Information Extraction;
Named Entity Recognition;
Knowledge Engineering;
Machine Learning;
Web Mining;
D O I:
10.5220/0006407002200229
中图分类号:
TP31 [计算机软件];
学科分类号:
081202 ;
0835 ;
摘要:
After the initial development and deployment, keeping software applications and their execution environments up to date comes with some challenges. System administrators have no insight into the internals of the applications running on their infrastructure, thus if an update is available for the interpreter or for a library packaged separately on which an application depends, they do not know if the new release will bring some changes that will break parts of the application. It is up to the development team to assess the changes and to support the new version. Such tasks take time to accomplish. In this paper we propose an approach consisting of automatic analysis of the applications and automatic verification if the changes in a new version of a software dependency affect them. With this solution, system administrators will have an insight and will not depend on the developers of the applications in such situations, and the latter will be able to find out faster which is the impact of the new release on their applications.