Research On Patent Mapping Technology For Product Conceptual Design

Patent information is an important factor in the phase of product conceptual design. On one hand, making full use of the knowledge embedded in patents is helpful to promote the creative activities in the phase. On the other hand, it can avoid technique redeveloping and potential patent conflicts to take patent factor into account in the conceptual design phase, which can minimize the expense.Patent mapping is an important tool of patent analysis. Patent mapping this paper referred to is a series of mapping process with patent information, especially those unstructured information, such as patent claims, details and patent illustrations.Although there are lots of research and software relative to patent mapping, most of them just focus on English invention patents, which are localized in language sorts and patent sorts. In China, the research on patent mapping has just taken the first step, and there is still no research on patent mapping aiming at conceptual design. This paper concentrates on Chinese patents and presents the technique of patent mapping for product conceptual design.In the view of the patent corpus, the main part of invention and utility model patents is text assisted with illustration. In this paper, invention and utility model patents are entitled as technique patents and Natural Language Processing technology was used to analyze them. The main part of design patents is illustration assisted with text. This paper used Content Based Image Retrieval technology in the analyzing process.The text of technique patents is usually expatiatory and hard to understand. So word processing technology was firstly investigated including methods of new words identification based on statistics and rules, words semantic measurement based on TongYiCi CiLin, and new words explanation. Consequentially, Patent Key words Map was produced with the fore mentioned technology. Then patent specifications and patent claims were structurally processed. The characteristics of the patents were extracted based on similar-semanteme-word groups, which were taken to estimate the similarity of patents. Finally, according to the judgment of "those documents that have mass of same concepts are similar", dynamic SOM model was adopted to cluster the patents.As far as patents of design, in order to mine the abundant information implicated in the patent illustrations, content based image retrieval technology was proposed to extract the characteristics of the design patent illustrations' color, texture and shape. The weights of these characteristics would be adjusted according to relevant feedback. So the similarity of two patents can be measured to carry out the patent-clustering. Patent mapping visualization and interaction technology were also discussed. This paper proposed a method of projecting the patents clustering tree along three directions to create Tree-Shaped Patent Map, Scatter Plot Patent Map and Parallel Coordinates Patent Map. Words Distribution Patent Map was also developed based on words semantic measurement and frequency. A visualization and interaction structure was set up based on Flex technology.Finally, the patent mapping system for conceptual design was developed, named GIST (GIS for patent). And some practical examples were presented such as lock and plastic stool.