H. Yamada et al., A CONSTRUCTING METHOD OF FUNCTIONAL-MODEL BY INTEGRATED LEARNING FROMEXAMPLES OF SOFTWARE MODIFICATION, IEICE transactions on information and systems, E78D(9), 1995, pp. 1133-1141
One approach to develop software efficiently is to reuse existing soft
ware by modifying a part of it. However, modifying software will often
introduce unexpected side effects into other parts of it. As a result
, it costs much time and care to modify the software. So, in order to
modify software efficiently, we have proposed a functional model to re
present information about side effects caused by modification and a mo
del based supporting system for modifying software. So far, however, a
n expert software developer must describe the entire functional model
of the target software through the analysis of practical modifying pro
cesses. This will be an unnecessary burden on him. Moreover, the large
r target software becomes, the harder the model construction becomes.
Therefore, an automatic constructing method of the functional model is
needed in order to solve this problem. So, this paper considers a met
hod of acquiring useful interaction information by learning from train
ing examples of modification. However, in our application domain, it s
eems that it is impossible to make complete domain theory and to prepa
re a large number of training examples in advance. Therefore, our lear
ning method involves an integration of explanation-based learning (EEL
) from positive examples of modification generated by the user and Sim
ilarity-based learning (SBL) from positive or negative examples genera
ted by the user and the learning system. As a result, our method can a
cquire valid knowledge about the interaction from not so many examples
under incomplete theory. Then, this paper presents a constructing met
hod, in which our proposed learning method is incorporated, of a funct
ional model. Finally, this paper demonstrates construction of the func
tional model in the domain of an event-driven queueing simulation prog
ram according to our learning method.