Aiming at the lack of named entity corpus in the satellite domain and the low recognition performance of existing algorithms, a satellite domain entity labeling method considering fuzzy boundaries was proposed, constructed a corpus containing 8 common satellite domain entities where the granularity was finer and the coverage was wider in comparison with the existing corpora in this field. Based on this, a transfer learning and multi-network fusion satellite domain entity recognition algorithm was proposed. Algorithm used pretrained bidirectional encoder representations for transformers to smoothly transfer the semantics of the corpus for subword-level features, a BiLSTM (bi-directional long-short term memory) network for capturing contextual information to determine boundaries, and label prediction was achieved using a conditional random field as a decoder. Experimental results show that, compared with traditional models such as BiLSTM, the proposed algorithm has better recognition performance where the F₁-score in 8 entities is all above 92% and the micro-average F₁-score reaches 96.10%.

Matching in-orbit satellites with International Telecommunication Union (ITU) satellite network declaration filings is crucial for the design, selection, declaration, and coordination of satellite frequencies and orbits. Due of their low matching efficiency and high domain knowledge requirements, traditional manual matching algorithms frequently encounter difficulties. To address these issues, we propose an unsupervised prediction of matching between in-orbit satellites and satellite network filings (UPMIS) method. This method establishes a prediction indicator system and knowledge graphs for both in-orbit satellites and satellite network filings. By integrating domain knowledge and graph partitioning, we design a three-tier filtering framework comprising a time parameter module, a numerical orbit parameter module, and a character-based societal parameter module. This framework enables fast and accurate matching between in-orbit satellites and satellite network filings. Experimental results demonstrate that UPMIS achieves a H₁₀ score of 0.8542 on real datasets, outperforming other comparative models. Additionally, the average runtime reaches millisecond-level efficiency. Additionally, the trials offer helpful references for future matching relationship predictions by recommending values for parameters like filtering quantity and aggregation depth.