Identification and Prediction of Emerging Technologies in Healthy Food Packaging under Multi-source Data Using BERTopic-SVR

ZHU Wenshuang; CHEN Hejie; LIANG Shuli

doi:10.19554/j.cnki.1001-3563.2025.23.010

PDF(1964 KB)

Packaging Engineering ›› 2025, Vol. 46 ›› Issue (23) : 96-105. DOI: 10.19554/j.cnki.1001-3563.2025.23.010

Special Topic on Open Packaging Innovation Driving the Industrialization of Healthy and Nutritional Food

Identification and Prediction of Emerging Technologies in Healthy Food Packaging under Multi-source Data Using BERTopic-SVR

ZHU Wenshuang, CHEN Hejie, LIANG Shuli^*

Author information +

History +

Abstract

The work aims to enhance the accuracy and depth of forward-looking assessment in healthy food packaging technology and clarify the corresponding innovative evolution path. By integrating multi-source data from academic publications, patents, and social media, the BERTopic model (Bidirectional encoder representations from transformers for topic modeling) was used to mine technical topics. And cutting-edge technologies were identified through quantitative indicators of "influence, growth potential, coherence, innovation, uncertainty/ambiguity". Social media sentiment analysis was further applied to evaluate social acceptance, and support vector regression (SVR) was employed to predict future development trends. Five technologies emerged as promising directions with high growth potential and market penetration prospects: bio-based biodegradable materials, plant-derived antimicrobial coatings, RFID-based nutrition tracking, multi-spectral freshness sensing, and microencapsulated targeted nutrient delivery. By incorporating market demand into technology identification and forecasting emerging trends, this research provides a novel perspective for R&D planning and strategic decision-making in healthy food packaging.

Key words

health food packaging / BERTopic model / multi-source data fusion / Support Vector Regression (SVR) / emerging technology

Cite this article

EndNote

Ris (Procite)

Bibtex

Download Citations

ZHU Wenshuang, CHEN Hejie, LIANG Shuli. Identification and Prediction of Emerging Technologies in Healthy Food Packaging under Multi-source Data Using BERTopic-SVR[J]. Packaging Engineering. 2025, 46(23): 96-105 https://doi.org/10.19554/j.cnki.1001-3563.2025.23.010

References

[1] 国务院. “健康中国2030”规划纲要[Z]. 北京: 人民出版社, 2016.
General Office of the State Council of The People's Republic of China. Outline of the Healthy China 2030 Plan[Z]. Beijing: People's Publishing House, 2016.
[2] DAY G S, SCHOEMAKER WITH ROBERT E GUNTHER P J H. Wharton on Managing Emerging Technologies[M]. New York: Wiley, 2000.
[3] ROTOLO D, HICKS D, MARTIN B R.What Is an Emerging Technology[J]. Research Policy, 2015, 44(10): 1827-1843.
[4] 李昌, 杨中楷, 董坤. 基于多维属性动态变化特征的新兴技术识别研究[J]. 情报学报, 2022, 41(5): 463-474.
LI C, YANG Z K, DONG K.Recognition of Emerging Technologies Based on Dynamic Characteristics of Multi-Dimensional Attributes[J]. Journal of the China Society for Scientific and Technical Information, 2022, 41(5): 463-474.
[5] 徐硕, 郝丽媛, 安欣. 基于药物专利信息资源的新兴技术探测研究[J]. 中国发明与专利, 2021, 18(6): 14-23.
XU S, HAO L Y, AN X.Research on Emerging Technology Detection Based on Drug Patent Information Resources[J]. China Invention & Patent, 2021, 18(6): 14-23.
[6] 周云泽, 闵超. 基于LDA模型与共享语义空间的新兴技术识别——以自动驾驶汽车为例[J]. 数据分析与知识发现, 2022(2): 55-66.
ZHOU Y Z, MIN C.Identifying Emerging Technology with LDA Model and Shared Semantic Space—Case Study of Autonomous Vehicles[J]. Data Analysis and Knowledge Discovery, 2022(2): 55-66.
[7] 王宏, 刘沁莹, 胡玉峰, 等. 基于多源数据融合的新兴技术识别方法研究[J]. 科技进步与对策, 2025, 42(5): 21-31.
WANG H, LIU Q Y, HU Y F, et al.An Approach to Identifying Emerging Technologies by Fusing Multi-Source Data[J]. Science & Technology Progress and Policy, 2025, 42(5): 21-31.
[8] 冯立杰, 秦浩, 王金凤, 等. 融合专利数据与社交媒体数据的潜在颠覆性技术识别——基于深度学习模型[J]. 情报学报, 2024, 43(2): 181-197.
FENG L J, QIN H, WANG J F, et al.A Deep Learning Approach for Identification of Potentially Disruptive Technologies by Integrating Patent Data and Social Media[J]. Journal of the China Society for Scientific and Technical Information, 2024, 43(2): 181-197.
[9] GROOTENDORST M.BERTopic: Neural Topic Modeling with a Class-based TF-IDF Procedure[C]// Proceedings of the 36th Conference on Neural Information Processing Systems, New Orleans, 2022: 352-356.
[10] 扈剑颖, 鲁莹, 李龙珠, 等. 基于BERTopic-Kano模型的老年人健康管理服务系统设计[J]. 包装工程, 2025, 46(16): 81-91.
HU J Y, LU Y, LI L Z, et al.Design of Elderly Health Management Service System Based on BERTopic-Kano Model[J]. Packaging Engineering, 2025, 46(16): 81-91.
[11] 黄欢, 王楚尧. 基于BERTopic和长短期记忆网络模型的我国生成式人工智能技术演进、发展趋势与创新生态分析[J]. 科技管理研究, 2025, 45(17): 178-190.
HUANG H, WANG C Y.Analysis of the Evolution, Development Trends, and Innovation Ecosystem of China's Artificial Intelligence Generated Content Technology Based on BERTopic and Long Short-Term Memory[J]. Science and Technology Management Research, 2025, 45(17): 178-190.
[12] DRUCKER H, BURGES C J C, KAUFMAN L, et al. Support Vector Regression Machines[C]// Advances in Neural Information Processing Systems 9, Cambridge, 1997: 155-161.
[13] 杨辉斌, 郑德仁, 王贺龙, 等. 基于GA-SVR的管网异常漏损检测[J]. 水电能源科学, 2024, 42(3): 133-136.
YANG H B, ZHENG D R, WANG H L, et al.Abnormal Leakage Detection of Pipe Network Based on GA-SVR[J]. Water Resources and Power, 2024, 42(3): 133-136.
[14] 汤繁稀, 杨钦, 高尚卿. 基于KE-PSO-SVR模型感性意向座椅设计及其适用性研究[J]. 包装工程, 2025, 46(12): 449-459.
TANG F X, YANG Q, GAO S Q.Perceptual Intention Seat Design and Its Applicability Based on KE-PSO-SVR Model[J]. Packaging Engineering, 2025, 46(12): 449-459.
[15] 陈远东, 孟辉, 李猛克, 等. 基于支持向量机的变压器碳排放预测模型[J]. 包装工程, 2024, 45(1): 254-261.
CHEN Y D, MENG H, LI M K, et al.Transformer Carbon Emission Prediction Model Based on Support Vector Machine[J]. Packaging Engineering, 2024, 45(1): 254-261.
[16] 张曦, 崔荣荣, 申欣悦, 等. 基于在线评论用户感性意象的食品包装设计智能生成[J]. 包装工程, 2025, 46(16): 332-339.
ZHANG X, CUI R R, SHEN X Y, et al.Intelligent Generation of Food Packaging Design Based on User's Perceptual Images of Online Reviews[J]. Packaging Engineering, 2025, 46(16): 332-339.
[17] 冯缨, 曹颖. 基于主题建模与情感分析的在线健康社区用户使用评价研究[J]. 图书馆学研究, 2025(8): 112-125.
FENG Y, CAO Y.Research on User Evaluation of Online Health Communities Based on Topic Modeling and Sentiment Analysis[J]. Research on Library Science, 2025(8): 112-125.
[18] LUNDBERG S M, LEE S I.A Unified Approach to Interpreting Model Predictions[J]. Advances in Neural Information Processing Systems, 2017, 30: 4765-4774.