Invention Grant
US08504998B2 Extracting training scenarios from test cases for user-interface component recognition 有权
从用户界面组件识别的测试用例中提取训练场景

Extracting training scenarios from test cases for user-interface component recognition
Abstract:
In one embodiment, a method includes accessing a set of use cases of a software application and selecting, from the set of use cases, a subset of use cases that has a goodness-to-cost ratio that satisfies a predetermined requirement. Computing the goodness-to-cost ratio for the subset of use cases includes: for each use case from the subset of use cases, computing a second cost as a sum of the first costs associated with the user actions of the use case; computing a third cost as a sum of the second costs; for each user-interface component that is acted upon by at least one of the user actions of at least one of the use cases from the subset of use cases, computing a first goodness measure that indicates a level of user-action coverage that the subset of use cases provides on the user-interface component; computing a second goodness measure as a sum of the first goodness measures; and computing the goodness-to-cost ratio as a ratio between the second goodness measure and the third cost.
Information query
Patent Agency Ranking
0/0