Urban green land park relative performance evaluation method and device

Abstract

The invention provides an urban green park relative performance evaluation method and device. A green park automatic classification technology is constructed based on different core functions of a green park; two sets of corresponding index systems of benefit output and resource occupation are established based on multi-modal space-time big data, the benefit output and the resource occupation of different types of green land parks are evaluated by using an analytic hierarchy process (AHP) and a cyclic random forest regression method, and the utilization efficiency of the different types of green land parks is comprehensively evaluated according to the relationship between the benefit output and the resource occupation. High-efficiency and low-efficiency green land parks are identified; a Shapley additive interpretation (SHAP) method is utilized to excavate main driving factors influencing the low-efficiency green land park, and the importance of resource occupation indexes is quantified. Classified utilization efficiency evaluation is performed on the green land park based on space-time big data and a deep learning method, a low-efficiency space is identified, a space pattern in which benefit output is not matched with resource occupation is analyzed, and key causes of the space pattern are mined.

Description

Technical Field

[0001] The present invention belongs to the technical field of urban land use efficiency evaluation, and in particular relates to a relative performance evaluation method and device for urban green park. Background Art

[0002] Providing green parks in urban areas is an important component of a healthy, livable and sustainable urban environment. It is interrelated with the city's economic development, social security, community cohesion, quality of life, and various other development issues such as environmental and climate change. However, with the rapid development of urbanization, differences in resource allocation and economic development have led to not all newly built green parks being fully utilized. Sometimes due to poor planning, insufficient market demand or poor management, some green parks cannot attract enough people and activities. This leaves some green parks unmanaged and unused, unable to realize their potential value, resulting in inefficient space utilization. Therefore, how to evaluate the utilization efficiency of urban green parks and rationally plan the allocation of spatial resources has become the key to urban governance.

[0003] Considering the important role of green parks in cities, many scholars have conducted qualitative and quantitative analysis of urban green parks from different levels and multiple perspectives. For example, they have studied the landscape planning, spatial distribution, quality evaluation and main driving factors of urban green parks; they have studied the utilization efficiency and benefits of green parks, and provided effective suggestions for the effective use of space and the main functions of green parks. They have conducted research from multiple perspectives such as society, history, transportation, economy, and population, but few have analyzed the spatial differentiation laws of all types of urban green parks from the perspective of spatial resource input and output, and evaluated the utilization efficiency of urban green parks as a whole.

[0004] In order to effectively identify inefficient urban green parks, current research is generally conducted at macro and meso scales such as cities or at small micro scales based on traditional surveys. Usually, the inefficient space is directly determined by combining weighting and threshold partitioning. With the emergence of spatiotemporal big data products, such as social media data and mobile phone signaling data, it is possible to conduct large-scale and refined analysis and research on the utilization efficiency of urban green parks. Summary of the invention

[0005] In order to overcome the problem that the above-mentioned existing urban land use efficiency evaluation methods lack large-scale and refined comprehensive evaluation or at least partially solve the above-mentioned problem, an embodiment of the present invention provides a relative performance evaluation method and device for urban green parks.

[0006] According to a first aspect of an embodiment of the present invention, a method for evaluating relative performance of an urban green park is provided, comprising the following steps:

[0007] (1) Based on the different core functions of green parks, an automatic classification technology for green parks is constructed, and green parks are divided into five types: comprehensive parks, community parks, ecological parks, urban squares, and theme spaces;

[0008] (2) Based on multimodal spatiotemporal big data, two corresponding indicator systems of benefit output and resource occupation are established to characterize the multi-dimensional real benefit output and resource input and location environment characteristics of green parks respectively. The analytic hierarchy process (AHP) is used to evaluate the benefit output of different types of green parks. The recurrent random forest regression method is used to model the correlation between the benefit output and resource occupation evaluation indicator system, classify and predict resource occupation, and comprehensively evaluate the utilization efficiency of different types of green parks based on the ratio of benefit output to resource occupation, and identify efficient and inefficient green parks.

[0009] Step (2) establishes two corresponding indicator systems for benefit output and resource occupation by characterizing the multi-dimensional real benefit output, resource input and location environment characteristics of green parks, including:

[0010] The benefit output evaluation index system is centered around the three dimensions of society, economy, and ecological environment, with a total of 7 indicators, including GDP, passenger flow, travel score, net primary productivity, PM2.5, surface temperature, and biomass;

[0011] The resource occupation evaluation index system focuses on five dimensions: land resources, transportation resources, landscape resources, environmental resources, and social resources, with a total of 22 indicators, including area, floor area ratio, accessibility to community residents, accessibility to industry, commerce and institutions, accessibility to transportation facilities, green space coverage, forest coverage, water body coverage, road density, infrastructure configuration coverage, landscape abundance, landscape dominance, leaf area index, normalized vegetation index, evapotranspiration index, average canopy height, population density, coverage of commercial service facilities, coverage of leisure and entertainment facilities, coverage of scientific and educational institutions facilities, coverage of accommodation service facilities, and coverage of tourist attractions facilities.

[0012] In step (2), the cyclic random forest regression method is used to model the correlation between benefit output and resource occupation evaluation index system. This method adds a cyclic structure of the original sample on the basis of the random forest regression model. The specific steps include:

[0013] In each cycle, each green park in each type of green park is used as a test sample for model training in a cycle, and all remaining samples in this type of green park are included in the random forest regression model for training, so that the predicted expected value is the resource occupancy value of the green park.

[0014] Step (2) quantifies the utilization efficiency of different types of green parks by considering the ratio of benefit output to resource occupation, and thereby identifies efficient and inefficient green parks;

[0015] Specifically, green parks with a severe imbalance between benefit output BO and resource occupation RO are extracted as efficient spaces or inefficient spaces, and the evaluation results are divided into three categories: 1) Inefficient space, BO is much smaller than RO, indicating that the green park unit is inefficient and has not been fully utilized; 2) Efficient space, BO is much larger than RO, indicating that the green park unit is fully and effectively utilized; 3) Normal space, BO is relatively close to RO, indicating that the utilization efficiency of spatial resources has reached a normal state and has not been overused or wasted;

[0016] Accordingly, by considering the ratio of benefit output to resource occupation, the utilization efficiency of green park UE is quantified, namely:

[0017]

[0018] Based on this, efficient space and inefficient space are extracted. The specific extraction formula is:

[0019]

[0020]

[0021]

[0022] Among them, UE is the utilization efficiency, i is the type of green park, and i represents inefficient space, normal space and efficient space respectively.

[0023] (3) The Shapley additive interpretation (SHAP) method was used to explore the main driving factors affecting inefficient green parks and quantify the importance of resource occupation indicators.

[0024] According to a second aspect of an embodiment of the present invention, a device for evaluating relative performance of an urban green park is provided, comprising:

[0025] Green park function classification module, used to automatically classify green parks according to different core function types;

[0026] The inefficient space identification module is used to establish two corresponding indicator systems for benefit output and resource occupation. The hierarchical analysis method and cyclic random forest regression are used to evaluate the benefit output and resource occupation of different types of green parks. Based on the relationship between the two, the utilization efficiency of different types of green parks is comprehensively evaluated to identify efficient and inefficient green parks.

[0027] The influencing factor mining module is used to explore the main driving factors affecting inefficient green parks and quantify the importance of resource occupation indicators.

[0028] According to a third aspect of an embodiment of the present invention, there is also provided an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor calls the program instructions to execute a relative performance evaluation method for urban green parks provided by any one of the various possible implementations of the first aspect.

[0029] According to the fourth aspect of an embodiment of the present invention, a non-transitory computer-readable storage medium is also provided, wherein the non-transitory computer-readable storage medium stores computer instructions, and the computer instructions enable the computer to execute the urban green park relative performance evaluation method provided by any possible implementation method of the various possible implementation methods of the first aspect.

[0030] The embodiment of the present invention provides a relative performance evaluation method and device for urban green parks. The method targets the problem of urban land use efficiency evaluation, combines multimodal spatiotemporal big data and machine learning methods, classifies green parks according to their core functions, and classifies and comprehensively evaluates their utilization efficiency, identifies high-efficiency and low-efficiency spaces, explores the key factors that affect the mismatch between benefit output and resource occupation, provides suggestions for the rational allocation of spatial resources and environment, and promotes sustainable development. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0032] Figure 1 A schematic diagram of the overall process of the relative performance evaluation method for urban green parks provided by an embodiment of the present invention;

[0033] Figure 2 A schematic diagram of a recurrent random forest regression model in a relative performance evaluation method for urban green parks provided in an embodiment of the present invention;

[0034] Figure 3 A schematic diagram of the overall structure of a relative performance evaluation device for urban green parks provided by an embodiment of the present invention;

[0035] Figure 4 A schematic diagram of the overall structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0036] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0037] In one embodiment of the present invention, a relative performance evaluation method for urban green parks is provided. Figure 1 The overall flow diagram of the relative performance evaluation method of urban green parks provided by the embodiment of the present invention includes:

[0038] S1, build green park automatic classification technology based on different core functions of green parks;

[0039] Among them, green parks are divided into five types according to their core functions: comprehensive parks, community parks, ecological parks, urban squares and theme spaces. Among them, comprehensive parks refer to parks with a variety of functions, aiming to provide urban residents with rich and diverse leisure and entertainment and natural and cultural experiences; community parks are parks that meet the needs of residents in a certain community to carry out daily leisure and entertainment and outdoor activities nearby; ecological parks are green parks based on ecological functions, focusing on natural ecological protection, improving the urban environment, or providing scientific education and leisure and recreation activities for urban residents; urban squares are public open activity places for urban residents to gather, relax and socialize, and their green area accounts for less than 65%; theme spaces are based on economic functions and provide public open activity spaces for urban residents to engage in sports and fitness, indoor and outdoor leisure and entertainment.

[0040] The data sets were classified using random forest classification by taking the water systems, woodlands and green spaces in the Amap AOI data and the functional area data as the original data sets and combining them with the resource occupation evaluation index system.

[0041] S2, based on multimodal spatiotemporal big data, two corresponding indicator systems of benefit output and resource occupation were established. The benefit output and resource occupation of different types of green parks were evaluated using the analytic hierarchy process (AHP) and the cyclic random forest regression method respectively. The utilization efficiency of different types of green parks was comprehensively evaluated based on the relationship between the two, and efficient and inefficient green parks were identified.

[0042] Among them, the two corresponding indicator systems of benefit output and resource occupation respectively represent the multi-dimensional real benefit output, resource input and location environment characteristics of green parks, including:

[0043] The benefit output evaluation index system is centered around the three dimensions of society, economy, and ecological environment, with a total of 7 indicators, including GDP, passenger flow, travel score, net primary productivity, PM2.5, surface temperature, and biomass;

[0044] The resource occupation evaluation index system focuses on five dimensions: land resources, transportation resources, landscape resources, environmental resources, and social resources, with a total of 22 indicators, including area, floor area ratio, accessibility to community residents, accessibility to industry, commerce and institutions, accessibility to transportation facilities, green space coverage, forest coverage, water body coverage, road density, infrastructure configuration coverage, landscape abundance, landscape dominance, leaf area index, normalized vegetation index, evapotranspiration index, average canopy height, population density, coverage of commercial service facilities, coverage of leisure and entertainment facilities, coverage of scientific and educational institutions facilities, coverage of accommodation service facilities, and coverage of tourist attractions facilities.

[0045] The calculation of the above indicators mainly uses a variety of data such as functional area data, POI data, remote sensing data, road network density data, population density data, etc., and is automatically extracted through a variety of spatial analysis methods such as buffer analysis, overlay analysis, and spatial statistical analysis.

[0046] Based on the above benefit output evaluation index system, the benefit output in three dimensions of multiple social, economic and ecological environment is calculated in combination with the analytic hierarchy process (AHP). Among them, the expert scoring method is used to compare every two influencing factors, and a relative comparison coefficient is assigned to each influencing factor to construct a judgment matrix. The consistency of the matrix is ​​verified by calculating the consistency ratio (CR) (CR<0.10 means passing the consistency test), so as to determine the weight of each evaluation index. The specific formula is as follows:

[0047]

[0048]

[0049] Where RI is the stochastic index, λ max represents the main eigenvalue and n represents the order of the matrix.

[0050] Based on this, the normalized weighted comprehensive benefit output calculation formula is constructed:

[0051] BO c =a i ×GDP index +b i ×FF index +d i ×TR index +e i ×NPP index +f i ×PM25 index +gi ×LST index +h i ×BM index (c∈type(i), 1≤i≤5)

[0052] Among them, BO represents benefit output, FF represents passenger flow, TR represents travel fraction, NPP represents net primary productivity, LST represents land surface temperature, BM represents biomass, a is the weight of GDP, b is the weight of FF, d is the weight of TR, e is the weight of NPP, f is the weight of LST, g is the weight of PM25, and h is the weight of BM.

[0053] Based on the above resource occupation evaluation index system and benefit output value, the correlation between the two is modeled using the cyclic random forest regression method. This method adds a cyclic structure of the original sample on the basis of the random forest regression model, such as Figure 2 As shown, the specific steps include:

[0054] Initially, each regression tree is trained using a bagging algorithm with random sampling and replacement. At each tree node, binary segmentation is performed based on the contribution of each index to the model. Finally, the prediction results of all trees are averaged to obtain the regression result. The cyclic random forest regression method adds a cyclic structure of the original sample on the basis of the random forest regression model. In each cycle, each green park in each type of green park is used as a test sample for model training in a cycle, and all remaining samples in this type of green park are included in the random forest regression model for training, so that the expected value predicted is the resource occupancy value of the green park. The number of cycles of this cyclic structure is determined by the number of samples of that type of green park participating in the training each time.

[0055] Based on the calculation results of the above benefit output and resource occupation, the utilization efficiency of different types of green parks is quantified, and efficient and inefficient green parks are identified accordingly.

[0056] Specifically, green parks with a severe imbalance between benefit output (BO) and resource occupation (RO) are extracted as efficient spaces or inefficient spaces, and the evaluation results are divided into three categories: 1) Inefficient space, BO is much smaller than RO, indicating that the green park unit is inefficiently utilized and has not been fully utilized; 2) Efficient space, BO is much larger than RO, indicating that the green park unit is fully and effectively utilized; 3) Normal space, BO is relatively close to RO, indicating that the utilization efficiency of spatial resources has reached a normal state and is not overused or wasted.

[0057] Accordingly, by considering the ratio of benefit output to resource occupation, the utilization efficiency (UE) of green park is quantified, namely:

[0058]

[0059] Based on this, efficient space and inefficient space are extracted. The specific extraction formula is:

[0060]

[0061]

[0062]

[0063] Among them, UE is the utilization efficiency, i is the type of green park, and represents inefficient space, normal space and efficient space respectively.

[0064] S3, using the Shapley additive explanation (SHAP) method to explore the main driving factors affecting inefficient green parks and quantify the importance of resource occupation indicators;

[0065] This example uses the SHAP method to quantify the importance of each indicator. Shapley Additive Interpretation (SHAP) is an additive feature attribution method based on cooperative game theory, which mainly uses Shapley value to determine the importance of features. The Shapley value of a feature value is its contribution to the model, and all possible combinations of feature values ​​are weighted and added:

[0066]

[0067] where represents the subset of features used in the model, represents the vector of feature values ​​for the instances to be explained, and represents the number of features. refers to the prediction made for the feature values ​​in the set, obtained by marginalizing over the features not included in the set, i.e.:

[0068]

[0069] A more reliable measure of feature importance is achieved by averaging the absolute Shapley values ​​of multiple instances, as shown below:

[0070]

[0071] This aggregation method can comprehensively evaluate the impact of each feature on model prediction, and the features with higher absolute Shapley values ​​are more important in the model prediction process.

[0072] In another embodiment of the present invention, a device for evaluating the relative performance of urban green parks is provided, which is used to implement the methods in the aforementioned embodiments. Therefore, the description and definition in the embodiments of the aforementioned methods for evaluating the relative performance of urban green parks can be used for understanding the various execution modules in the embodiments of the present invention. Figure 3The overall structure diagram of the relative performance evaluation device for urban green parks provided by an embodiment of the present invention includes a green park function classification module 301, an inefficient space identification module 302 and an influencing factor mining module 303;

[0073] Among them, the green park function classification module 301 is used to automatically classify green parks according to different core function types;

[0074] The inefficient space identification module 302 is used to establish two corresponding indicator systems for benefit output and resource occupation, and use the hierarchical analysis method and cyclic random forest regression to evaluate the benefit output and resource occupation of different types of green parks, and comprehensively evaluate the utilization efficiency of different types of green parks based on the relationship between the two, and identify efficient and inefficient green parks;

[0075] The influencing factor mining module 303 is used to mine the main driving factors affecting the inefficient green park and quantify the importance of resource occupation indicators.

[0076] This embodiment provides an electronic device, Figure 4 The overall structure diagram of the electronic device provided by the embodiment of the present invention includes: at least one processor 401, at least one memory 402 and a bus 403; wherein,

[0077] The processor 401 and the memory 402 communicate with each other via the bus 403;

[0078] The memory 402 stores program instructions that can be executed by the processor 401. The processor calls the program instructions to execute the methods provided by the above-mentioned method embodiments, for example, including: constructing an automatic classification technology for green parks based on different core functions of green parks, and dividing green parks into five types: comprehensive parks, community parks, ecological parks, urban squares and theme spaces; establishing two corresponding indicator systems of benefit output and resource occupation based on multimodal spatiotemporal big data, respectively characterizing the multi-dimensional real benefit output and resource input and location environment characteristics of green parks, using the hierarchical analysis method (AHP) to evaluate the benefit output of different types of green parks, using the cyclic random forest regression method to model the correlation between the benefit output and resource occupation evaluation indicator system, classifying and predicting resource occupation, and based on the ratio relationship between benefit output and resource occupation, comprehensively evaluating the utilization efficiency of different types of green parks, and identifying efficient and inefficient green parks; using the Shapley additive interpretation (SHAP) method to explore the main driving factors affecting inefficient green parks and quantify the importance of resource occupation indicators.

[0079] The present embodiment provides a non-transitory computer-readable storage medium, which stores computer instructions. The computer instructions enable a computer to execute the methods provided by the above-mentioned method embodiments, for example, including: constructing an automatic classification technology for green parks based on different core functions of green parks, and dividing green parks into five types: comprehensive parks, community parks, ecological parks, urban squares and theme spaces; establishing two corresponding indicator systems of benefit output and resource occupation based on multimodal spatiotemporal big data, respectively characterizing the multi-dimensional real benefit output and resource input and location environment characteristics of green parks, using the analytic hierarchy process (AHP) to evaluate the benefit output of different types of green parks, using the cyclic random forest regression method to model the correlation between the benefit output and resource occupation evaluation indicator system, classifying and predicting resource occupation, and based on the ratio relationship between benefit output and resource occupation, comprehensively evaluating the utilization efficiency of different types of green parks, and identifying efficient and inefficient green parks; using the Shapley additive interpretation (SHAP) method to explore the main driving factors affecting inefficient green parks and quantify the importance of resource occupation indicators.

[0080] A person of ordinary skill in the art can understand that all or part of the steps of implementing the above method embodiment can be completed by hardware related to program instructions, and the aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it executes the steps of the above method embodiment; and the aforementioned storage medium includes: ROM, RAM, magnetic disk or optical disk, etc., various media that can store program codes.

[0081] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0082] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0083] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A relative performance evaluation method for urban green parks, characterized in that: The following steps are involved: (1) Build an automatic classification technology for green parks based on their different core functions; (2) Based on multimodal spatiotemporal big data, two corresponding indicator systems were established: benefit output and resource occupation. The benefit output and resource occupation of different types of green parks were evaluated using the analytic hierarchy process (AHP) and the cyclic random forest regression method, respectively. Based on the relationship between the two, the utilization efficiency of different types of green parks was comprehensively evaluated to identify efficient and inefficient green parks. (3) The Shapley additive interpretation (SHAP) method was used to explore the main driving factors affecting inefficient green parks and quantify the importance of resource occupation indicators.

2. The relative performance evaluation method of urban green parks according to claim 1 is characterized in that: In step (1), the random forest classification method is used to classify green parks into five types: comprehensive parks, community parks, ecological parks, urban squares and theme spaces.

3. The relative performance evaluation method of urban green parks according to claim 1 is characterized in that: Step (2) establishes two corresponding indicator systems for benefit output and resource occupation by characterizing the multi-dimensional real benefit output, resource input and location environment characteristics of green parks, including: The benefit output evaluation index system is centered around the three dimensions of society, economy, and ecological environment, with a total of 7 indicators, including GDP, passenger flow, travel score, net primary productivity, PM2.5, surface temperature, and biomass; The resource occupation evaluation index system focuses on five dimensions: land resources, transportation resources, landscape resources, environmental resources, and social resources, with a total of 22 indicators, including area, floor area ratio, accessibility to community residents, accessibility to industry, commerce and institutions, accessibility to transportation facilities, green space coverage, forest coverage, water body coverage, road density, infrastructure configuration coverage, landscape abundance, landscape dominance, leaf area index, normalized vegetation index, evapotranspiration index, average canopy height, population density, coverage of commercial service facilities, coverage of leisure and entertainment facilities, coverage of scientific and educational institutions facilities, coverage of accommodation service facilities, and coverage of tourist attractions facilities.

4. The relative performance evaluation method of urban green parks according to claim 1 is characterized in that: In step (2), the cyclic random forest regression method is used to model the correlation between benefit output and resource occupation evaluation index system. This method adds a cyclic structure of the original sample on the basis of the random forest regression model. The specific steps include: In each cycle, each green park in each type of green park is used as a test sample for model training in a cycle, and all remaining samples in this type of green park are included in the random forest regression model for training, so that the predicted expected value is the resource occupancy value of the green park.

5. The relative performance evaluation method of urban green parks according to claim 1 is characterized in that: Step (2) quantifies the utilization efficiency of different types of green parks by considering the ratio of benefit output to resource occupation, and thereby identifies efficient and inefficient green parks; Specifically, green parks with a severe imbalance between benefit output BO and resource occupation RO are extracted as efficient spaces or inefficient spaces, and the evaluation results are divided into three categories: 1) Inefficient space, BO is much smaller than RO, indicating that the green park unit is inefficient and has not been fully utilized; 2) Efficient space, BO is much larger than RO, indicating that the green park unit is fully and effectively utilized; 3) Normal space, BO is relatively close to RO, indicating that the utilization efficiency of spatial resources has reached a normal state and has not been overused or wasted; Accordingly, by considering the ratio of benefit output to resource occupation, the utilization efficiency of green park UE is quantified, namely: Based on this, efficient space and inefficient space are extracted. The specific extraction formula is: Among them, UE is the utilization efficiency, i is the type of green park, and i represents inefficient space, normal space and efficient space respectively.

6. The relative performance evaluation device for urban green parks is characterized by: include: Green park function classification module, used to automatically classify green parks according to different core function types; The inefficient space identification module is used to establish two corresponding indicator systems for benefit output and resource occupation. The hierarchical analysis method and cyclic random forest regression are used to evaluate the benefit output and resource occupation of different types of green parks. Based on the relationship between the two, the utilization efficiency of different types of green parks is comprehensively evaluated to identify efficient and inefficient green parks. The influencing factor mining module is used to explore the main driving factors affecting inefficient green parks and quantify the importance of resource occupation indicators.

7. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the steps of the relative performance evaluation method of urban green parks as described in any one of claims 1 to 5 are implemented.

8. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the relative performance evaluation method of urban green parks as described in any one of claims 1 to 5 are implemented.

Images