In legal texts, named entity recognition (NER) is researched using deep learning models. First, the bidirectional (Bi)-long short-term memory (LSTM)-conditional random field (CRF) model for studying NER in legal texts is established. Second, different annotation methods are used to compare and analyze the entity recognition effect of the Bi-LSTM-CRF model. Finally, other objective loss functions are set to compare and analyze the entity recognition effect of the Bi-LSTM-CRF model. The research results show that the F1 value of the model trained on the word sequence labeling corpus on the named entity is 88.13%, higher than that of the word sequence labeling corpus. For the two types of entities, place names and organization names, the F1 values obtained by the Bi-LSTM-CRF model using word segmentation are 67.60% and 89.45%, respectively, higher than the F1 values obtained by the model using character segmentation. Therefore, the Bi-LSTM-CRF model using word segmentation is more suitable for recognizing extended entities. The parameter learning result using log-likelihood is better than that using the maximum interval criterion, and it is ideal for the Bi-LSTM-CRF model. This method provides ideas for the research of legal text recognition and has a particular value.

No abstract

In this paper, a nonlinear stochastic matrix approach is used to conduct an in-depth study and analysis of modern urban governance, and an urban information model is designed for practical application based on it. Random matrix theory can calculate and give the average result of all interactions contained in a complex system. By comparing the different characteristics of statistical properties of real systems and random matrices, the special nonrandom properties of real systems are deduced. The dissipative structure theory of open complex systems, the evolution of network entropy describing the robustness of complex networks, can explain the empirical results of random matrix theory from the point of view of system evolution; through the open development of fair data and prevention tools to achieve algorithm correction, promote the design optimization of intelligent models, intelligent platforms, and intelligent programs; and commit to the benign development of technology. At the institutional level, it is necessary to keep the smart city construction synchronized with the rule of law construction, with a two-pronged approach of policy incentives and legal constraints, in addition to the overall promotion of government transformation and pluralistic governance patterns. It exceeds the carrying capacity and adaptability of social systems and institutions. In fact, the productive forces with science and technology as the core are the most active forces in social change and development. The urban information model is a technical and conceptual change triggered by the wave of information in the field of urban planning and construction and urban management, which takes the 3D model as a carrier to realize digital information application and unification in the whole life cycle of territorial spatial planning, urban design, and engineering and construction projects through the simulation of spatial planning and urban design spatial elements as well as the entry, management, and application of engineering and construction project information. Digital technology is applied in the whole life cycle of land spatial planning, urban design, and engineering projects to achieve fine governance and efficient management of urban planning.

Dispatchable energy storage system (ESS) plays a critical role in the smart grid through energy shift and power support. However, it exhibits different operational strategies and economic benefits in different application scenarios due to its inherent degradation behaviour. This paper aims to explore the technical and economic feasibility of the flexible traction power supply system (FTPSS) integrating ESS and renewable energy sources (RES) based on the traction load characteristics. First, a battery degradation model applicable in its frequent charging and discharging operating conditions is derived. Then this paper develops an operational‐sizing co‐optimization framework for the ESS in the FTPSS, where the operation decisions are made considering the degradation costs varying with the sizes and energy throughput. To solve this large‐scale non‐linear intertemporal decision‐making problem, an iterative method with a linear programming (LP) core is proposed to achieve the trade‐off between computational efficiency and accuracy. The results of the extensive comparative cases show that the proposed approach can achieve approximately 10% higher economic benefits than the existing bi‐level sizing strategies for FTPSS.