[目的/意义] 通过对比分析两类主流方法（基于参考文献和基于主题内容）的有效性及局限性，为构建多维协同的知识新颖性评价框架提供参考，为知识创新价值统一度量指标的形成提供借鉴。[方法/过程] 选取B. Uzzi等提出的非传统组合方法（参考文献维度）与P. Savov的主题年份预测方法（主题内容维度），通过核密度分布、相关性分析（Spearman检验）和多元回归模型，系统验证两类方法对知识新颖性的表征能力，并结合案例解读，揭示其内在机制差异。[结果/结论] 研究发现：①核密度分布一致性以及文献摘要解读证实两种方法一定程度的有效性，与被引量相关性的年度变化和文献解读证实两种方法倾向于表征知识新颖性的不同层面；②非传统组合虽受参考文献期刊频次影响，但其识别的高创新性期刊对组合具有一定有效性；主题年份预测方法虽受年份影响较大，但相同年份的识别具有一定有效性；③通过创新值均值年度变化和不同阶段的累积密度分布进一步证实主题年份预测方法受年份影响较大；在影响因素分析中，被引量对主题年份预测方法的高正影响力（0.495）显性化该方法受年份影响的程度，同时显性化期刊数目对非传统组合方法的正影响力（0.193）。
[Purpose/Significance] This study comparatively analyzes the effectiveness and limitations of two mainstream methodologies (reference-based and topic content-based approaches) to provide insights for constructing a multi-dimensional collaborative evaluation framework for knowledge novelty and references for developing a unified measurement approach for assessing the value of knowledge innovation. [Method/Process] This study employed atypical combination approach (reference dimension) by B. Uzzi et al. and topic-year prediction method (topic content dimension) by P. Savov. Through kernel density distribution analysis, correlation analysis (Spearman’s test), and multivariate regression models, it systematically validated the representational capacity of these two methods for knowledge novelty. Case studies further elucidated their intrinsic mechanistic differences. [Result/Conclusion] Key findings are as follows. ①Consistency in kernel density distributions and case studies of literature abstracts confirm the partial validity of both methods. Annual variations in their correlations with citation counts and textual interpretations reveal their tendencies to capture distinct dimensions of knowledge novelty. ②While the atypical combination method is influenced by the citation frequency of reference sources, its identification of highly innovative sources demonstrates validity for combination analysis. The topic-year prediction method, though significantly affected by temporal factors, exhibits validity in identifying novelty within the same year. ③Annual variations in mean innovation values and cumulative density distributions across stages further corroborate the pronounced temporal bias of the topic-year prediction method. In factor analysis, citation count exerts a strong positive influence (0.495) on the topic-year prediction method, highlighting its temporal dependency, while the number of reference sources positively impacts the atypical combination method (0.193).
[1] SCHNEIDER J, COSTAS R. Identifying potential “breakthrough” publications using refined citation analyses: three related explorative approaches[J]. Journal of the Association for Information Science and Technology, 2017, 68(3): 709-723.
[2] MIN C, DING Y, LI J, et al. Innovation or imitation: the diffusion of citations[J]. Journal of the Association for Information Science and Technology, 2018, 69(10): 1271-1282.
[3] MUKHERJEE S, ROMERO D, JONES B, et al. The nearly universal link between the age of past knowledge and tomorrow’s breakthroughs in science and technology: the hotspot[J]. Science advances, 2017, 3(4): e1601315.
[4] UZZI B, MUKHERJEE S, STRINGER M, et al. Atypical combinations and scientific impact[J]. Science, 2013, 342(6157): 468-472.
[5] SHIBAYAMA S, WANG J. Measuring originality in science[J]. Scientometrics, 2020, 122(1): 409-427.
[6] LI X, WEN Y, JIANG J, et al. Identifying potential breakthrough research: a machine learning method using scientific papers and Twitter data[J]. Technological forecasting and social change, 2022, 184: 122042.
[7] YAN Y, TIAN S, ZHANG J. The impact of a paper’s new combinations and new components on its citation[J]. Scientometrics, 2020, 122(2): 895-913.
[8] HOFSTRA B, KULKARNI V, GALVEZ S, et al. The diversity-innovation paradox in science[J]. Proceedings of the National Academy of Sciences, 2020, 117(17): 9284-9291.
[9] MIN C, BU Y, SUN J. Predicting scientific breakthroughs based on knowledge structure variations[J]. Technological forecasting and social change, 2021, 164: 120502.
[10] MIN C, BU Y, WU D, et al. Identifying citation patterns of scientific breakthroughs: a perspective of dynamic citation process[J]. Information processing & management, 2021, 58(1): 102428.
[11] XU H, LUO R, WINNINK J, et al. A methodology for identifying breakthrough topics using structural entropy[J]. Information processing & management, 2022, 59(2): 102862.
[12] WANG S, MA Y, MAO J, et al. Quantifying scientific breakthroughs by a novel disruption indicator based on knowledge entities[J]. Journal of the Association for Information Science and Technology, 2023, 74(2): 150-167.
[13] 温浩, 何茜茹. 学术文摘创新点挖掘的认知分析方法[J]. 情报学报, 2021, 40(5): 489-499. (WEN H, HE Q R. Cognitive analysis method for mining innovation points in academic abstracts[J]. Journal of the China Society for Scientific and Technical Information, 2021, 40(5): 489-499.)
[14] 曹树金,闫欣阳,张倩等. 中外情报学论文创新性特征研究[J]. 图书情报工作, 2020, 64(1): 80-92. (CAO S J, YAN X Y, ZHANG Q, et al. Research on characteristics of innovation in Chinese and international academic literature of information science[J]. Library and information service, 2020, 64(1): 80-92.)
[15] SUN B, KOLESNIKOV S, GOLDSTEIN A, et al. A dynamic approach for identifying technological breakthroughs with an application in solar photovoltaics[J]. Technological forecasting and social change, 2021, 165: 120534.
[16] LUO Z, LU W, HE J, et al. Combination of research questions and methods: a new measurement of scientific novelty[J]. Journal of informetrics, 2022, 16(2): 101282.
[17] SAVOV P, JATOWT A & NIELEK R. Identifying breakthrough scientific papers[J]. Information processing & management, 2020, 57(2): 102168.
[18] HICKS D, WOUTERS P, WALTMAN L, et al. Bibliometrics: the Leiden Manifesto for research metrics[J]. Nature, 2015, 520(7548): 429-431.
[19] WU L, KITTUR A, YOUN H, et al. Metrics and mechanisms: measuring the unmeasurable in the science of science[J]. Journal of informetrics, 2022, 16(2): 101290.
[20] WAGNER C, WHETSELL T, MUKHERJEE S. International research collaboration: novelty, conventionality, and atypicality in knowledge recombination[J]. Research policy, 2019, 48(5): 1260-1270.
[21] BOUDREAU K, GUINAN E, LAKHANI K, et al. Looking across and looking beyond the knowledge frontier: intellectual distance, novelty, and resource allocation in science[J]. Management science, 2016, 62(10): 2765-2783.
[22] FONTANA M, IORI M, MONTOBBIO F, et al. New and atypical combinations: an assessment of novelty and interdisciplinarity[J]. Research policy, 2020, 49(7): 104063.
[23] BORNMANN L, TEKLES A. Convergent validity of several indicators measuring disruptiveness with milestone assignments to physics papers by experts[J]. Journal of informetrics, 2021, 15(3): 101159.
[24] ABBASIHAROFTEH M, KOGLER D, LENGYEL B. Atypical combinations of technologies in regional co-inventor networks[J]. Research policy, 2023, 52(10): 104886.
[25] MUKHERJEE S, UZZI B, JONES B, et al. A new method for identifying recombinations of existing knowledge associated with high-impact innovation[J]. Journal of product innovation management, 2016, 33(2): 224-336.
