Behavior perception and automatic intervention for cooperative learning in network environment

Cooperative learning is that learners learn in different places through the Internet, which greatly improves the effectiveness and convenience of learning. Although some empirical researches show cooperative learning is effective, how to guarantee maximum efficiency of cooperative learning and overcome negative effects producing by communication that still remain a problem to be analyzed and studied further. This paper introduces a quantification method to reveal dynamic mechanism and intrinsic law of distributed pair programming, a professional cooperative learning method. A behavior prediction method in pairs is presented from the viewing of game. On the basis of prediction, an automatic intervention method is proposed for irrational behavior affecting results of pair programming. Experiments show that the methods in this paper can guarantee high efficiency in pairing real time.