Research On Knowledge Acquisition From Forestry Machinery Patent And Aided Innovation Platform

The patent is not only the innovative achievements, but also is an important source of knowledge to expand the space of design knowledge and promote the product innovation. Obtaining the knowledge hidden in the patent of forestry machinery and aiding the developers to innovate could shorten the development time effectively and achieve the valuable design.This paper proposes a knowledge acquisition method for the patent of forestry machinery based on the integration of closed weighted frequent patterns and a synonym thesaurus of forestry machinery. Extract the information of characteristic components from the claims of patent through the pretreatment. In order to ensure the accuracy of knowledge acquisition results, apply the synonym thesaurus to convert the characteristic components and normalize the definition expressed. Finally, using closed weighted pattern mining algorithm based on FP-growth obtains the function and structure knowledge. Because the patent of forestry machinery contains a lot of characteristic components, we use weighted constraints to obtain the functional knowledge which has high value and use close constraints to control the number of patterns and reduce useless patterns. Experimental results show that the method proposed is more stability and reliability to acquire knowledge from the patent of forestry machinery.In order to speed up the process of innovative design and assist the developers accomplishing mechanical innovation effectively, a conceptual design method which integrates the theory of TRIZ and the functional analysis based on the knowledge acquired from the patent of forestry machinery is proposed. Through the functional analysis we can find the improving direction from the function and structure knowledge of forestry machinery. Apply the theory of contradiction matrix of TRIZ to solve the problem and achieve the conceptual design in the end.Construct the aided innovation platform of forestry machinery based on the base of functional knowledge of forestry machinery and the conceptual design method put forward before. First, the platform provides the necessary patent retrieval method for developers, such as contradiction retrieval, inventive principle retrieval and so on. Second, provides the function of knowledge query. It can feed back users the method to achieve function by the homogeneous and heterogeneous patents. Finally, accomplish the function of concept design. It can modularize the whole progress of concept design, save the users’ new design schemes and assist users to innovation design scientifically.