A Study Of Patent Quality Valuation Based On Deep Learning Models

Intellectual property refers to the achievement of intellectual creative labor,and it is a lawful right that intellectual workers enjoy legally according to their results.With the economic globalization and the establishment and development of the knowledge economy,all countries,companies and even individuals have greatly strengthened the protection of intellectual property rights,especially the protection of patents.In the era of large data,many researchers have invested in patent mining related fields,different from traditional text data,and patent document data have a well defined structure,in-cluding the front page,detailed description,declaration,appended drawings,and so on.However,patent documents are often tedious and are filled with a variety of obscure ter-minology,which requires a patent value appraiser to have a multidisciplinary expertise and need to spend a lot of manpower to analyze it.Patenting is of significant importance to protect intellectual properties for individuals,organizations and companies.One of practical demands is to automatically evaluate the quality of new patents,i.e.,patent valuation,which can be used for patent indemnification and patent portfolio.However,to solve this problem,most traditional methods just conducted simple statistical analy-ses based on patent citation networks,while ignoring much crucial information,such as patent text materials and many other useful attributes.To that end,in this dissertation,we propose a Deep Learning based Patent Quality Valuation(DLPQV)model which can integrate the above information to evaluate the quality of patents.It consists of two parts:Attribute Network Embedding(ANE)and Attention-based Convolutional Neu-ral Network(ACNN).ANE learns the patent embedding from citation networks and attributes,and ACNN extracts the semantic representation from patent text materials.Then their outputs are concatenated to predict the quality of new patents.The exper-imental results on a real-world patent dataset show our method outperforms baselines significantly with respect to patent valuation.