Hunan Sports Bureau

This study aimed to conduct archival collection and search and rescue of traditional mountain village landscapes through the application of geographic information system (GIS) technology, addressing potential cultural heritage losses during natural disasters or human-induced events. Initially, on-site surveys were conducted to gather fundamental information about the geographical locations, natural environments, architectural structures, and cultural activities of traditional mountain villages. Subsequently, GIS technology was employed to spatially locate and digitize the collected data, constructing 3D landscape models of these villages. Building upon this foundation, a GIS-based search and rescue path planning algorithm was devised to enhance the efficiency of landscape archival retrieval during mountain disasters. On this basis, this study combined GIS-based terrain modeling with trust region policy optimization, an artificial intelligence reinforcement learning method, to achieve real-time search and rescue path planning in disaster scenarios. Compared with traditional path planning methods, the proposed model shows significant advantages in three typical mountain village scenarios: the average path length is shortened by 5 km (p < 0.01), the path coordinate deviation is controlled between 0.3 and 0.6 km, and the average search and rescue success rate is increased by 12.3% (p < 0.05). This integrated framework of a GIS and artificial intelligence has a significant effect on improving the efficiency and spatial accuracy of landscape archive retrieval and emergency response. In conclusion, this study not only provides novel technical means for the collection and preservation of landscape archives in traditional mountain villages but also highlights the significant application value of GIS technology in the field of search and rescue.

Abstract One of the important directions for the reform of ideology and politics is the integration of big data with ideological and political education. As a type of big data technology, knowledge graphs can display a clear knowledge architecture, thereby optimizing the educational impact of courses in ideology and politics. In this paper, we first enhance the CRF model for processing sequence annotation, which is based on BiLSTM. We then propose features for judging entity relationships. Next, we use the classification model to classify the relationships within it. Finally, we integrate and store the knowledge of both civics and political science to create a comprehensive civics and political science knowledge graph. We found that the experimental class, applying knowledge mapping for Civics education, increased its total Civics score by 5.88 points and outperformed the control group in the knowledge memorization and application tests by 6.33 and 3.54 points, respectively. The experimental class’s post-test score for deep teaching interest reached 22.38, surpassing the control group’s evaluation, and the surface orientation decreased by 1.91 points. Research has demonstrated the benefits of using knowledge mapping in teaching to enhance students’ civics performance and learning interest.