Scholarlink

Background <p>HPV vaccination remains the most effective strategy for cervical cancer prevention, yet hesitancy impedes global uptake. Given the rapidly expanding and fragmented literature, traditional reviews struggle to capture dynamic thematic shifts. This study utilized a machine-learning-enhanced bibliometric approach to map the knowledge domain and identify evolutionary trends in HPV vaccine hesitancy research.</p> Methods <p>We analyzed publications retrieved from the Web of Science Core Collection (WoSCC) between 2011 and 2025. VOSviewer, CiteSpace, and Bibliometrix were employed to construct co-authorship and citation networks. Furthermore, we applied Latent Dirichlet Allocation (LDA), a probabilistic topic modeling algorithm, to conduct an unsupervised analysis of abstract texts, enabling the detection of latent semantic themes and their temporal evolution.</p> Results <p>A total of 711 publications were identified, showing an exponential growth trajectory since 2017. The United States contributed over half of the global output (53.4%), revealing a geographical imbalance compared to low- and middle-income countries (LMICs). LDA modeling unveiled three distinct thematic clusters: (1) determinants of adolescent vaccination and parental decision-making; (2) public health strategies for improving uptake and knowledge; and (3) the disrupting impact of social media and COVID-19-related misinformation. While early research focused on “access” and “safety”, post-2020 topics have heavily shifted towards “infodemics” and “trust”.</p> Conclusion <p>The research landscape on HPV vaccine hesitancy was characterized by significant geographic disparities, with high-income countries dominating the discourse despite the higher disease burden in LMICs. The application of LDA revealed a stagnation in traditional barrier studies and a critical pivot towards digital information ecosystems. These findings highlighted a disconnect between scholarly output and practical vaccination coverage, underscoring the need for research to pivot from descriptive surveys to intervention-based studies in underrepresented regions.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Mapping the global research landscape on HPV vaccine hesitancy: a machine-learning based bibliometric analysis

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年2月16日

查看全文
  • Jialao Ma,
  • Peiyan Deng,
  • Bowen Lin,
  • Pincheng Luo,
  • Wei Wu,
  • Sijia Liu

摘要

Background

HPV vaccination remains the most effective strategy for cervical cancer prevention, yet hesitancy impedes global uptake. Given the rapidly expanding and fragmented literature, traditional reviews struggle to capture dynamic thematic shifts. This study utilized a machine-learning-enhanced bibliometric approach to map the knowledge domain and identify evolutionary trends in HPV vaccine hesitancy research.

Methods

We analyzed publications retrieved from the Web of Science Core Collection (WoSCC) between 2011 and 2025. VOSviewer, CiteSpace, and Bibliometrix were employed to construct co-authorship and citation networks. Furthermore, we applied Latent Dirichlet Allocation (LDA), a probabilistic topic modeling algorithm, to conduct an unsupervised analysis of abstract texts, enabling the detection of latent semantic themes and their temporal evolution.

Results

A total of 711 publications were identified, showing an exponential growth trajectory since 2017. The United States contributed over half of the global output (53.4%), revealing a geographical imbalance compared to low- and middle-income countries (LMICs). LDA modeling unveiled three distinct thematic clusters: (1) determinants of adolescent vaccination and parental decision-making; (2) public health strategies for improving uptake and knowledge; and (3) the disrupting impact of social media and COVID-19-related misinformation. While early research focused on “access” and “safety”, post-2020 topics have heavily shifted towards “infodemics” and “trust”.

Conclusion

The research landscape on HPV vaccine hesitancy was characterized by significant geographic disparities, with high-income countries dominating the discourse despite the higher disease burden in LMICs. The application of LDA revealed a stagnation in traditional barrier studies and a critical pivot towards digital information ecosystems. These findings highlighted a disconnect between scholarly output and practical vaccination coverage, underscoring the need for research to pivot from descriptive surveys to intervention-based studies in underrepresented regions.