User Story Mapping Construction And Application For Human-Computer Collaboration

The development of technology and the continuous promotion of the Internet+,the digital transformation has become an urgent need of many users.Rapid economic development has also led to a shift in software from traditional basic services to professional-oriented and closer integration with industry.Agile development is a software development method that emerged with software engineering techniques,especially for responding to rapidly changing needs.Agile development focuses more on the human impact of the development process.It is believed that face-to-face discussions and close collaboration between development groups and business experts can produce deliverable versions of software faster.This,in addition to compact,self-organizing teams,can better respond to changing requirements and adapt to the way teams are organized.Agile development is centered on the changing needs of users,and user stories are the main form of documenting requirements in agile development.Therefore the management of user stories is especially important.User story mapping is a common way to manage user stories.Many dominant tools use user story mapping.It can not only organize user stories from the macro level,but also edit those from the micro level,which can help relevant personnel to keep track of development progress and assist in version planning.Once collected,users create a user story mapping by manually sorting through the relationships between stories.In practice,however,a amount of user stories are generated as the demand changes.Due to the large number and the constant changes,manually grouping and sorting the logic to create a user story mapping would be an extremely time-consuming process as well as a significant additional development effort.In this view,we propose a study on the construction and application of user story mapping for human-computer collaboration.It is mainly divided into two parts.?.User story grouping.The user story map is a two-dimensional sheet,which needs to be grouped to form a two-dimensional structure.In this part,a natural language processing based user story clustering method is proposed.We mainly do the following work.1.Proposed a user story meta-model.The syntax attribute and context attribute are added to the unified user story metamodel to help syntactic analysis and element extraction.2.Apply natural language processing on user stories.Based on the unified meta-model of user stories,we analyze the sentence structure and extract the verb-object structure of the representative task,which provides the basis for the following clustering analysis.3.The clustering algorithm of user stories based on connectivity graph clustering is proposed.The adjacency matrix is obtained by similarity calculation,then an undirected graph is constructed to achieve clustering and good results are obtained.4.For multiple datasets,we compared with various existing methods in Recall and Precision and F-score,and the results showed that our method is better than the existing methods.?.User story acceptance test case generation.After the formation of the user story map,test cases are generated for each user story in conjunction with the acceptance criteria.In this section,a framework for generating test cases based on user stories and acceptance criteria is proposed.We mainly do the following.1.We have formulated a standard expression template to express user stories and acceptance criteria,which provides an analysis basis for subsequent automatic generation of test cases.2.We introduce the Z3 solver,which considers the constraints between each other when solving the test cases and optimizes the quality of the generated test cases.Our method has a good ability to support the constraints expressed by Z3.3.The method is applied to a Web case to verify the effectiveness of the method.