[Purpose/significance] Exploring the individual career development of scientists and the transforming laws of research topics can not only reveal the internal mechanism of the development of scientific productivity, but also help provide better policy guidance and support for the development of scientific undertakings.[Method/process] Based on the representative discipline data of natural sciences, social science, art and humanities, this article identified the career peaks of scientists. The career peak was used as the basis for dividing the academic career of scientists. The Top2Vec topic modeling method in natural language processing was used to identify research topics, and the topic similarity and topic transfer probability of the research topics at different stages of the academic career of scientists were measured.[Result/conclusion] The research results show that scientists in various disciplines generally change research topics more frequently after experiencing their career peaks, while elite scholars have more specific research topics after experiencing their career peaks.
[1] 周建中, 闫昊, 孙粒. 我国科研人员职业生涯成长轨迹与影响因素研究[J]. 科研管理, 2019, 40(10):126-141.
[2] MERTON R K. The matthew effect in science[J]. International journal of dermatology, 1968, 27(3810):56-63.
[3] LIU L, WANG Y, SINATRA R, et al. Hot streaks in artistic, cultural, and scientific careers[J]. Nature, 2018, 559(7714):396-399.
[4] 中共中央, 国务院. 关于进一步弘扬科学家精神加强作风和学风建设的意见[EB/OL].[2021-07-18]. http://www.gov.cn/zhengce/2019-06/11/content_5399239.htm.
[5] RUAN W, HOU H, HU Z. Detecting dynamics of hot topics with alluvial diagrams:a timeline visualization[J]. Journal of data and information science, 2017, 2(3):37-48.
[6] 邱均平, 余厚强. 科学家黄金年龄影响因素的综合分析[J]. 情报杂志, 2014, 33(3):11-15, 5.
[7] COLE S. Age and scientific performance[J]. American journal of sociology, 1979, 84(4):958-977.
[8] JONES B F, WEINBERG B A. Age dynamics in scientific creativity[J]. Proceedings of the national academy of sciences, 2011, 108(47):18910-18914.
[9] SIMONTON D K. Career landmarks in science:individual differences and interdisciplinary contrasts[J]. Developmental psychology, 1991, 27(1):119.
[10] SIMONTON D K. Age and outstanding achievement:what do we know after a century of research?[J]. Psychological bulletin, 1988, 104(2):251.
[11] BRODETSKY S. Newton:scientist and man[J]. Nature, 1942, 150(3816):698-699.
[12] STEPHAN P, LEVIN S. Age and the Nobel Prize revisited[J]. Scientometrics, 1993, 28(3):387-399.
[13] LI J, YIN Y, FORTUNATO S, et al. Scientific elite revisited:patterns of productivity, collaboration, authorship and impact[J]. Journal of the royal society interface, 2020, 17(165):20200135.
[14] JONES B F. The burden of knowledge and the "death of the renaissance man":is innovation getting harder?[J]. The review of economic studies, 2009, 76(1):283-317.
[15] COKOL M, IOSSIFOV I, WEINREB C, et al. Emergent behavior of growing knowledge about molecular interactions[J]. Nature biotechnology, 2005, 23(10):1243-1247.
[16] SINATRA R, DEVILLE P, SZELL M, et al. A century of physics[J]. Nature physics, 2015, 11(10):791-796.
[17] PETERSEN A M, FORTUNATO S, PAN R K, et al. Reputation and impact in academic careers[J]. Proceedings of the national academy of sciences, 2014, 111(43):15316-15321.
[18] PETERSEN A M. Quantifying the impact of weak, strong, and super ties in scientific careers[J]. Proceedings of the national academy of sciences, 2015, 112(34):e4671-e4680.
[19] 史庆伟, 乔晓东, 徐硕, 等. 作者主题演化模型及其在研究兴趣演化分析中的应用[J]. 情报学报, 2013, 32(9):912-919.
[20] 陈立雪, 郭思月, 滕广青, 等. 科研人员研究主题的聚焦与迁移研究[J]. 数字图书馆论坛, 2019(12):9-17.
[21] UZZI B, MUKHERJEE S, STRINGER M, et al. Atypical combinations and scientific impact[J]. Science, 2013, 342(6157):468-472.
[22] GUIMERA R, UZZI B, SPIRO J, et al. Team assembly mechanisms determine collaboration network structure and team performance[J]. Science, 2005, 308(5722):697-702.
[23] BOURDIEU P. The specificity of the scientific field and the social conditions of the progress of reason[J]. Social science information, 1975, 14(6):19-47.
[24] HOONLOR A, SZYMANSKI B K, ZAKI M J. Trends in computer science research[J]. Communications of the ACM, 2013, 56(10):74-83.
[25] RZHETSKY A, FOSTER J G, FOSTER I T, et al. Choosing experiments to accelerate collective discovery[J]. Proceedings of the national academy of sciences, 2015, 112(47):14569-14574.
[26] JIA T, WANG D, SZYMANSKI B K. Quantifying patterns of research-interest evolution[J]. Nature human behaviour, 2017, 1(4):78.
[27] ZENG A, SHEN Z, ZHOU J, et al. Increasing trend of scientists to switch between topics[J]. Nature communications, 2019, 10(1):1-11.
[28] HOFMANN T. Probabilistic latent semantic indexing[C]//Proceedings of the 22nd annual international ACM SIGIR conference on research and development in information retrieval. New York:ACM, 1999:50-57.
[29] BLEI D M, NG A Y, JORDAN M I. Latent dirichlet allocation[J]. Journal of machine learning research, 2003, 3(1):993-1022.
[30] ANGELOV D. Top2Vec:distributed representations of topics[EB/OL].[2021-02-18]. https://arxiv.org/pdf/2008.09470.pdf.
[31] SALTON G, YU C T. On the construction of effective vocabularies for information retrieval[J]. Acm sigplan notices, 1973, 10(1):48-60.
[32] SINATRA R, WANG D, DEVILLE P, et al. Quantifying the evolution of individual scientific impact[J]. Science, 2016, 354(6312):596.
[33] FROSCH K H. Workforce age and innovation:a literature survey[J]. International journal of management reviews, 2011, 13(4):414-430.
