Synchronous Distributed Knowledge Construction for Learning

Manuel Prieto, Jose A. Olivas, Aurora Vizcaíno,

Laboratorio de Cognición, Colaboración y Aprendizaje

Departamento de Informática

Universidad de Castilla-La Mancha

 

Ronda de Calatrava 5

Ciudad Real 13071 España

Tel: + 34 926 295300 Fax: +34 926 295354

E-mail: {mprieto|avizcain|jaolivas}@inf-cr.uclm.es

http: www.inf-cr.uclm.es/~lcca


COVER PAGE

Synchronous Distributed Knowledge Construction for Learning

 

During the second half of the Twentieth Century, Psychology, Computing and other Sciences have contributed with their principles and experiences to establish a great amount of knowledge about cognition and its regulating function of human activity.

There have been produced some models about knowledge structure, the ways they are represented in mind and the complex retrieval procedures. As consequence, we have new instruments to perform better human learning. In Cognitive Science, learning is understood a process of subsequent modifications of cognitive structures that leitmotivs human behavior.

Learning and Development

There have been several theoretical views about relations of learning and development.

In this point, Vygotsky gives a different grasp with the concept of Zone of Proximal Development (ZPD) defined as the distance between the real development level settled by a problem solving capacity, and the potential development level determined by the problem resolution under the assistance of another - more capable - person.

The Cultural-Historical Theory of Human Development perceive the psychological activity as a permanent reorganization of internal forms and the concurrent control of conditional or cultural forms involving operations with different symbols that are external at the beginning and later becomes internal. In other words, development can be explained as a knowledge acquisition produced within the personal process of embodiment to the human culture.

The understanding of thought and language relation, the importance of using mediation, the use of ZPD principle and the social –collaborative – conception are essential aspects when constructing net-computer based learning-aid systems.

Cooperation, Negotiation and Learning

If human knowledge presumes the capacity to perform specific operations, it’s obviously necessary to create the adequate environmental conditions making good use of previous already mastered actions and, building with them, the new operations.

To perform the construction of operations it’s necessary, first of all, The Search. The search is normally oriented by the Problem. The problem itself aims the Action Plan accomplishing the function of anticipatory scheme. This operational schematic sketching may be general at the beginning. Then, and thanks to the spontaneous trial process, the General Anticipatory scheme becomes more and more structured and differentiated providing a new operation.

It is necessary to assist the learner orienting him/her to search new operations in the right direction. This orientative capacity could be achieved by the problem itself. In this operation development process, the activity must be socially organized, for instance, with group work. Only under cooperation conditions, the students begin to understand other point of views, differentiating them from her/his own. The adaptation process to other person’s positions produces more dynamic and logical thinking.

To build concepts and operations it’s important to carry out general discussions. During discussions and negotiations, the learners discover other positions and different solution proposals. Understanding these differences, they must find links between distinct positions and finally build a unified system with all.

Negotiations are process through which parties try to reduce or remove a conflict between them. A successful negotiation, then, is one that allows these parties to reach a compromise using communication, adaptation and persuasion. There are many important and interesting problems related with the negotiation processes intended to construct shared knowledge during learning

 

Conceptual and Procedural Learning

In spite of it’s difficult to separate both kinds of learning in real live, it is convenient to split them for planning and organizational proposes:

The concept is associated with theory, information and identification algorithm. Procedures with practice, operations and transformation algorithm. The concept field of study is description while in procedures is execution. We are intended to develop facilities for general procedure learning instead of those for specific concept acquisition. Generally speaking, procedural learning could be given both in cognitive and psycho-motor domains.

Knowledge Representation

The aim is to get a method for representing procedural learning using collective (maybe distributed) knowledge representation. We present an application of O-A-V-tables (Object-Attribute-Value) and A-O-D-tables (Action-Objects-Description) and arrive at a method for representing Collaborative Knowledge. This method then allows us to summarize the partners knowledge using a previously established paradigm and describe the procedural output (or final) processes.

 

A system for procedural learning using collective (maybe distributed) knowledge representation

Method consists on giving empty tables to the partners, to fill in by means of Cooperation and Negotiation processes, to get:

  1. A list of all the objects of the problem, with its attributes and values.
  2. All the possible actions with the objects involved and a brief description.
  3. A Knowledge Map: The sequence of operations to get the goal.

Then, the Knowledge Based System (KBS), interacts with the team members, to generate the algorithm accomplishing the goal. Let’s show a real example of an O-A-V-table: The problem of assigning and optimizing emergency resources is typical on distributed and collaborative knowledge. The table corresponds to the object "Free Resource":

Object:

Free Resource

 

Attribute

Value

Observations

Position

(x,y)

 

Type

Truck/Patrol/Helicopter

 

Mobility(M)

M [0,1]

Linguistic labels given by the users that describes the mobility of the resource. (Fuzzy logic).

Proximity(P)

list (P)

P ={p1, p2, ... ,pn}

( pi [0,1])

pi is the degree of proximity of the resource to the Significant Point si.The KBS get it from the distance in Km. with a distribution function. (Fuzzy logic).
Degree of covering to the respective Significant Points.

list (C)

Each element of the list identifies the degree in that a Point is covered for this resource, result of combination of mobility and proximity to each point and the relevance of each point.
General covering level(G)

G [0,1]

Combination of all the elements of C. Measure of the good behaviour of each resource. (Fuzzy logic).

Nowadays, we are trying to introduce Prototypes Theory to our method. As a parallel example to our case we can compare the act of driving a car in a hail storm: if it starts to hail while we are driving, we adjust to our "driving under potentially dangerous conditions" scheme. In other words, we associate a fact or a set of facts with a paradigm so that the paradigm interprets the situation and the actions we carry out depend on it.

To generalise, many of the actions we carry out in our daily life depend on our constituting an interpretation, on our finding the most similar paradigm or prototype for the circumstances of the problem. In our case as well, the KBS simulates the partner’s capacity for interpreting the situation and finding the prototype of the problem that is most appropriate in the current conditions.

Bibliography

Olivas, J. A. and Sobrino, A. (1994) An Application of Zadeh's Prototype Theory to the Prediction of Forest Fire in a Knowledge-based System. Proceedings of the 5th. International IPMU (Information Processing and Management of Uncertainty in Knowledge-based Systems), VOL. II, Paris, 747-752.

Ruiz, F. Prieto, M. Ortega, M. Bravo, J. Sanz, J y Flores, J. (1996) Cooperative distance learning with a Integrated System for Computer Assisted Laboratory Work. In Diaz y Fernández Lecture Notes in Computer Science: Computer Aided Learning and Instruction. Springer Verlag. Berlin.

Vygotsky, L. S. (1962) Thought and Language. Cambridge MA MIT Press.

Zadeh, L. A. (1982) A note on prototype set theory and fuzzy sets. Cognition, 12, 291-297.