联系我们

论文查询结果	返回搜索

论文编号：	14726
作者编号：	2120223636
上传时间：	2024/6/7 6:59:48
中文题目：	专利新颖性及其价值研究 ——基于技术认可与影响力的视角
英文题目：	Research on Patent Novelty and Value: Based on the Perspective of Technology Recognition and Influence
指导老师：	林润辉教授
中文关键字：	专利新颖性;专利价值评价;技术认可;技术影响力;BERT模型
英文关键字：	Patent Novelty;Patent Value Evaluation;Technology Recognition;Technology Influence;BERT Model
中文摘要：	随着全球科技创新竞争的加剧，发明专利新颖性的准确识别对于企业创新战略和知识产权保护至关重要。专利作为技术创新的重要载体，其新颖性评估不仅关系到企业的研发投入是否能够得到合理的回报，也是衡量研究机构创新能力的重要指标。专利新颖性的准确识别是创新研究的重要课题，它直接关系到专利质量和市场竞争力。本研究针对现有专利新颖性评估方法的不足，提出了一种基于BERT预训练语言模型的新颖性评估方法，旨在运用深度学习语言模型，对中文专利文本进行新颖性测度。通过这种方法，本文不仅能够更准确地识别出具有创新性的专利，而且还能在此基础上，进一步分析专利新颖性与其价值的关系，为企业和研究机构提供技术创新战略的理论指导，同时也为授权机构审查流程提供决策依据。 基于BERT文本分析方法构建并实施专利新颖性识别模型，并通过与传统引文网络方法的对比研究，以客观的指标验证了所构建模型的准确性。本文探讨了专利新颖性及其价值的相关关系，揭示了技术创新与技术认可和影响力互动作用，提出了创新主体核心竞争力提升的完整路径。研究发现，专利新颖性与授权概率之间存在显著的正相关性，技术新颖性与技术认可和影响力之间也存在正相关性，证实了新颖性在专利质量和市场竞争力中的重要作用。不仅为企业技术创新战略提供了理论指导，还为授权机构审查流程提供了决策依据，助力国家创新驱动和技术引领战略实施。此外，本文系统地对高新颖性专利的特征和价值进行了量化分析，揭示了高新颖性专利的重要价值。不仅为创新理论中专利创新特征识别提供了新的视角，还为企业在研发初期阶段识别高新颖性的发明提供了理论支持。
英文摘要：	As the competition in global technological innovation grows fiercer, precise identification of the novelty of invention patents becomes imperative for corporate innovation strategies and intellectual property protection. The assessment of patent novelty, as a crucial element of technological innovation, not only influences whether an enterprise can achieve a reasonable return on its R&D investment but also serves as a key indicator of its innovation capabilities. This study addresses the limitations of existing patent novelty assessment methods and proposes a novel approach based on the BERT pre-trained language model. The goal is to leverage a deep learning Chinese language model to evaluate the novelty of Chinese patent texts, analyze the correlation between patent novelty and its value from the perspective of technology recognition and influence, and identify high-novelty patents to bolster the technological innovation and market competitiveness of enterprises and research institutions. A patent novelty identification model was developed and implemented using the BERT text analysis method. Through comparative research with traditional citation network methods, the accuracy of the model was validated using objective indicators. This article delves into the relationship between patent novelty and its value, uncovering the interplay between technological innovation, technological recognition, and influence. It proposes a comprehensive approach to enhance the core competitiveness of innovative entities, offering theoretical guidance for corporate technological innovation strategies, and serving as a decision-making framework for the review process of authorized agencies. Additionally, the article conducts a systematic quantitative analysis of the characteristics and value of high-novelty patents, highlighting their significant value. It not only offers a fresh perspective for identifying patent innovation characteristics in innovation theory but also provides theoretical support for enterprises to identify high-novelty inventions in the early stages of research and development.
查看全文：	预览  下载(下载需要进行登录)