This paper compares the goals and implementation effects of the red tourism policy, thus helping public policy makers to allocate and optimize tourism resources. Meanwhile, it is of great significance to help market participants to adjust the development and innovation of red tourism products. Specifically, we firstly integrate government service texts and Sina Weibo texts with the theme of "red tourism" as well as relevant statistical data to form a panel dataset of 31 provincial-level regions in China from 2011 to 2019. Then, the text semantic analysis is carried out to identify policy gap. Furthermore, the spatial weight matrices are built from the perspective of policy spatial relevance. Finally, the spatial panel Durbin model is constructed to estimate the spatial spillover effect of red tourism policy on online attention from the perspective of policy spatial correlation. The findings are as follows. (1) The policy gap exists in the development of red tourism to some extent. (2) The public online attention to the red tourism is significantly dependent on geographic and policy space. The main reason for its spillover effect is that government services attach great importance to the goal of spreading the red culture, but the stability of the policy is negatively correlated with the online attention. The unstable competition and cooperation relationship among governments and the low-level steady state of the policy development environment are possible reasons that cause the public to pay less attention to red tourism than expected. (3) Under the effect of geographical and policy spatial correlation, the regional public service capacity presents a positive spillover effect, while the information variables show a negative spillover effect. Strengthening regional cooperation and integration of information service and other elements is the direction of red tourism policy innovation.

The development efficiency of red tourism represents the high-quality development level of red tourism. Taking the red tourism region in Northern and Western Guizhou province as the research object, we used DEA model and Malmquist index to measure the red tourism development efficiency of 17 counties in the study area from 2014 to 2019, and applied the kernel density function and multiple linear regression model to analyze the characteristics of its spatial evolution and key influencing factors. The research found that: (1) The overall efficiency of red tourism development in the study area is low. Thanks to the technological progress, the development efficiency of red tourism in most of the counties (cities, districts) is steadily improving. (2) The spatial distribution of red tourism development efficiency has the characteristics of agglomeration and heterogeneity, showing an "M" bimodal pattern as a whole, and the distribution of red tourism development efficiency varies greatly among counties (districts and cities). (3) Technological innovation, the advantage of red tourism resources, and the convenience of transportation are the main factors that affect the development efficiency of red tourism in the region. The economic intervention from the government to the tourism, the level of economic development, and the scale of talents have little impact on the efficiency of red tourism development in the region. The positive impact of structure on improving the efficiency of red tourism development in the region has not yet been highlighted.

Red tourism, as an important way to inherit revolutionary culture, can promote the patriotism education publicly and improve the economic development of old revolutionary areas. This paper studies the spatial distribution and heterogeneity of influencing factors of 763 scenic spots in 300 published red tourism classic attractions of China, by constructing a theoretical framework system, spatial analysis methods and regression models, so as to put forward the development mode and countermeasures. The results show that: (1) The spatial distribution of classic red tourism scenic spots is affected by four subsystems: resources, environment, society and economy. (2) In terms of spatial distribution, the core areas are more dispersed compared with other tourist attractions, which presents a pattern of seven first-class core areas, seven second-class core areas, two contiguous second-class core areas, low-density, lower-density areas and so on. The spatial correlations are quite different, and the correlation effect weakens with the reduction of spatial scale. (3) The influencing factors of classic red tourism scenic spots are different in different regions. The influence degree is red tourism resources > altitude > policy intensity > slope > transportation facilities > market demand > resource linkage > consumption level. (4) In the future, the construction of classic red tourism scenic spots should focus on the development mode and countermeasures of network, node, corridor, symbiosis linkage and cross-regional coordination. The conclusions can provide scientific decision-making basis for the red tourism development, management and cross-regional cooperation.