World Patent Information最新文献_第3页

AI-assisted patent drafting tools: A patent landscape & future prospectives 人工智能辅助专利起草工具：专利前景与未来展望

IF 1.9 Q2 INFORMATION SCIENCE & LIBRARY SCIENCE

World Patent Information

Pub Date : 2025-12-01 Epub Date: 2025-10-07 DOI: 10.1016/j.wpi.2025.102402

Narender Yadav , Mukesh Kumar Kumawat , Gufran Ajmal , Khalid Bashir Mir

As of early 2020, the global patent landscape for patent drafting tools has undergone significant transformation, driven by technological advancements, increased automation, and the integration of artificial intelligence. This patent landscape analysis examines global patent filings on drafting tools using Lens.org and Scifinder to identify innovation and strategic opportunities amidst the increasing complexity and volume of worldwide patent filings. It highlights key trends, innovative features, claim types, leading inventors and applicants, jurisdictional distribution, and the technological characteristics of 122 relevant patent applications spanning 56 extended patent families, thereby providing a holistic view of ongoing innovations and research activities. Furthermore, whitespace analysis reveals technological gaps and underrepresented areas including infringement analysis tools, collaboration platforms, and automated patent drawing systems pointing to promising avenues for strategic innovation and market differentiation. In addition, this study reviews 41 patent drafting tools currently available worldwide, outlining their innovative features and associated patent applications. Overall, the study offer valuable insights for researchers, innovators, patent practitioners, and stakeholders, providing foundational knowledge and clear guidance for future developments and investments in the rapidly evolving field of patent drafting technologies.

截至2020年初，在技术进步、自动化程度提高和人工智能整合的推动下，专利起草工具的全球专利格局发生了重大转变。本专利格局分析使用Lens.org和Scifinder对起草工具的全球专利申请进行了分析，以在日益复杂和日益庞大的全球专利申请中识别创新和战略机会。它突出了主要趋势、创新特征、权利要求类型、主要发明人和申请人、管辖权分布以及跨越56个扩展专利家族的122项相关专利申请的技术特征，从而提供了正在进行的创新和研究活动的整体视图。此外，空白分析揭示了技术差距和代表性不足的领域，包括侵权分析工具、协作平台和自动专利绘图系统，指出了战略创新和市场差异化的有前途的途径。此外，本研究还回顾了目前世界上可用的41种专利起草工具，概述了它们的创新特征和相关的专利申请。总体而言，该研究为研究人员、创新者、专利从业者和利益相关者提供了宝贵的见解，为快速发展的专利起草技术领域的未来发展和投资提供了基础知识和明确的指导。

{"title":"AI-assisted patent drafting tools: A patent landscape & future prospectives","authors":"Narender Yadav , Mukesh Kumar Kumawat , Gufran Ajmal , Khalid Bashir Mir","doi":"10.1016/j.wpi.2025.102402","DOIUrl":"10.1016/j.wpi.2025.102402","url":null,"abstract":"<div><div>As of early 2020, the global patent landscape for patent drafting tools has undergone significant transformation, driven by technological advancements, increased automation, and the integration of artificial intelligence. This patent landscape analysis examines global patent filings on drafting tools using Lens.org and Scifinder to identify innovation and strategic opportunities amidst the increasing complexity and volume of worldwide patent filings. It highlights key trends, innovative features, claim types, leading inventors and applicants, jurisdictional distribution, and the technological characteristics of 122 relevant patent applications spanning 56 extended patent families, thereby providing a holistic view of ongoing innovations and research activities. Furthermore, whitespace analysis reveals technological gaps and underrepresented areas including infringement analysis tools, collaboration platforms, and automated patent drawing systems pointing to promising avenues for strategic innovation and market differentiation. In addition, this study reviews 41 patent drafting tools currently available worldwide, outlining their innovative features and associated patent applications. Overall, the study offer valuable insights for researchers, innovators, patent practitioners, and stakeholders, providing foundational knowledge and clear guidance for future developments and investments in the rapidly evolving field of patent drafting technologies.</div></div>","PeriodicalId":51794,"journal":{"name":"World Patent Information","volume":"83 ","pages":"Article 102402"},"PeriodicalIF":1.9,"publicationDate":"2025-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145267826","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Modularizing patent knowledge for enhanced technological impact 模块化专利知识以增强技术影响

IF 1.9 Q2 INFORMATION SCIENCE & LIBRARY SCIENCE

World Patent Information

Pub Date : 2025-12-01 Epub Date: 2025-10-23 DOI: 10.1016/j.wpi.2025.102407

L. Siddharth

Modularity is a core design principle that enables technological artefacts to remain adaptable and evolvable, while creating value, maintaining technical robustness, and protecting intellectual property. Despite the perceived advantages, modularity has mostly been theoretically argued without sufficient empirical support—primarily due to the lack of large-scale datasets of structured representations of the knowledge of technological artefacts. In this study, we leverage a recently populated dataset of over 33,800 patent knowledge graphs built by recurrently extracting facts (head entity-relationship-tail entity) from patent sentences. Considering these knowledge graphs as explicit representations of patent knowledge, we investigate the influence of modularity upon the technological impact of patents, controlling for structural and semantic variables of patent knowledge graphs. We find a consistent positive influence of modularity on technological impact—quantified by short-term (5 years) and long-term (10 years) forward citation scores. Empirically substantiating the influence of modularity argued in design theories, we develop a predictive framework combining Graph Neural Networks (GNNs) and regression models to estimate citation scores from patent knowledge graphs. Using this framework, we showcase how modifications to patent knowledge—either through re-design or re-representation—can enhance the citation scores increasingly over 5- to 10-year periods—particularly for under- or un-cited patents.

模块化是一个核心设计原则，它使技术工件能够保持适应性和进化，同时创造价值，保持技术健壮性，并保护知识产权。尽管有明显的优势，但模块化在理论上的争论大多没有足够的经验支持，主要是由于缺乏技术工件知识的结构化表示的大规模数据集。在这项研究中，我们利用了一个最近填充的超过33,800个专利知识图谱的数据集，该数据集通过从专利句子中反复提取事实（头部实体-关系实体-尾部实体）而构建。考虑到这些知识图是专利知识的显式表示，我们在控制专利知识图的结构和语义变量的情况下，研究了模块化对专利技术影响的影响。通过短期（5年）和长期（10年）前向引用得分量化，我们发现模块化对技术影响具有一致的正向影响。为了实证证明设计理论中模块化的影响，我们开发了一个结合图神经网络（GNNs）和回归模型的预测框架，以估计专利知识图谱的引用分数。利用这个框架，我们展示了对专利知识的修改——无论是通过重新设计还是重新表示——如何在5到10年的时间内不断提高被引用分数，特别是对于未被引用或未被引用的专利。

{"title":"Modularizing patent knowledge for enhanced technological impact","authors":"L. Siddharth","doi":"10.1016/j.wpi.2025.102407","DOIUrl":"10.1016/j.wpi.2025.102407","url":null,"abstract":"<div><div>Modularity is a core design principle that enables technological artefacts to remain adaptable and evolvable, while creating value, maintaining technical robustness, and protecting intellectual property. Despite the perceived advantages, modularity has mostly been theoretically argued without sufficient empirical support—primarily due to the lack of large-scale datasets of structured representations of the knowledge of technological artefacts. In this study, we leverage a recently populated dataset of over 33,800 patent knowledge graphs built by recurrently extracting facts (head entity-relationship-tail entity) from patent sentences. Considering these knowledge graphs as explicit representations of patent knowledge, we investigate the influence of modularity upon the technological impact of patents, controlling for structural and semantic variables of patent knowledge graphs. We find a consistent positive influence of modularity on technological impact—quantified by short-term (5 years) and long-term (10 years) forward citation scores. Empirically substantiating the influence of modularity argued in design theories, we develop a predictive framework combining Graph Neural Networks (GNNs) and regression models to estimate citation scores from patent knowledge graphs. Using this framework, we showcase how modifications to patent knowledge—either through re-design or re-representation—can enhance the citation scores increasingly over 5- to 10-year periods—particularly for under- or un-cited patents.</div></div>","PeriodicalId":51794,"journal":{"name":"World Patent Information","volume":"83 ","pages":"Article 102407"},"PeriodicalIF":1.9,"publicationDate":"2025-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145362679","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Generative AI-based intelligent patent summarization for intellectual property knowledge communication and cooperation 基于生成式人工智能的知识产权知识交流与合作智能专利摘要

IF 1.9 Q2 INFORMATION SCIENCE & LIBRARY SCIENCE

World Patent Information

Pub Date : 2025-12-01 Epub Date: 2025-11-14 DOI: 10.1016/j.wpi.2025.102410

Amy J.C. Trappey , Yuga Y.C. Lin , Chun-Yi Wu

The substantial increase in patent applications has created notable challenges in retrieving, analyzing, and managing patent data. According to the World Intellectual Property Indicators (WIPI) report published by the World Intellectual Property Organization (WIPO) in 2024, the total number of patent applications worldwide surpassed 3.55 million in 2023. Traditional manual methods for extracting and interpreting key patented knowledge are usually time-consuming, expensive, subjective, and lack validation. To address the rise in patent filings and the growing need for effective patent knowledge management, we have developed an intelligent patent summarization system that utilizes large language model (LLM) technology to enhance the understanding and usability of patent documents. This research uses patents related to advanced vehicle-to-everything (V2X) technologies as case studies. Through empirical analysis, we show that the system can automatically condense large amounts of patent documents into concise and meaningful summaries. This intelligent patent summarization system runs efficiently on consumer-grade computers. Experimental results indicate that its semantic structure achieves nearly 90 % similarity compared to patents written by domain experts. This research aims to enhance the efficiency, accuracy, and accessibility of patent document processing, thereby significantly advancing the practical application of this technology.

专利申请的大量增加给检索、分析和管理专利数据带来了显著的挑战。根据世界知识产权组织（WIPO） 2024年发布的《世界知识产权指标》（WIPI）报告，2023年全球专利申请总量超过355万件。传统的人工提取和解释关键专利知识的方法通常耗时、昂贵、主观且缺乏验证。为了应对专利申请的增加和对有效专利知识管理的日益增长的需求，我们开发了一个智能专利摘要系统，该系统利用大语言模型（LLM）技术来提高对专利文献的理解和可用性。本研究使用与先进车联网（V2X）技术相关的专利作为案例研究。通过实证分析，我们发现该系统能够自动将大量的专利文献浓缩为简洁而有意义的摘要。这种智能专利摘要系统在消费级计算机上高效运行。实验结果表明，其语义结构与领域专家撰写的专利相似度接近90%。本研究旨在提高专利文献处理的效率、准确性和可及性，从而显著推进该技术的实际应用。

{"title":"Generative AI-based intelligent patent summarization for intellectual property knowledge communication and cooperation","authors":"Amy J.C. Trappey , Yuga Y.C. Lin , Chun-Yi Wu","doi":"10.1016/j.wpi.2025.102410","DOIUrl":"10.1016/j.wpi.2025.102410","url":null,"abstract":"<div><div>The substantial increase in patent applications has created notable challenges in retrieving, analyzing, and managing patent data. According to the World Intellectual Property Indicators (WIPI) report published by the World Intellectual Property Organization (WIPO) in 2024, the total number of patent applications worldwide surpassed 3.55 million in 2023. Traditional manual methods for extracting and interpreting key patented knowledge are usually time-consuming, expensive, subjective, and lack validation. To address the rise in patent filings and the growing need for effective patent knowledge management, we have developed an intelligent patent summarization system that utilizes large language model (LLM) technology to enhance the understanding and usability of patent documents. This research uses patents related to advanced vehicle-to-everything (V2X) technologies as case studies. Through empirical analysis, we show that the system can automatically condense large amounts of patent documents into concise and meaningful summaries. This intelligent patent summarization system runs efficiently on consumer-grade computers. Experimental results indicate that its semantic structure achieves nearly 90 % similarity compared to patents written by domain experts. This research aims to enhance the efficiency, accuracy, and accessibility of patent document processing, thereby significantly advancing the practical application of this technology.</div></div>","PeriodicalId":51794,"journal":{"name":"World Patent Information","volume":"83 ","pages":"Article 102410"},"PeriodicalIF":1.9,"publicationDate":"2025-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145528012","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

EigenPatent: A novel eigenvector-based ranking method for enhanced patent similarity detection 特征专利：一种基于特征向量的新型专利相似度排序方法

IF 1.9 Q2 INFORMATION SCIENCE & LIBRARY SCIENCE

World Patent Information

Pub Date : 2025-12-01 Epub Date: 2025-11-13 DOI: 10.1016/j.wpi.2025.102411

Ioannis Pontikis, Chen Li, Dimitrios Chrysostomou

Patent retrieval often faces unique challenges due to the complex structure and nature of technical documents. Traditional similarity measures often fail to capture the nuanced semantic relationships between inventions, marking it difficult for non-experts to retrieve relevant prior art. This study introduces a novel eigenvector-based ranking methodology for patent similarity that significantly outperforms traditional embedding approaches. We integrate covariance-matrix analysis with hyperplane projections, to capture both semantic and structural relationships between technical documents. Experiments within the F03D patent subclass demonstrate our approach achieves a similarity score of 83.37, substantially outperforming Word2Vec (23.3), ELMo (21.1), and SimCSE (44.8). This work addresses critical challenges in patent retrieval while introducing innovations applicable to broader technical document similarity tasks, enabling non-experts to efficiently identify relevant prior work without specialized knowledge of patent systems.

由于技术文献的复杂结构和性质，专利检索往往面临着独特的挑战。传统的相似性度量常常不能捕捉到发明之间细微的语义关系，这使得非专家很难检索相关的现有技术。本研究引入了一种新的基于特征向量的专利相似度排序方法，该方法显著优于传统的嵌入方法。我们将协方差矩阵分析与超平面投影相结合，以捕获技术文档之间的语义和结构关系。在F03D专利子类中的实验表明，我们的方法实现了83.37的相似度得分，大大优于Word2Vec (23.3), ELMo（21.1）和SimCSE（44.8）。这项工作解决了专利检索中的关键挑战，同时引入了适用于更广泛的技术文档相似性任务的创新，使非专家能够在没有专利系统专业知识的情况下有效地识别相关的先前工作。

引用次数: 0

New patent text similarity methods with a comprehensive understanding of SAO semantics 全面理解SAO语义的新专利文本相似方法

IF 1.9 Q2 INFORMATION SCIENCE & LIBRARY SCIENCE

World Patent Information

Pub Date : 2025-12-01 Epub Date: 2025-10-11 DOI: 10.1016/j.wpi.2025.102403

Nan Wang , Ziyi Wan , Hongyu Zhao , Yingtong Hu

Patent text similarity measurement is recognized as a critical component of semantic search, due diligence, infringement detection, and litigation in intellectual property management. With the continued growth in global patent filings, conventional keyword-, citation-, and classification-based approaches have been shown to inadequately capture the contextual semantics inherent in patent documents. Subject–Action–Object (SAO) structures provide a promising semantic representation; however, their effectiveness has been limited by the scarcity of specialized Semantic Text Similarity (STS) datasets and the lack of comprehensive evaluations. In this study, a novel and comprehensive framework for patent text similarity leveraging SAO semantics is proposed. Specialized patent STS datasets were constructed from USPTO examination decisions and PTAB appeal documents, comprising a 2-point scale similarity dataset and a ranking dataset for retrieval evaluation—the first openly available benchmarks of this kind. The framework integrates multiple SAO extraction techniques, novel weighting strategies including clustering-based methods, and a variety of similarity computation approaches ranging from lexical to deep learning models. Experimental evaluations show that the proposed SAO-based framework improves retrieval accuracy by 43 % over keyword-based baselines and by 26 % over standard document embedding methods. Vector-based similarity algorithms incorporating K-means clustering weights achieved a 32 % improvement over unweighted baselines. The vector-based similarity algorithm combined with K-means clustering weights improved by 32 % compared to the unweighted baseline, while the knowledge-based similarity threshold of 0.4–0.6 achieved the maximum distinction between similar and dissimilar patents. A systematic ablation analysis identified the optimal configuration as combining SAO embeddings derived from pre-trained patent vectors with clustering-based weighting, similarity thresholds, and semantic knowledge extensions. This configuration yielded superior performance in litigation support, infringement detection, and patent retrieval, reducing the average ranking position of relevant patents from 5.7 to 2.7 and achieving top-3 retrieval in all test cases.

专利文本相似度测量被认为是知识产权管理中语义搜索、尽职调查、侵权检测和诉讼的关键组成部分。随着全球专利申请的持续增长，传统的基于关键字、引文和分类的方法已被证明不能充分捕捉专利文献中固有的上下文语义。主体-动作-对象（SAO）结构提供了一种很有前途的语义表示；然而，由于缺乏专门的语义文本相似度（STS）数据集和缺乏全面的评估，它们的有效性受到限制。本文提出了一种利用SAO语义的专利文本相似度分析框架。专门的专利STS数据集是根据USPTO审查决定和PTAB上诉文件构建的，包括一个2分制的相似性数据集和一个用于检索评估的排名数据集——这是此类公开可用的基准。该框架集成了多种SAO提取技术、新的加权策略（包括基于聚类的方法）以及从词汇模型到深度学习模型的各种相似度计算方法。实验评估表明，该框架比基于关键字的基线检索精度提高了43%，比标准文档嵌入方法提高了26%。结合K-means聚类权重的基于向量的相似性算法比未加权的基线提高了32%。与未加权基线相比，基于向量的相似度算法结合K-means聚类权重提高了32%，而基于知识的相似度阈值为0.4-0.6，实现了相似和不相似专利的最大区分。系统的烧消分析确定了最优配置，即将基于预训练专利向量的SAO嵌入与基于聚类的加权、相似阈值和语义知识扩展相结合。这种配置在诉讼支持、侵权检测和专利检索方面产生了卓越的性能，将相关专利的平均排名从5.7降至2.7，并在所有测试用例中实现了前3名的检索。

{"title":"New patent text similarity methods with a comprehensive understanding of SAO semantics","authors":"Nan Wang , Ziyi Wan , Hongyu Zhao , Yingtong Hu","doi":"10.1016/j.wpi.2025.102403","DOIUrl":"10.1016/j.wpi.2025.102403","url":null,"abstract":"<div><div>Patent text similarity measurement is recognized as a critical component of semantic search, due diligence, infringement detection, and litigation in intellectual property management. With the continued growth in global patent filings, conventional keyword-, citation-, and classification-based approaches have been shown to inadequately capture the contextual semantics inherent in patent documents. Subject–Action–Object (SAO) structures provide a promising semantic representation; however, their effectiveness has been limited by the scarcity of specialized Semantic Text Similarity (STS) datasets and the lack of comprehensive evaluations. In this study, a novel and comprehensive framework for patent text similarity leveraging SAO semantics is proposed. Specialized patent STS datasets were constructed from USPTO examination decisions and PTAB appeal documents, comprising a 2-point scale similarity dataset and a ranking dataset for retrieval evaluation—the first openly available benchmarks of this kind. The framework integrates multiple SAO extraction techniques, novel weighting strategies including clustering-based methods, and a variety of similarity computation approaches ranging from lexical to deep learning models. Experimental evaluations show that the proposed SAO-based framework improves retrieval accuracy by 43 % over keyword-based baselines and by 26 % over standard document embedding methods. Vector-based similarity algorithms incorporating K-means clustering weights achieved a 32 % improvement over unweighted baselines. The vector-based similarity algorithm combined with K-means clustering weights improved by 32 % compared to the unweighted baseline, while the knowledge-based similarity threshold of 0.4–0.6 achieved the maximum distinction between similar and dissimilar patents. A systematic ablation analysis identified the optimal configuration as combining SAO embeddings derived from pre-trained patent vectors with clustering-based weighting, similarity thresholds, and semantic knowledge extensions. This configuration yielded superior performance in litigation support, infringement detection, and patent retrieval, reducing the average ranking position of relevant patents from 5.7 to 2.7 and achieving top-3 retrieval in all test cases.</div></div>","PeriodicalId":51794,"journal":{"name":"World Patent Information","volume":"83 ","pages":"Article 102403"},"PeriodicalIF":1.9,"publicationDate":"2025-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145267828","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Generating patent claims with semantic novelty 产生具有语义新颖性的专利权利要求

IF 1.9 Q2 INFORMATION SCIENCE & LIBRARY SCIENCE

World Patent Information

Pub Date : 2025-12-01 Epub Date: 2025-10-15 DOI: 10.1016/j.wpi.2025.102404

Jieh-Sheng Lee

This manuscript represents an initial step to explore the potential of leveraging generative AI to drive AI-assisted inventions. From a technical standpoint, large language models are transforming industries and driving innovation, with the patent domain being no exception. From a domain perspective, the USPTO’s recent inventorship guidance highlights that AI-assisted inventions are not categorically unpatentable, opening an exciting new frontier for inventors, patent professionals, and computer scientists. The goal of this research is to generate patent claims that exhibit a higher degree of novelty. Central to this study is the investigation of whether reinforcement learning can facilitate the generation of patent claims with such novelty. In patent law, a patent is granted when it fulfills various legal requirements, such as novelty, nonobviousness, utility, written description, and subject matter eligibility. This manuscript focuses on addressing the novelty requirement by employing reinforcement learning to generate patent claim text with a higher degree of “semantic novelty.” The semantic novelty is regarded as inversely proportional to sentence similarity, which is measured by sentence embeddings. Semantic novelty serves as a computational metric to approximate the concept of novelty as understood in patent law. In pursuit of empirical investigation, this study seeks to generate dependent claims with a higher degree of novelty relative to the independent claims, and vice versa. While the experiments presented in this manuscript are preliminary and not comprehensive, they demonstrate the efficacy of reinforcement learning and the model’s capacity to generate novel ideas, underscoring the potential of this research direction for AI-assisted inventions in the future.

这份手稿代表了探索利用生成式人工智能驱动人工智能辅助发明的潜力的第一步。从技术角度来看，大型语言模型正在改变行业并推动创新，专利领域也不例外。从领域的角度来看，美国专利商标局最近的发明指导强调，人工智能辅助的发明并不是绝对不可专利的，这为发明家、专利专业人士和计算机科学家开辟了一个令人兴奋的新领域。本研究的目的是产生具有较高新颖性的专利权利要求书。本研究的核心是调查强化学习是否可以促进具有这种新颖性的专利权利要求的产生。在专利法中，当专利满足各种法律要求，如新颖性、非显而易见性、实用性、书面描述和主题资格时，专利就被授予。本文的重点是通过使用强化学习来生成具有更高程度“语义新颖性”的专利权利要求文本，从而解决新颖性要求。语义新颖性与句子相似度成反比，句子相似度是通过句子嵌入来衡量的。语义新颖性作为一种计算度量来近似专利法中新颖性的概念。为了追求实证调查，本研究试图产生相对于独立主张具有更高新颖性的依赖主张，反之亦然。虽然本文中提出的实验是初步的，并不全面，但它们证明了强化学习的有效性和模型产生新想法的能力，强调了这一研究方向在未来人工智能辅助发明方面的潜力。

{"title":"Generating patent claims with semantic novelty","authors":"Jieh-Sheng Lee","doi":"10.1016/j.wpi.2025.102404","DOIUrl":"10.1016/j.wpi.2025.102404","url":null,"abstract":"<div><div>This manuscript represents an initial step to explore the potential of leveraging generative AI to drive AI-assisted inventions. From a technical standpoint, large language models are transforming industries and driving innovation, with the patent domain being no exception. From a domain perspective, the USPTO’s recent inventorship guidance highlights that AI-assisted inventions are not categorically unpatentable, opening an exciting new frontier for inventors, patent professionals, and computer scientists. The goal of this research is to generate patent claims that exhibit a higher degree of novelty. Central to this study is the investigation of whether reinforcement learning can facilitate the generation of patent claims with such novelty. In patent law, a patent is granted when it fulfills various legal requirements, such as novelty, nonobviousness, utility, written description, and subject matter eligibility. This manuscript focuses on addressing the novelty requirement by employing reinforcement learning to generate patent claim text with a higher degree of “semantic novelty.” The semantic novelty is regarded as inversely proportional to sentence similarity, which is measured by sentence embeddings. Semantic novelty serves as a computational metric to approximate the concept of novelty as understood in patent law. In pursuit of empirical investigation, this study seeks to generate dependent claims with a higher degree of novelty relative to the independent claims, and vice versa. While the experiments presented in this manuscript are preliminary and not comprehensive, they demonstrate the efficacy of reinforcement learning and the model’s capacity to generate novel ideas, underscoring the potential of this research direction for AI-assisted inventions in the future.</div></div>","PeriodicalId":51794,"journal":{"name":"World Patent Information","volume":"83 ","pages":"Article 102404"},"PeriodicalIF":1.9,"publicationDate":"2025-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145333156","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Towards a new paradigm for patent experimentation: WPI+ 迈向专利实验的新范式：WPI+

IF 1.9 Q2 INFORMATION SCIENCE & LIBRARY SCIENCE

World Patent Information

Pub Date : 2025-12-01 Epub Date: 2025-09-29 DOI: 10.1016/j.wpi.2025.102389

Michail Salampasis , Eleni Kamateri , Vasileios Stamatis , Mihai Lupu , Allan Hanbury , Florina Piroi

We enhance the WPI patent research collection, which is publicly accessible and free of charge, to facilitate more comparable, transparent, and reproducible experiments. This is accomplished through what we call “soft standardization” advocating the adoption of consistent methods in using the test collection. We offer data statistics, predefined collection subsets, ground-truth data for additional tasks, and open-source tools for using the collection, all on a public GitHub repository. These resources not only relieve researchers from performing essential collection analysis tasks but also implicitly guide them toward sound methods for conducting experiments with the collection. Our initiative is primarily motivated by the goal of enhancing comparability and reproducibility of patent research. This is achieved through the development of a carefully designed resource that will be continuously expanded and maintained. Our work is also driven by the observation that highly integrated Information Retrieval experiment platforms for large scale evaluation are not widely adopted by researchers. We provide examples of how the WPI+ resource/collection can be used for research on multiple patent specific tasks, including prior-art search, patent classification, and summarization. Overall, our work shows that the traditional concept of a test collection—limited to just a corpus, topics, and relevance assessments—can be broadened to support more efficient and reliable scientific experimentation.

我们加强了WPI专利研究收集，该收集是公开和免费的，以促进更具可比性、透明度和可重复性的实验。这是通过我们所谓的“软标准化”来实现的，提倡在使用测试集合时采用一致的方法。我们提供数据统计，预定义的集合子集，额外任务的真实数据，以及用于使用集合的开源工具，所有这些都在公共GitHub存储库中。这些资源不仅使研究人员从执行基本的收集分析任务中解脱出来，而且还隐含地指导他们采用合理的方法进行收集实验。我们的倡议主要是为了提高专利研究的可比性和可重复性。这是通过开发一个精心设计的资源来实现的，这个资源将不断扩大和维护。我们的工作也是由于观察到用于大规模评估的高度集成的信息检索实验平台并未被研究人员广泛采用。我们提供了如何将WPI+资源/集合用于多个专利特定任务的研究的示例，包括现有技术搜索、专利分类和摘要。总的来说，我们的工作表明，传统的测试集合概念——仅限于语料库、主题和相关评估——可以被扩展，以支持更有效和可靠的科学实验。

{"title":"Towards a new paradigm for patent experimentation: WPI+","authors":"Michail Salampasis , Eleni Kamateri , Vasileios Stamatis , Mihai Lupu , Allan Hanbury , Florina Piroi","doi":"10.1016/j.wpi.2025.102389","DOIUrl":"10.1016/j.wpi.2025.102389","url":null,"abstract":"<div><div>We enhance the WPI patent research collection, which is publicly accessible and free of charge, to facilitate more comparable, transparent, and reproducible experiments. This is accomplished through what we call “soft standardization” advocating the adoption of consistent methods in using the test collection. We offer data statistics, predefined collection subsets, ground-truth data for additional tasks, and open-source tools for using the collection, all on a public GitHub repository. These resources not only relieve researchers from performing essential collection analysis tasks but also implicitly guide them toward sound methods for conducting experiments with the collection. Our initiative is primarily motivated by the goal of enhancing comparability and reproducibility of patent research. This is achieved through the development of a carefully designed resource that will be continuously expanded and maintained. Our work is also driven by the observation that highly integrated Information Retrieval experiment platforms for large scale evaluation are not widely adopted by researchers. We provide examples of how the WPI+ resource/collection can be used for research on multiple patent specific tasks, including prior-art search, patent classification, and summarization. Overall, our work shows that the traditional concept of a test collection—limited to just a corpus, topics, and relevance assessments—can be broadened to support more efficient and reliable scientific experimentation.</div></div>","PeriodicalId":51794,"journal":{"name":"World Patent Information","volume":"83 ","pages":"Article 102389"},"PeriodicalIF":1.9,"publicationDate":"2025-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145227418","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Innovation trends and evolutionary paths of electrocatalytic hydrogen evolution reaction technology: A global patent analysis 电催化析氢反应技术的创新趋势与演进路径：全球专利分析

IF 1.9 Q2 INFORMATION SCIENCE & LIBRARY SCIENCE

World Patent Information

Pub Date : 2025-12-01 Epub Date: 2025-09-04 DOI: 10.1016/j.wpi.2025.102388

Wenting Jin , Ji Huang

Electrocatalytic hydrogen evolution reaction has emerged as a key driver of technological innovation and industrial advancement in the hydrogen energy sector. By conducting statistical analysis on patent information in this technology field, we can effectively grasp the trends and directions of technological research and development (R&D), thereby providing a critical basis for scientific policy making and industrial deployment strategies in related fields. Based on search results from the IncoPat database, this study integrates text mining with KeyBERT algorithm, CiteSpace visualization analytics, and Logistic model to conduct a comprehensive investigation from multiple dimensions including patent quantity and quality, R&D hotspots and frontiers, as well as technology lifecycle. The results indicate that: (1) The patented technologies in this field predominantly originate from core innovation clusters in China, Japan, the United States, and South Korea. China maintains an unequivocal dominance in the volume of technological outputs and has made effective strides in catching up with developed countries in terms of patent quality. However, the industrial application of Chinese patents may encounter certain difficulties. In contrast, the technological innovations of the United States and Japan maintain comparative advantages in terms of global influence and market presence. (2) The R&D hotspots in this field are concentrated primarily on topics such as precious metal-based catalysts and transition metal-based catalysts. (3) The evolutionary trajectory of this technology can be delineated into three distinct phases, with each phase featuring distinct R&D focuses and mainstream paths. (4) The technology is currently in a rapid growth phase, with forecasts suggesting it will enter the technological maturity stage by 2026 and the decline stage by 2036.

电催化析氢反应已成为氢能源领域技术创新和产业进步的关键驱动力。通过对该技术领域的专利信息进行统计分析，可以有效掌握技术研发的趋势和方向，从而为相关领域的科学政策制定和产业部署战略提供重要依据。本研究以IncoPat数据库的检索结果为基础，将文本挖掘与KeyBERT算法、CiteSpace可视化分析、Logistic模型相结合，从专利数量与质量、研发热点与前沿、技术生命周期等多个维度进行综合调查。结果表明：(1)该领域的专利技术主要来源于中国、日本、美国和韩国的核心创新集群。中国在技术产出量方面保持着无可置疑的优势，在专利质量方面取得了有效进展，正在追赶发达国家。然而，中国专利的产业化应用可能会遇到一定的困难。相比之下，美国和日本的技术创新在全球影响力和市场占有率方面保持着比较优势。(2)该领域的研发热点主要集中在贵金属基催化剂和过渡金属基催化剂等课题上。(3)该技术的演进轨迹可以划分为三个不同的阶段，每个阶段都有不同的研发重点和主流路径。(4)该技术目前处于快速成长阶段，预测到2026年将进入技术成熟期，到2036年将进入衰退期。

{"title":"Innovation trends and evolutionary paths of electrocatalytic hydrogen evolution reaction technology: A global patent analysis","authors":"Wenting Jin , Ji Huang","doi":"10.1016/j.wpi.2025.102388","DOIUrl":"10.1016/j.wpi.2025.102388","url":null,"abstract":"<div><div>Electrocatalytic hydrogen evolution reaction has emerged as a key driver of technological innovation and industrial advancement in the hydrogen energy sector. By conducting statistical analysis on patent information in this technology field, we can effectively grasp the trends and directions of technological research and development (R&D), thereby providing a critical basis for scientific policy making and industrial deployment strategies in related fields. Based on search results from the IncoPat database, this study integrates text mining with KeyBERT algorithm, CiteSpace visualization analytics, and Logistic model to conduct a comprehensive investigation from multiple dimensions including patent quantity and quality, R&D hotspots and frontiers, as well as technology lifecycle. The results indicate that: (1) The patented technologies in this field predominantly originate from core innovation clusters in China, Japan, the United States, and South Korea. China maintains an unequivocal dominance in the volume of technological outputs and has made effective strides in catching up with developed countries in terms of patent quality. However, the industrial application of Chinese patents may encounter certain difficulties. In contrast, the technological innovations of the United States and Japan maintain comparative advantages in terms of global influence and market presence. (2) The R&D hotspots in this field are concentrated primarily on topics such as precious metal-based catalysts and transition metal-based catalysts. (3) The evolutionary trajectory of this technology can be delineated into three distinct phases, with each phase featuring distinct R&D focuses and mainstream paths. (4) The technology is currently in a rapid growth phase, with forecasts suggesting it will enter the technological maturity stage by 2026 and the decline stage by 2036.</div></div>","PeriodicalId":51794,"journal":{"name":"World Patent Information","volume":"83 ","pages":"Article 102388"},"PeriodicalIF":1.9,"publicationDate":"2025-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144997631","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Technological and patent landscape of biosensors for leukemia detection: Evolution, maturity, and future prospects 用于白血病检测的生物传感器的技术和专利前景：演变、成熟和未来展望

IF 1.9 Q2 INFORMATION SCIENCE & LIBRARY SCIENCE

World Patent Information

Pub Date : 2025-12-01 Epub Date: 2025-10-22 DOI: 10.1016/j.wpi.2025.102405

María C. Otálora Trujillo, Rubén Camargo-Amado, Fiderman Machuca-Martínez

The growing demand for rapid, sensitive, and accessible diagnostic tools for early leukemia detection has driven the advancement of biosensors, particularly those based on electrochemical principles. These devices stand out for their low cost, portability, and integration into point-of-care systems, making them attractive alternatives to conventional diagnostic methods. However, doubts remain about their technological maturity and long-term prospects.

This study evaluated the evolution and technological life cycle of biosensors for leukemia detection through an integrated bibliometric and patentometric analysis. A total of 1278 scientific articles indexed in Scopus and 4756 patent families retrieved from Orbit Intelligence between 2004 and 2024 were analyzed. Of these, 2332 granted and in-force patent families were considered for technological maturity and forecasting analyses. Bibliometric mapping was performed using VOSviewer, while patent classification followed International Patent Classification codes. Logistic S-curve models (Loglet Lab 4) and Yoon's parameters were applied to estimate maturity levels and future projections.

Results reveal sustained growth in publications and patents, indicating that biosensors entered an early maturity stage in 2025, with projected technological and market relevance until approximately 2050. Among detection techniques employed in electrochemical biosensors, cyclic voltammetry accounts for the largest projected number of publications and patents, while electrochemical impedance spectroscopy shows the longest technological longevity. Differential pulse voltammetry, although less represented, remains a complementary technique.

In conclusion, biosensors represent a consolidating technology with strong potential to transform leukemia diagnostics. Patentometric analysis provides strategic intelligence to guide innovation policies, strengthen intellectual property management, and support technology transfer in this developing area.

对快速、灵敏、易获得的早期白血病检测诊断工具的需求不断增长，推动了生物传感器的发展，特别是那些基于电化学原理的生物传感器。这些设备以其低成本，便携性和集成到护理点系统而脱颖而出，使其成为传统诊断方法的有吸引力的替代品。然而，人们对它们的技术成熟度和长期前景仍然心存疑虑。本研究通过综合文献计量学和专利计量学分析，评估了用于白血病检测的生物传感器的发展和技术生命周期。分析了2004 - 2024年间Scopus检索的1278篇科学论文和Orbit Intelligence检索的4756个专利族。其中，2332个已授权和有效的专利家族被考虑用于技术成熟度和预测分析。使用VOSviewer进行文献计量制图，专利分类遵循国际专利分类代码。Logistic s曲线模型（Loglet Lab 4）和Yoon的参数用于估计成熟度水平和未来预测。结果显示，出版物和专利数量持续增长，表明生物传感器在2025年进入早期成熟阶段，预计到2050年左右才具有技术和市场相关性。在电化学生物传感器中使用的检测技术中，循环伏安法预计发表和专利数量最多，而电化学阻抗谱则显示出最长的技术寿命。差分脉冲伏安法，虽然较少代表，仍然是一种补充技术。总之，生物传感器代表了一种整合技术，具有改变白血病诊断的强大潜力。专利计量分析为指导创新政策、加强知识产权管理和支持这一发展中地区的技术转让提供了战略情报。

{"title":"Technological and patent landscape of biosensors for leukemia detection: Evolution, maturity, and future prospects","authors":"María C. Otálora Trujillo, Rubén Camargo-Amado, Fiderman Machuca-Martínez","doi":"10.1016/j.wpi.2025.102405","DOIUrl":"10.1016/j.wpi.2025.102405","url":null,"abstract":"<div><div>The growing demand for rapid, sensitive, and accessible diagnostic tools for early leukemia detection has driven the advancement of biosensors, particularly those based on electrochemical principles. These devices stand out for their low cost, portability, and integration into point-of-care systems, making them attractive alternatives to conventional diagnostic methods. However, doubts remain about their technological maturity and long-term prospects.</div><div>This study evaluated the evolution and technological life cycle of biosensors for leukemia detection through an integrated bibliometric and patentometric analysis. A total of 1278 scientific articles indexed in Scopus and 4756 patent families retrieved from Orbit Intelligence between 2004 and 2024 were analyzed. Of these, 2332 granted and in-force patent families were considered for technological maturity and forecasting analyses. Bibliometric mapping was performed using VOSviewer, while patent classification followed International Patent Classification codes. Logistic S-curve models (Loglet Lab 4) and Yoon's parameters were applied to estimate maturity levels and future projections.</div><div>Results reveal sustained growth in publications and patents, indicating that biosensors entered an early maturity stage in 2025, with projected technological and market relevance until approximately 2050. Among detection techniques employed in electrochemical biosensors, cyclic voltammetry accounts for the largest projected number of publications and patents, while electrochemical impedance spectroscopy shows the longest technological longevity. Differential pulse voltammetry, although less represented, remains a complementary technique.</div><div>In conclusion, biosensors represent a consolidating technology with strong potential to transform leukemia diagnostics. Patentometric analysis provides strategic intelligence to guide innovation policies, strengthen intellectual property management, and support technology transfer in this developing area.</div></div>","PeriodicalId":51794,"journal":{"name":"World Patent Information","volume":"83 ","pages":"Article 102405"},"PeriodicalIF":1.9,"publicationDate":"2025-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145362681","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Patentability of AI and global harmonization: An analysis of the current guidelines in Brazil and the IP5 offices 人工智能的可专利性和全球协调：对巴西和五局现行指南的分析

IF 1.9 Q2 INFORMATION SCIENCE & LIBRARY SCIENCE

World Patent Information

Pub Date : 2025-12-01 Epub Date: 2025-09-26 DOI: 10.1016/j.wpi.2025.102400

Tarso Mesquita Machado , Eduardo Winter , Hernane Borges de Barros Pereira

Innovations related to artificial intelligence can impact many technological fields and several sectors from industry, including patent offices and the patent system. Major patent offices have been updating their examination guidelines to address the particularities of AI inventions, providing legal certainty and predictability to the players in the patent system. The present study has explored the responses of the IP5 patent offices and Brazil's INPI to these challenges, revealing key areas of harmonization, divergence, and areas requiring further development. The analysis shows that, while the IP5 offices have taken steps to adapt their patent guidelines to account for the unique features of AI technologies, Brazil's INPI lags behind in terms of clarity and specificity. This works also analyses the level of harmonization among the IP5, where we conclude that significant differences remain between their approaches, especially in the case of the USPTO, which continues to rely heavily on judicial interpretations. Brazil's INPI must continue to evolve its guidelines, and there is an opportunity to observe the behavior of the IP5 to incorporate best examination practices in Brazil. By aligning with international best practices and offering clear, detailed guidance, patent offices can provide the legal certainty necessary to foster sustained investment in AI, ensuring that these transformative technologies benefit both inventors and society at large.

与人工智能相关的创新可以影响许多技术领域和工业的几个部门，包括专利局和专利制度。主要专利局一直在更新其审查指南，以解决人工智能发明的特殊性，为专利制度中的参与者提供法律确定性和可预测性。本研究探讨了五国专利局和巴西国家知识产权局对这些挑战的反应，揭示了协调、分歧和需要进一步发展的关键领域。分析表明，虽然五国知识产权局已采取措施调整其专利指南，以考虑人工智能技术的独特性，但巴西的INPI在清晰度和特异性方面落后。本著作还分析了五国知识产权局之间的协调程度，我们得出结论，它们的方法之间仍然存在显著差异，特别是在美国专利商标局的情况下，它继续严重依赖司法解释。巴西的INPI必须继续发展其指导方针，并且有机会观察IP5的行为，以纳入巴西的最佳审查实践。通过与国际最佳实践保持一致并提供清晰、详细的指导，专利局可以提供必要的法律确定性，以促进对人工智能的持续投资，确保这些变革性技术使发明者和整个社会受益。

{"title":"Patentability of AI and global harmonization: An analysis of the current guidelines in Brazil and the IP5 offices","authors":"Tarso Mesquita Machado , Eduardo Winter , Hernane Borges de Barros Pereira","doi":"10.1016/j.wpi.2025.102400","DOIUrl":"10.1016/j.wpi.2025.102400","url":null,"abstract":"<div><div>Innovations related to artificial intelligence can impact many technological fields and several sectors from industry, including patent offices and the patent system. Major patent offices have been updating their examination guidelines to address the particularities of AI inventions, providing legal certainty and predictability to the players in the patent system. The present study has explored the responses of the IP5 patent offices and Brazil's INPI to these challenges, revealing key areas of harmonization, divergence, and areas requiring further development. The analysis shows that, while the IP5 offices have taken steps to adapt their patent guidelines to account for the unique features of AI technologies, Brazil's INPI lags behind in terms of clarity and specificity. This works also analyses the level of harmonization among the IP5, where we conclude that significant differences remain between their approaches, especially in the case of the USPTO, which continues to rely heavily on judicial interpretations. Brazil's INPI must continue to evolve its guidelines, and there is an opportunity to observe the behavior of the IP5 to incorporate best examination practices in Brazil. By aligning with international best practices and offering clear, detailed guidance, patent offices can provide the legal certainty necessary to foster sustained investment in AI, ensuring that these transformative technologies benefit both inventors and society at large.</div></div>","PeriodicalId":51794,"journal":{"name":"World Patent Information","volume":"83 ","pages":"Article 102400"},"PeriodicalIF":1.9,"publicationDate":"2025-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145158505","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0