps, concepts may have multiple parents in Concept Maps,
  • Nodes represent concepts,
  • There are both downward links and crosslinks,
  • Links are just as important as nodes,
  • Links are labeled and describe type of relationships between concepts,
  • Crosslinks can tie together different domains of knowledge in the Map, and
  • During the creation of the Concept Map, crosslinks can often represent creative leaps.
Often, instead of a stand-alone tool, it is part of a larger, automated system. The Concept Map is often the User Interface for a Knowledge Management System or a Computational Reasoning System. Knowledge can be input, or presented, through a Concept Map.

We can see how some simple computational reasoning might work. Structured knowledge enters a system using a Concept Map. If incomplete knowledge enters the system and matches the structure of the Map, then the system may be able to automatically infer missing relationships. As a simple example, we have a Concept Map representing the structure of a family, and knowledge enters the system saying Bob "is the son of" Lisa and Chuck "is the son of" Lisa. Based upon the knowledge input to the system through the "Family" Concept Map, the system can infer that Bob "is the brother of" Chuck, and that Chuck "is the brother of" Bob.

This Concept Map, from the same paper as the image above, represents the seasons. It presents all the information about how the seasons are determined and how they relate to each other. As you can see, a relatively small, simple Map is packed with a lot of knowledge. If you wanted to learn about seasons, this would be an effective way to do it. It would also be a good way for you to communicate what you know about seasons to someone else.

Now that I've introduced both Mindmaps and Concept Maps, next time I'll talk about when you might want to use a Mindmap rather than a Concept Map, and vice versa. If you have any questions, just post them as comments, and I'll respond to them as quickly as is possible.

N.B. Thanks to Joseph D. Novak and Alberto J. Cañas for the images used in this post, from their paper:
The Theory Underlying Concept Maps and How to Construct Them, J. D Novak & A. J. Cañas (2006), Technical Report IHMC CmapTools 2006-01, Florida Institute for Human and Machine Cognition, 2006.