A Classification And Retrieval Of Mobile Platform User Interface Pattern Library

The design of the user interface based on interface pattern is an effective way to improve the quality and efficiency of interface design. In the user interface design process, UI pattern library is the key to realize the interface pattern reuse, and it is getting more attention and application of domestic and foreign academic circles. In practice, the number of interface pattern is growing, and interface pattern continues evolution, which results in interface pattern exposing, so lead to the problem of using and retrieving pattern. Hence, we should face the challenge to solve the urgent problem that how to retrieve UI pattern efficiently and accurately in large pattern library.Firstly, this paper analyzed the existing classification in the current user interface pattern library, and these method couldn’t meet the requirements of pattern developer and user well. Then we propose pattern developer category and pattern user category that can meet the user requirements efficiently. We pointed out the existed retrieve methods that cannot match the requirement of users with correctly pattern. This paper proposes a pattern retrieval method based on the word segmentation. The method mainly includes three parts: the user need in natural language as a search condition and word segmentation processing; we parse the interface pattern library with the XML description of interface pattern into pattern analytical library, which describe with natural language; then match the segmentation results with analytical library. This paper presents the concept of matching degree and similarity matching algorithm. We improve the segmentation method and the segmentation dictionary structure. Finally, we build a retrieval tool which used our method. Then we do the experiment to compare the retrieval method proposed in this paper with the traditional method. The experimental results show that the classification and retrieval method proposed in this paper is more effective than the traditional retrieval methods, and has a certain theoretical significance and practical value.