With the coming of information era,IT technology has been applied to all walks of life quickly.It has become a necessary ability for a modern enterprise to survive and develop.But many problems are still need to be solved in the process of software project management.Especially,the features of large investment and high uncertainty of the software project has lead to high risk of software project.Thus,it is a key problem to manage the risk for the software project.So far,how to improve the effectiveness of the software project risk management is still a difficult problem.Risk exists in the whole process of the project construction.How to identify,control and avoid the risk effectively,to a great extent,affects the success of the project.Among them,risk identification has decisive influence on the project.On the basis of existing research,this thesis made further research and discussion on the theory and methods of project risk identification.Ontology theory is applied to the area of software project risk management.This thesis established a model of software project risk identification based on ontology,with its advantage in aspect of knowledge representation.This thesis put forward an ontology modeling method suitable for this area,using the “seven step method” as the guiding ideology,and combining with the characteristic of this field.First of all,a method is put forward to do the concept acquisition combining literature statistics method,semantic analysis method and expert method because there is no authority term vocabulary can be used in this area so far.And Semantic similarity computation was made for the statistical results in order to avoid the problem of heterogeneous with the word in the process of literature statistics,and synonyms was excluded in order to improve the comprehensiveness and accuracy of concept extraction.Secondly,this thesis made the model respectively from two process of the project management and systems development in view of the specific characteristics of software project risk,such as large uncertainty,large range,abstract concepts,the absence of classification basis and so on.The sub ontology is built according to each knowledge module in the process firstly.Then the final ontology model is got by using the methods of database model integration.At last,it was implemented through the protégé 2000.Finally,a software project risk query system is designed and implemented,and it also verified the practicability and utility of the system.The model based on ontology we constructed is a helpful supplement of the methods for risk identification and it will be certain guidingsignificance for the practice of software project management. |