engleski

Measurement of Information Density in Educational Material Production

Different kinds of media are used in foreign language learning methodology. One of the newest increasingly used media is a video clip. Main information in foreign language learning is transferred from the speaker to the listener by the sound wave but it is important to take care of the visual part of video clip because it should not demand too many cognitive resources. Scientists have to take this into consideration due to the constant growth of visual activity in video clips nowadays. Several authors have pointed out that direct consequence of the high information density in visual media causes viewer’s information overload which results in the fall of his cognitive skills. Lang introduced and empirically proved LC4MP model that defines a moment of information overload. However, the problem with this model is that only trained observers can analyze video clips. In this paper, the new automatic method for information density measuring has been presented. It is based on the computer vision algorithm used for background subtraction. Many background subtraction algorithms are available in various papers but in our method, Multilayer BGS algorithm has been used. Our method has been tested on two groups of 10 short video clips from two YouTube foreign language learning channels. The difference between the two video clip groups is easily discernible by anyone and it is also statistically significant. This significant difference is proved by the data obtained from the proposed method. Our method can be used for information density measuring under the condition that some researches related to defining limits of acceptable visual activity are conducted first.

information density, background subtraction, educational material production, foreign languages, LC4MP

nije evidentirano