The Medical Subject Headings (MeSH) thesaurus is a controlled vocabulary widely used in biomedical knowledge systems. We propose a novel framework, termed MeSHelper, that employs a dynamic knowledge graph to predict whether the MeSH main headings (MHs) will evolve while also predicting their corresponding revision type. We parsed the whole PubMed database and all MeSH releases to construct a dynamic semantic tree (DST) and a dynamic knowledge network (DKN) to characterise the evolutionary patterns of MHs and create prediction models. Our results show that DST-related features play a major role in predicting whether the MHs will be revised. Our prediction performance achieved an F1 score of 92.07%. Both DST- and DKN-related features play a crucial role in predicting which types of MHs will evolve. The prediction performance achieved a Macro-F1 score of 72.15%, a Micro-F1 score of 84.09% and a Weighted-F1 score of 84.55%. The findings of this work aid both in constructing an automatic update model for domain thesauruses and in detecting evolutionary trends of the domain knowledge system.
KastrinAHristovskiD. Disentangling the evolution of MEDLINE bibliographic database: a complex network perspective. J Biomed Inform2019; 89: 101–113.
2.
LiangZMaoJLuK, et al. Combining deep neural network and bibliometric indicator for emerging research topic prediction. Inform Process Manag2021; 58(5): 102611.
3.
YangJLuWHuJ, et al. A novel emerging topic detection method: a knowledge ecology perspective. Inform Process Manag2022; 59(2): 102843.
4.
DjebbariAKaramychevaSHoweE, et al. MeSHer: identifying biological concepts in microarray assays based on PubMed references and MeSH terms. Bioinformatics2005; 21(15): 3324–3326.
5.
MorkJDemner-FushmanDSchmidtS, et al. Recent enhancements to the NLM Medical Text Indexer. In: Proceedings of the working notes for the conference and labs of the evaluation forum (CLEF), Sheffield, 15–18 September 2014, CEUR-WS.org, pp. 1328–1336.
6.
LiuKPengSWuJ, et al. MeSHLabeler: improving the accuracy of large-scale MeSH indexing by integrating diverse evidence. Bioinformatics2015; 31(12): i339–i347.
7.
PengSYouRWangH, et al. DeepMeSH: deep semantic representation for improving large-scale MeSH indexing. Bioinformatics2016; 32(12): i70–i79.
8.
DaiSYouRLuZ, et al. FullMeSH: improving large-scale MeSH indexing with full text. Bioinformatics2020; 36(5): 1533–1541.
9.
YouRLiuYMamitsukaH, et al. BERTMeSH: deep contextual representation learning for large-scale high-performance MeSH indexing with full text. Bioinformatics2021; 37(5): 684–692.
10.
HuoCMaSLiuX. Hotness prediction of scientific topics based on a bibliographic knowledge graph. Inform Process Manag2022; 59(4): 102980.
11.
TsatsaronisGVarlamisIKanhabuaN, et al. Temporal classifiers for predicting the expansion of medical subject headings. In: Proceedings of the 14th international conference on computational linguistics and intelligent text processing (CICLing’2013) (Part I 14), Samos, 24–30 March 2013, pp. 98–113. Berlin; Heidelberg: Springer.
12.
CardosoSDPruskiCDa SilveiraM. Supporting biomedical ontology evolution by identifying outdated concepts and the required type of change. J Biomed Inform2018; 87: 1–11.
13.
NguyenHChenHChenJ, et al. Construction and evaluation of a domain-specific knowledge graph for knowledge discovery. Inf Discov Deliv2023; 51(4): 358–370.
14.
NoyNFKleinM. Ontology evolution: not the same as schema evolution. Knowl Inf Syst2004; 6: 428–440.
15.
WardhanaHAshariAKartikaA. Review of ontology evolution process. Int J Comput Appl2018; 45: 26–33.
16.
WanBDongSChuD, et al. A deep neural network model for coreference resolution in geological domain. Inform Process Manag2023; 60(3): 103268.
17.
McCrayATLeeK. Taxonomic change as a reflection of progress in a scientific discipline. In: KuppersBOHahnUArtmannS (eds) Evolution of semantic systems. Berlin; Heidelberg: Springer, 2013, pp. 189–208.
18.
GuoYTangYKaoH. Genealogical-based method for multiple ontology self-extension in MeSH. IEEE T Nanobiosci2014; 13(2): 124–130.
19.
BaloghSGZagyvaDPollnerP, et al. Time evolution of the hierarchical networks between PubMed MeSH terms. PLoS One2019; 14(8): e0220648.
20.
CardosoSDDa SilveiraMPruskiC. Construction and exploitation of an historical knowledge graph to deal with the evolution of ontologies. Knowl-Based Syst2020; 194: 105508.
21.
AnastasiosNAnastasiaKGrigoriosT, et al. What is all this new MeSH about?Int J Digit Libr2021; 22(4): 319–337.
22.
ZhangYChenHLuJ, et al. Detecting and predicting the topic change of Knowledge-based Systems: a topic-based bibliometric analysis from 1991 to 2016. Knowl-Based Syst2017; 133: 255–268.
23.
ChenCChenYHorowitzM, et al. Towards an explanatory and computational theory of scientific discovery. J Informetr2009; 3(3): 191–209.
24.
LvYDingYSongM, et al. Topology-driven trend analysis for drug discovery. J Informetr2018; 12(3): 893–905.
25.
KesslerMM. Bibliographic coupling between scientific papers. Am Doc1963; 14(1): 10–25.
26.
ZhangZLiangXZuoY, et al. Improving unsupervised keyphrase extraction by modeling hierarchical multi-granularity features. Inform Process Manag2023; 60(4): 103356.
27.
YiSChoiJ. The organization of scientific knowledge: the structural characteristics of keyword networks. Scientometrics2012; 90(3): 1015–1026.
28.
KwonY. Understanding of structural changes of keyword networks in the computer engineering field. KIPS Trans Softw Data Eng2013; 2(3): 187–194.
29.
YangHJungW-S. Structural dynamics of keyword networks: liquid crystal display and plasma display panel cases. J Eng Technol Manage2016; 40: 64–75.
30.
BehrouziSSarmoorZSHajsadeghiK, et al. Predicting scientific research trends based on link prediction in keyword networks. J Informetr2020; 14(4): 101079.
31.
ChoudhuryNFaisalFKhushiM. Mining temporal evolution of knowledge graphs and genealogical features for literature-based discovery prediction. J Informetr2020; 14(3): 101057.
32.
ChenBTsutsuiSDingY, et al. Understanding the topic evolution in a scientific domain: an exploratory study for the field of information retrieval. J Informetr2017; 11(4): 1175–1189.
33.
HuKLuoQQiK, et al. Understanding the topic evolution of scientific literatures like an evolving city: using Google Word2Vec model and spatial autocorrelation analysis. Inform Process Manag2019; 56(4): 1185–1203.
34.
GaoQLiangZWangP, et al. Potential index: revealing the future impact of research topics based on current knowledge networks. J Informetr2021; 15(3): 101165.
35.
JiSPanSCambriaE, et al. A survey on knowledge graphs: representation, acquisition, and applications. IEEE T Neur Net Lear2022; 33(2): 494–514.
36.
LipscombCE. Medical subject headings (MeSH). Bull Med Libr Assoc2000; 88(3): 265–266.
37.
FriedrichsKSprangerMPalaniappanS. Evolution of research in biomedical sciences – a network-based characterization based on PubMed. bioRxiv2020; 11: 2020–2009.
38.
LiuXWuJZhouZ. Exploratory undersampling for class-imbalance learning. IEEE T Syst Man Cy B2009; 39(2): 539–550.
BatistaGPratiRMonardM. A study of the behavior of several methods for balancing machine learning training data. ACM SIGKDD Explor Newslett2004; 6(1): 20–29.
41.
BatageljVMrvarA. A subquadratic triad census algorithm for large sparse networks with small maximum degree. Soc Networks2001; 23(3): 237–243.
42.
LundbergSLeeS. A unified approach to interpreting model predictions. Adv Neur In2017; 30: 4768–4777.
43.
HuZCuiJLinA. Identifying potentially excellent publications using a citation-based machine learning approach. Inform Process Manag2023; 60(3): 103323.
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
