Deep learning-based patent value evaluation for power grid enterprises

doi:10.11772/j.issn.1001-9081.2025070850

Journal of Computer Applications ›› 2026, Vol. 46 ›› Issue (5): 1468-1474.DOI: 10.11772/j.issn.1001-9081.2025070850

• Artificial intelligence • Previous Articles

Deep learning-based patent value evaluation for power grid enterprises

Xing SHENG¹, Sunxian WENG², Kuosong CHEN², Zhongping WANG², Ruifeng REN³, Yong LIU³()

^1.State Grid Corporation of China，Beijing 100031，China
^2.Electric Power Research Institute，State Grid Fujian Electric Power Company Limited，Fuzhou Fujian 350007，China
^3.Gaoling School of Artificial Intelligence，Renmin University of China，Beijing 100872，China

Received:2025-07-29 Revised:2025-09-10 Accepted:2025-09-12 Online:2025-09-25 Published:2026-05-10
Contact: Yong LIU
About author:SHENG Xing， born in 1984. M. S.， senior engineer. His research interests include science and technology innovation policies and paradigms， intellectual property operation and achievement transformation.
WENG Sunxian， born in 1987， Ph. D.， senior engineer. His research interests include power grid environmental protection， intellectual property operation and achievement transformation.
CHEN Kuosong， born in 1982， Ph. D.， senior economist. His research interests include transformation of scientific and technological achievements， intellectual property management.
WANG Zhongping， born in 1986， M. S.， engineer. His research interests include patent value evaluation， high-value patent cultivation.
REN Ruifeng， born in 2001， Ph. D. candidate. His research interests include machine learning algorithms， mechanistic analysis of large language models.
Supported by:
Science and Technology Project of State Grid Corporation of China Headquarters(1400-202421297A-1-1-ZN)

基于深度学习的电网企业专利价值评估

盛兴¹, 翁孙贤², 陈扩松², 王忠平², 任芮锋³, 刘勇³()

^1.国家电网有限公司，北京 100031
^2.国网福建省电力有限公司电力科学研究院，福州 350007
^3.中国人民大学高瓴人工智能学院，北京 100872

通讯作者: 刘勇
作者简介:盛兴（1984—），男，内蒙古通辽人，高级工程师，硕士，主要研究方向：科技创新政策与范式、知识产权运营与成果转化
翁孙贤（1987—），男，福建福州人，高级工程师，博士，主要研究方向：电网环保、知识产权运营与成果转化
陈扩松（1982—），男，福建福州人，高级经济师，博士，主要研究方向：科技成果转化、知识产权管理
王忠平（1986—），男，福建南平人，工程师，硕士，主要研究方向：专利价值评估、高价值专利培育
任芮锋（2001—），男，山西霍州人，博士研究生，主要研究方向：机器学习算法、大语言模型机理分析
基金资助:
国家电网有限公司总部科技项目(1400-202421297A-1-1-ZN)

Abstract

Abstract:

Patent value evaluation is a crucial tool for optimizing resource allocation and guiding intellectual property strategy decisions. However， traditional manual evaluation methods are limited by subjective expert experience and low evaluation efficiency， making it difficult for enterprises in the digital economy era to meet the large-scale demand for patent value evaluation. In recent years， machine learning technologies， with their powerful high-dimensional feature extraction capabilities， have provided a feasible technological approach to innovating patent value evaluation paradigms. However， existing studies mainly focus on small-scale models within single technical dimensions and do not fully explore the potential of Large Language Models （LLMs） for quantifying value indicators. Moreover， current methods struggle to handle cases where some patent data indicators are missing. To address the challenges of processing unstructured textual data and incomplete patent value indicators in the patent database of power grid enterprises， a deep learning-based patent value evaluation method for power grid enterprises was proposed. It used Large Language Model （LLM） technology to process unstructured textual information in power grid enterprise patents； adopted a Semi-Supervised Learning （SSL） paradigm to expand the labeled patent database used for training； and employed ensemble learning techniques to train the model on the power grid enterprise patent database and conduct patent value evaluation. Empirical results demonstrate that the proposed method can efficiently evaluate the patent value of power grid enterprises with low evaluation error.

Key words: deep learning, patent value evaluation, Semi-Supervised Learning (SSL), Large Language Model (LLM), ensemble learning

摘要：

专利价值评估是优化资源配置、指导知识产权战略决策的重要手段，但传统人工评估方法受限于专家主观经验、评估效率低下等不足，难以适应数字经济时代企业对于专利价值评估的规模化需求。近年来，机器学习技术凭借强大的高维特征提取优势，为专利价值评估范式的革新提供了技术可行性。然而，现有研究多聚焦于单一技术维度的小规模模型，尚未充分探索大语言模型（LLM）在价值指标量化中的潜力，且现有方法难以适用于专利数据部分指标缺失的场景。针对电网企业专利数据库中的非结构化文本数据较难处理，部分专利价值指标不完整等问题，提出基于深度学习的电网企业专利价值评估方法，主要包括：利用LLM处理电网企业专利的非结构化文本信息；采取半监督学习（SSL）范式扩充用于训练的已标注专利数据库；通过集成学习方式在电网企业专利数据库上训练模型并进行专利价值评估。实证结果表明，所提方法能够高效评估电网企业专利价值，且评估误差较小。

关键词: 深度学习, 专利价值评估, 半监督学习, 大语言模型, 集成学习

CLC Number:

TP391.4

Xing SHENG, Sunxian WENG, Kuosong CHEN, Zhongping WANG, Ruifeng REN, Yong LIU. Deep learning-based patent value evaluation for power grid enterprises[J]. Journal of Computer Applications, 2026, 46(5): 1468-1474.

盛兴, 翁孙贤, 陈扩松, 王忠平, 任芮锋, 刘勇. 基于深度学习的电网企业专利价值评估[J]. 《计算机应用》唯一官方网站, 2026, 46(5): 1468-1474.

Figures/Tables 7

Fig. 1 Quantitative representation of patent data

Fig. 2 Differences between supervised learning and semi-supervised learning in patent value evaluation

Fig. 3 Schematic diagram of expanding training data using SSL method

Tab. 1 Comparison of SSL combined with MoE module and classical machine learning methods

方法	MAE	MSE	MAPE/%
RF	0.443 5	0.291 1	6.488 9
KNN	0.428 2	0.292 3	6.332 2
MLP	0.433 0	0.299 7	6.405 6
SVR	0.427 5	0.285 1	6.369 5
SSL+MoE模块	0.420 1	0.279 4	6.198 4

Tab. 2 Experimental results of SSL combined with MoE module

方法	MAE	MSE	MAPE/%
基线方法	0.433 0	0.299 7	6.405 6
+SSL	0.433 2	0.300 2	6.357 2
+MoE模块	0.420 7	0.282 1	6.212 6
+SSL与MoE模块	0.420 1	0.279 4	6.198 4

Fig. 4 Impact of different α on performance

Fig. 5 Impact of different β on performance

References 33

[1]	国网福建省电力有限公司. 一种基于深度学习的专利价值评估方法与系统：202510637850.0［P］. 2025-06-17.
	State Grid Fujian Electric Power Company Limited. A patent value evaluation method and system based on deep learning： 202510637850.0［P］. 2025-06-17.
[2]	王子焉，刘文涛，倪渊，等.专利价值评估研究综述［J］.科技管理研究，2019，39（16）：181-190.
	WANG Z Y， LIU W T， NI Y， et al. A review of patent value evaluation［J］. Science and Technology Management Research， 2019， 39（16）： 181-190.
[3]	肖国华，王春，姜禾，等.专利分析评价指标体系的设计与构建［J］.图书情报工作，2008，52（3）：96-99.
	XIAO G H， WANG C， JIANG H， et al. Design and construction of indicator system for patent analysis and evaluation［J］. Library and Information Service， 2008， 52（3）： 96-99.
[4]	万小丽，朱雪忠.专利价值的评估指标体系及模糊综合评价［J］.科研管理，2008，29（2）：185-191.
	WAN X L， ZHU X Z. The indicator system and fuzzy comprehensive evaluation of patent value［J］. Science Research Management， 2008， 29（2）： 185-191.
[5]	谢萍，王秀红，卢章平.企业专利价值评估方法及实证分析［J］. 情报杂志，2015，34（2）：93-98.
	XIE P， WANG X H， LU Z P. Evaluation method of enterprise patent value and empirical analysis［J］. Journal of Intelligence， 2015， 34（2）： 93-98.
[6]	BASS S D， KURGAN L A. Discovery of factors influencing patent value based on machine learning in patents in the field of nanotechnology［J］. Scientometrics， 2010， 82（2）： 217-241.
[7]	ERCAN S， KAYAKUTLU G. Patent value analysis using support vector machines［J］. Soft Computing， 2014， 18（2）： 313-328.
[8]	LEE C， KWON O， KIM M， et al. Early identification of emerging technologies： a machine learning approach using multiple patent indicators［J］. Technological Forecasting and Social Change， 2018， 127： 291-303.
[9]	TRAPPER A J C， TRAPPEY C V， GOVINDARAJAN U H， et al. Patent value analysis using deep learning models： the case of IoT technology mining for the manufacturing industry［J］. IEEE Transactions on Engineering Management， 2021， 68（5）： 1334-1346.
[10]	XU G， DONG F， FENG J. Mapping the technological landscape of emerging industry value chain through a patent lens： an integrated framework with deep learning［J］. IEEE Transactions on Engineering Management， 2022， 69（6）： 3367-3378.
[11]	XIAO S， YAN J， LI C， et al. On modeling and predicting individual paper citation count over time［C］// Proceedings of the 25th International Joint Conference on Artificial Intelligence. Palo Alto： AAAI Press， 2016： 2676-2682.
[12]	LIU X， YAN J， XIAO S， et al. On predictive patent valuation： forecasting patent citations and their types［C］// Proceedings of the 31st AAAI Conference on Artificial Intelligence. Palo Alto： AAAI Press， 2017： 1438-1444.
[13]	赵蕴华，张静，李岩，等.基于机器学习的专利价值评估方法研究［J］.情报科学，2013，31（12）：15-18.
	ZHAO Y H， ZHANG J， LI Y， et al. Study on evaluation for patent value based on machine learning［J］. Information Science， 2013， 31（12）： 15-18.
[14]	冯岭，彭智勇，刘斌，等.一种基于潜在引用网络的专利价值评估方法［J］.计算机研究与发展，2015，52（3）：649-660.
	FENG L， PENG Z Y， LIU B， et al. A latent-citation-network based patent value evaluation method［J］. Journal of Computer Research and Development， 2015， 52（3）： 649-660.
[15]	林弘杰.基于深度学习的专利价值评估方法研究［D］.合肥：中国科学技术大学，2018：37-48.
	LIN H J. A study of patent quality evaluation based on deep learning models［D］. Hefei： University of Science and Technology of China， 2018： 37-48.
[16]	王思培，韩涛.基于随机森林算法的潜在高价值专利预测方法研究［J］.情报科学，2020，38（5）：6.
	WANG S P， HAN T. Research on prediction method of potential high-value patents based on random forest algorithm［J］. Information Science， 2020， 38（5）： 6.
[17]	杨树婷.基于人工智能的专利价值评估［D］.北京：中央财经大学，2022：62-67.
	YANG S T. Patent valuation based on artificial intelligence［D］. Beijing： Central University of Finance and Economics， 2022： 62-67.
[18]	VASWANI A， SHAZEER N， PARMAR N， et al. Attention is all you need［C］// Proceedings of the 31st International Conference on Neural Information Processing Systems. Red Hook： Curran Associates Inc.， 2017： 6000-6010.
[19]	KOROTEEV M V. BERT： a review of applications in natural language processing and understanding［EB/OL］. ［2025-01-19］..
[20]	BROWN T B， MANN B， RYDER N， et al. Language models are few-shot learners［C］// Proceedings of the 34th International Conference on Neural Information Processing Systems. Red Hook： Curran Associates Inc.， 2020： 1877-1901.
[21]	DONG Q， LI L， DAI D， et al. A survey on in-context learning［C］// Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing. Stroudsburg： ACL， 2024： 1107-1128.
[22]	VAN ENGELEN J E， HOOS H H. A survey on semi-supervised learning［J］. Machine Learning， 2020， 109（2）： 373-440.
[23]	OUALI Y， HUDELOT C， TAMI M. An overview of deep semi-supervised learning［EB/OL］. ［2025-04-02］..
[24]	XUE F， ZHENG Z， FU Y， et al. OpenMoE： an early effort on open mixture-of-experts language models［C］// Proceedings of the 41st International Conference on Machine Learning. New York： JMLR.org， 2024： 55625-55655.
[25]	FEDUS W， ZOPH B， SHAZEER N. Switch Transformers： scaling to trillion parameter models with simple and efficient sparsity［J］. Journal of Machine Learning Research， 2022， 23： 1-39.
[26]	CAI W， JIANG J， WANG F， et al. A survey on mixture of experts in large language models［J］. IEEE Transactions on Knowledge and Data Engineering， 2025， 37（7）： 3896-3915.
[27]	Team GLM. ChatGLM： a family of large language models from GLM-130B to GLM-4 all tools［EB/OL］. ［2025-05-11］..
[28]	XU M. Text2vec： text to vector［CP/OL］. ［2025-04-20］..
[29]	KINGMA D P， BA J L. Adam： a method for stochastic optimization［EB/OL］. ［2025-03-20］..
[30]	BREIMAN L. Random forests［J］. Machine Learning， 2001， 45（1）： 5-32.
[31]	GUO G， WANG H， BELL D， et al. KNN model-based approach in classification［C］// Proceedings of the 2003 OTM Confederated International Conferences “On the Move to Meaningful Internet Systems”， LNCS 2888. Berlin： Springer， 2003： 986-996.
[32]	MURTAGH F. Multilayer perceptrons for classification and regression［J］. Neurocomputing， 1991， 2（5/6）： 183-197.
[33]	AWAD M， KHANNA R. Support vector regression［M］// Efficient learning machines： theories， concepts， and applications for engineers and system designers. Berkeley： Apress， 2015： 67-80.

Deep learning-based patent value evaluation for power grid enterprises

基于深度学习的电网企业专利价值评估

RichHTML

PDF

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 7

References 33

Related Articles 15

Recommended Articles

Metrics

[1]	Xingcan LI, Lizhong DING, Junyu ZHANG, Chunhui ZHANG. ContraStacker： an ensemble approach for extremely imbalanced fraud detection [J]. Journal of Computer Applications, 2026, 46(5): 1363-1369.
[2]	Jiali ZHENG, Gang ZHOU, Jing CHEN, Shunhang LI. Adaptive multi-feature fusion detection method for AI-generated text [J]. Journal of Computer Applications, 2026, 46(5): 1433-1440.
[3]	Qianfei WANG, Yang LI, Deyu LI, Suge WANG. Dual-channel feature fusion representation method for short-text clustering based on large language model [J]. Journal of Computer Applications, 2026, 46(5): 1441-1449.
[4]	Haoxuan CHEN, Peichang YE, Lei LIU, Chengming LIU, Wenhua HU. Survey of automated code edit suggestion [J]. Journal of Computer Applications, 2026, 46(4): 1227-1237.
[5]	Xiaoyu WANG, Xin LI, Di XUE, Zhangtao JIANG, Wei WANG, Yanjun XIAO. Vulnerability classification framework for video surveillance network security based on large language models [J]. Journal of Computer Applications, 2026, 46(4): 1158-1170.
[6]	Kaizhou SHI, Xuan HE, Guoyi HOU, Gen LI, Shuanggao LI, Xiang HUANG. Airborne product metrological traceability knowledge graph construction method based on large language models [J]. Journal of Computer Applications, 2026, 46(4): 1086-1095.
[7]	Chuandong QIN, Zhiqiang SUO. Skin cancer classification integrating improved ResNet50 with ensemble classifier [J]. Journal of Computer Applications, 2026, 46(4): 1354-1362.
[8]	Jian ZHANG, Jianbo YU, Jian TANG. Municipal solid waste incineration state recognition method based on multilayer preprocessing [J]. Journal of Computer Applications, 2026, 46(3): 940-949.
[9]	Haoyang ZHANG, Liping ZHANG, Sheng YAN, Na LI, Xuefei ZHANG. Review of large language model methods for knowledge graph completion [J]. Journal of Computer Applications, 2026, 46(3): 683-695.
[10]	Yongwei JIANG, Xiaoqing CHEN, Linjie FU. Elastic medical image registration model with high-frequency preservation based on spectrum decomposition [J]. Journal of Computer Applications, 2026, 46(3): 924-932.
[11]	Enkang XI, Jing FAN, Yadong JIN, Hua DONG, Hao YU, Yihang SUN. Review of threats faced by federated learning in privacy and security field [J]. Journal of Computer Applications, 2026, 46(3): 798-808.
[12]	Yiming HUANG, Xihua ZOU, Guo DENG, Di ZHENG. Pre-answering and retrieval filtering： dual-stage optimization method for RAG-based question-answering systems [J]. Journal of Computer Applications, 2026, 46(3): 696-707.
[13]	Songsen YU, Huang HE, Guopeng XUE, Hengtuo CUI. Quantitation and grading method for ceramic tile chromatic aberration based on improved fractal encoding network [J]. Journal of Computer Applications, 2026, 46(3): 959-968.
[14]	Dingjia WU, Zhe CUI. MG-SQL： SQL generation framework with enhanced schema linking and multi-generator collaboration [J]. Journal of Computer Applications, 2026, 46(3): 723-731.
[15]	Rilong WANG, Zhenping LI, Xiaosong LI, Qiang GAO, Ya HE, Yong ZHONG, Yingxiao ZHAO. Multi-Agent collaborative knowledge reasoning framework [J]. Journal of Computer Applications, 2026, 46(3): 708-714.