The invention discloses a service marking
quality optimization method based on effective human-computer interaction. The method includes the following steps: encapsulating each operation as a markingand test unit, opening an interface to the outside, inputting a parameter
list and outputting the parameter
list; marking a
list of each operational parameter, each parameter associated with a candidate
label pool; carrying out a calling test for each operation, providing feedback and propagating the correct
annotation and the generated result instances of each
successful operation to the failed operation, and optimizing the
semantic annotation; recording the success ratio of each round of update, calculating the integration efficiency of automatic update, and judging the efficiency and the size of the given threshold; extracting untested operations and clustering to obtain the highest similarity operation; carrying out
manual annotation based on candidate annotations, and giving the constraint conditions of ontology concept; classifying each instance in detail; updating operations that are still failing; applying the generated correct annotations to
semantic Web services to further perform
service discovery, composition, and recommendation.