Topic Evolution and Technology Opportunity Emergence in Embodied Intelligence: A Dual-Source Analysis of Publications and Patents

Yan Zeng

doi:10.54691/mnvwq956

Authors

Yan Zeng School of Economics and Management, Beijing Jiaotong University, Beijing, China

DOI:

https://doi.org/10.54691/mnvwq956

Keywords:

Embodied intelligence, dynamic LDA, topic evolution, publication-patent dual-source analysis, technology opportunity identification.

Abstract

Embodied intelligence has become a major research frontier in artificial intelligence, yet the evolution of this field has rarely been examined through the combined lens of scientific publications and technological patents. This paper introduces a Topic Evolution and Technology Opportunity (TETO) framework to investigate how research themes evolve and how technology opportunities emerge across the two knowledge systems. A Dynamic Latent Dirichlet Allocation (LDA) model with temporal dependence is employed to characterize topic evolution over time, after which publication and patent trajectories are examined comparatively. The analysis reveals a persistent divergence between academic research and technological innovation. Scientific publications place greater emphasis on cognitive mechanisms and Human-Machine interaction, whereas patents are concentrated more heavily on motion control and system integration. At the same time, several research themes exhibit convergent evolutionary patterns, reflecting increasing interaction between scientific exploration and technological development. Based on these cross-domain dynamics, technology opportunities are classified into three categories: frontier-exploration, industry-driven, and co-evolutionary. The proposed framework extends existing approaches to technology forecasting and provides a systematic basis for strengthening alignment between academic research and industrial innovation.

Downloads

Download data is not yet available.

References

[1] Honores-Marín, G., Zolfani, S. H., & Chipulu, M. (2026). From patents to predictive analytics: Leveraging R-GCNs for technological opportunity discovery in converging industries. Technological Forecasting and Social Change, 227, 124613.

[2] Gao, X., & Rai, V. (2023). Knowledge acquisition and innovation quality: The moderating role of geographical characteristics of technology. Technovation, 125, 102766.

[3] Liu, H., Guo, D., & Cangelosi, A. (2025). Embodied intelligence: A synergy of morphology, action, perception and learning. ACM Computing Surveys, 57(7), 1–36.

[4] Liu, H., Guo, D., & Huang, K. (2025). Learning for embodiment and embodiment for learning. Nature Reviews Electrical Engineering, 2, 651–653.

[5] Sutton, R. S., & Barto, A. G. (1998). Reinforcement learning: An introduction. IEEE Transactions on Neural Networks, 9(5), 1054.

[6] Kober, J., Bagnell, J. A., & Peters, J. (2013). Reinforcement learning in robotics: A survey. The International Journal of Robotics Research, 32(11), 1238–1274.

[7] Small, H. (1973). Co-citation in the scientific literature: A new measure of the relationship between two documents. Journal of the American Society for Information Science, 24(4), 265–269.

[8] Cobo, M. J., López-Herrera, A. G., Herrera-Viedma, E., & Herrera, F. (2011). Science mapping software tools: Review, analysis, and cooperative study among tools. Journal of the American Society for Information Science and Technology, 62(7), 1382–1402.

[9] Blei, D. M., Ng, A. Y., & Jordan, M. I. (2003). Latent Dirichlet allocation. Journal of Machine Learning Research, 3, 993–1022.

[10] Griffiths, T. L., & Steyvers, M. (2004). Finding scientific topics. Proceedings of the National Academy of Sciences, 101(Suppl. 1), 5228–5235.

[11] Wang, X., & McCallum, A. (2006). Topics over time: A non-Markov continuous-time model of topical trends. In Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (pp. 424–433).

[12] Callon, M., Courtial, J. P., Turner, W. A., et al. (1983). From translations to problematic networks: An introduction to co-word analysis. Social Science Information, 22(2), 191–235.

[13] Hall, D., Jurafsky, D., & Manning, C. D. (2008). Studying the history of ideas using topic models. In Proceedings of EMNLP (pp. 363–371).

[14] Wang, X., McCallum, A., & Wei, X. (2007). Topical n-grams: Phrase and topic discovery, with an application to information retrieval. In Proceedings of ICDM.

[15] Porter, A. L., & Detampel, M. J. (1995). Technology opportunities analysis. Technological Forecasting and Social Change, 49(3), 237–255.

[16] Yoon, B., & Park, Y. (2004). A text-mining-based patent network: Analytical tool for high-technology trend. Journal of High Technology Management Research, 15(1), 37–50.

[17] Lee, S., Yoon, B., & Park, Y. (2009). An approach to discovering new technology opportunities: Keyword-based patent map approach. Technovation, 29(6–7), 481–497.

[18] Narin, F., Hamilton, K. S., & Olivastro, D. (1997). The increasing linkage between U.S. technology and public science. Research Policy, 26(3), 317–330.

[19] Glänzel, W., & Meyer, M. (2003). Patents cited in the scientific literature: An exploratory study of reverse citation relations. Scientometrics, 58(2), 415–428.