Semantic-Enhanced Rete Algorithm for Geological Data Quality Inspection Rule Engine

【Objective】Geological data are characterized by multi-source heterogeneity, complex semantics, strong spatial correlation and intensive professional rules, which serve as the fundamental support for national energy and resource security, ecological civilization construction, disaster prevention and mitigation, and other undertakings. At present, data are no longer merely a by-product of geological surveys but have evolved into core assets. It is particularly important to establish a scientific, rigorous and efficient data quality management system to guarantee the authenticity, accuracy, integrity and timeliness of geological survey data. Current geological data quality inspection mainly relies on fixed scripts and sampling inspection, which can hardly cope with the semantic heterogeneity and complex logical constraints of multi-source heterogeneous data. Although the classic Rete algorithm features high efficiency, it is limited to exact character matching and cannot recognize synonyms of geological terms, fuzzy descriptions or unstructured texts.【Methods】This paper proposes a geological data quality inspection rule engine based on the semantically enhanced Rete algorithm. By optimizing the classic Rete algorithm with node sharing, pre-data filtering and dynamic semantic nodes, a semantic enhanced reasoning network oriented to geological scenarios is constructed. A complete reasoning process covering semantic matching, rule activation, conflict resolution and rule execution is realized.【Results】Compared with the traditional rule engine, the proposed method achieves improvements in both recognition accuracy and recall rate, with the F1-score reaching 0.88, which verifies its superior capability in geological data quality inspection.【Conclusion】The rule engine realizes the decoupling of business logic and quality inspection rules, and significantly improves rule configurability and system maintainability. It provides an effective new technical approach for intelligent quality control of geological data, and is of great significance for promoting the high-quality development of geological informatization and strengthening the supporting capacity of geological data.