Introduction and Motivation
Abstract:
There at least five advantages a computer can offer as opposed to a book. These are introduced here.
Using computers to aid the teaching of languages has been done probably since the late 1960ies in a number of institutions especially at universities.
What was the driving force behind this?
Several advantages can be noted theoretically for the use of computers in this domain and for the use of computers for learning in general.
It is important to note that all of these aspects must be seen from the learner's perspective.
Why the area of computer-assisted language learning (CALL) might be interesting from the perspective of computational linguistics (CL) will shown later.
 Figure 1: Immediate feedback |
One of the main reasons for using computers in a learning scenario is the possibility for immediate feedback to learner input.
In a standard class room setting this can only be achieved by the teacher as opposed to doing exercises in a workbook.
Using the computer, every learner has the possibility to receive immediate feedback.
 Figure 2: Different types of interaction |
Not only immediate feedback is possible but also different types of interaction with the computer.
The learner may not only type and read, but look at a video, point at something with the mouse, or speak into a microphone.
 Figure 3: Time and place |
Additionally the use of computers is usually not restricted by time and place, whereas face to face lessons are always limited in this sense.
One consequence is that the pace of learning can be adjusted individually.
 Figure 4: Information in one box |
A fourth reason is the availability of various sources of information in one box.
When the learner tries to do an exercise s/he can consult e.g. an electronic dictionary, the program's help-files, or the
↗WWW.
The information can also be conveyed through different media (graphics, audio ...).
Everything is coded digitally in one box.
 Figure 5: Hypertext |
Finally the linear structure of printed material is lifted and material can be presented in a multi-level design, i.e. hypertext. Again this allows for an individual learning strategy according to the desired content and for a certain amount of knowledge discovery.
These advantages are the ones which concern the language learner independent of issues of computational linguistics (
↗CL).
However from a different perspective, there are some reasons why methods of computational linguistics can be applied to
↗CALL, which is then called intelligent computer-assisted language learning (
↗ICALL).
One main advantage is, that the language domain in language learning is usually limited.
The idea is that at least in beginners to medium language classes, language usage is restricted enough to handle it with the current state-of-the-art language technology.
Additionally to the limitation of the language, the world-knowledge domain can also be easily restricted with the design of the exercise.
As with the language a limitation in world knowledge increases the chances that the system will perform as expected by the learner.
Finally the field of learner language can be seen as a challenge to CL.
A lot of commercial CALL programs lack a lot of functionality and questionable types of exercises, as will be seen.
Language technology might be able to improve the functionality and provide more and suitable information about the language either entered by the learner or contained in the program.
Since much of the research in language technology is concerned with the analysis of language (vs. the generation), this might also be applied in ICALL-systems.
Analysing language input which is probably in some way erroneous makes an interesting aspect for error recognition applications.
For these reasons it might be interesting to choose CALL as a field of application.