Article Details

Abstract

This study conducts an in-depth investigation into key issues in Guizhou Province, such as fragmented logistics information, information silos, and irrational transportation routes caused by mountainous terrain and industrial disparities. The research aims to construct a unified knowledge base in this field to address the challenges of integrating multi-source heterogeneous data. To achieve this, the study employs innovative schema mapping and knowledge graph technology to effectively establish semantic associations and integrate fragmented data, including geographical features, policy documents, and real-time logistics operations. This process results in the creation of a unified semantic knowledge base, providing a reliable data foundation for dynamic reasoning and intelligent decision-making. The research also optimizes semantic retrieval based on this domain-specific knowledge base by introducing vectorization techniques, local dictionaries, and dynamic weight adjustment methods, significantly improving the accuracy and efficiency of information retrieval. Additionally, leveraging knowledge graph technology enables precise prediction of users' latent needs. The outcomes of this study lay a solid technical foundation for building a unified logistics information platform in Guizhou Province.

Keywords

multi-source knowledge base; semantic retrieval technology; Knowledge Graph; mixed measurement method; domain dictionary construction

References

1. [1] Luo Ya, Zhang Rongxing, Xue Xixi, etc Research on Karst Landform Classification Based on Positive and Negative Topography: A Case Study of Guizhou Karst Region [J]. Journal of Guizhou Normal University (Natural Science Edition), 2024, 42 (5): 9-19
2. [2] Yang Peng, Huang Jie, Wang Jiaoe, etc Measurement of Accessibility and Fairness of Rural Logistics Terminal Distribution Points in China [J]. Journal of Geography, 2024, 79 (11): 2739-2753
3. [3] Zhu Liangbing, Liu Fade Comparison and Integration of Large Language Models and Knowledge Graph [J]. Intelligence Exploration, 2024 (12): 8-16
4. [4] Gao Jin Research on Data Conversion Method Based on RDB-RDF Pattern Mapping [D]. Shenyang: Shenyang Jianzhu University, 2024
5. [5] Chen Zhaoyi, Feng Ke, Chen Zhibin, etc Incremental Learning Raman Spectroscopy Recognition Algorithm [J]. Journal of Army Engineering University, 2023, 2 (4): 62-69
6. [6] Yang Yi The Application of Word Vectors in Linguistic Research [J]. Corpus Linguistics, 2024, 11 (2): 115-126
7. [7] Zhang Guanping Design of English Literature Retrieval Model Based on Mathematical Expressions and Document Keywords [J]. Automation Technology and Applications, 2024, 43 (12): 107-109129
8. [8] Jiang Haoda, Zhao Chunlei, Chen Han, etc A method for constructing a domain sentiment dictionary based on improved TF-IDF and BERT [J]. Computer Science, 2024, 51 (S1): 162-170
9. [9] Cheng Xinyu, Xu Juan Research on Sentiment Analysis of International Chinese MOOC Comments Based on BERT Model [J]. International Chinese Culture Research, 2024: 254-274
10. [10] Li Lu, Li Baolin, Li Lihong Fuzzy Manhattan distance weighted optimal granularity selection algorithm [J]. Journal of North China University of Science and Technology (Natural Science Edition), 2023, 45 (4): 58-65

About the authors

Zhao Tingxin (1975-), male, Gelao ethnic group, from Shiqian, Guizhou, senior engineer, master's degree in engineering, research direction: traffic informatization; Yang Bai (1988-), male, Han nationality, from Zunyi, Guizhou, intermediate engineer, bachelor's degree, research direction: artificial intelligence; Peng Chencheng (1988-), male, Han ethnicity, from Weng'an, Guizhou, undergraduate degree, research direction: transportation informatization; Corresponding author: Chen Shuang (1978-), female, Buyi ethnic group, from Guiyang, Guizhou, professor, senior engineer, master's degree, research direction: intelligent transportation, computer vision.