Land consolidation and rehabilitation effectiveness evaluation method and system based on lasso regression and expert knowledge fusion

Abstract

This invention discloses a method and system for evaluating the effectiveness of land consolidation based on LASSO regression and expert knowledge fusion. The method includes: constructing an indicator evaluation system covering multiple dimensions; constructing an indicator observation dataset for each region to be evaluated; obtaining the expert comprehensive score of land consolidation projects in each region to be evaluated as a supervised learning label; using a LASSO regression model with dimensional structure constraints, training the model using the indicator observation dataset and supervised learning labels, and outputting the model training results; based on the expert review of the model training results, determining whether to adjust the indicator evaluation system; if so, adjusting the indicator evaluation system and continuing to train the model; otherwise, adjusting the weight ratio of each dimension of the model to achieve the evaluation of the effectiveness of land consolidation in each region to be evaluated. This invention can solve the problems of unstable evaluation results, distorted variable selection, and poor interpretability of purely data-driven models.

Description

Technical Field

[0001] This invention relates to a method and system for evaluating the effectiveness of land consolidation based on LASSO regression and expert knowledge fusion, belonging to the technical field of land consolidation effectiveness evaluation. Background Technology

[0002] In recent years, with the continuous deepening and advancement of land resource management policies, comprehensive land consolidation work, with townships as units, has been widely carried out across the country. Through large-scale investment and systematic consolidation, various regions have achieved significant economic, social, and ecological benefits. Land consolidation has become an important means of optimizing the spatial pattern of the national territory, promoting rural revitalization, and building ecological civilization. However, in the process of continuously advancing comprehensive land consolidation practices, the evaluation of consolidation effectiveness still faces many challenges. For example, the current evaluation system suffers from unreasonable indicator settings, incomplete indicator coverage, and a lack of basis for weighting, making it difficult to comprehensively and objectively reflect the consolidation effectiveness. To strengthen the whole-process management and post-evaluation of projects, relevant competent authorities have clearly required that, after the project is completed and passes acceptance, a systematic effectiveness evaluation should be organized according to the corresponding evaluation standards and technical specifications.

[0003] LASSO (Minimum Absolute Contraction and Selection Operator) regression, as a commonly used feature selection and modeling method, possesses strong variable screening and weight calculation capabilities. This method can automatically identify key factors that significantly influence the effectiveness of land consolidation when processing multidimensional indicator data, thereby establishing a predictive model. LASSO regression primarily relies on the numerical relationship between the indicators in the input data and the target variable, with its optimization objective being to minimize computational error. However, this method may overlook the policy orientation inherent in land consolidation itself, the multi-faceted attributes of the final indicator system, and the differences in biases across different projects when evaluating land consolidation.

[0004] On the other hand, expert knowledge, as an assessment tool widely used in land consolidation practice, possesses strong domain-specific guidance and experiential judgment capabilities. Experts can scientifically select and classify assessment indicators based on policy guidance, consolidation goals, and regional realities, and provide reasonable indicator interpretations and stratification suggestions. In terms of variable selection, weight reference, and indicator importance ranking, expert knowledge can serve as an important supplement to data-driven models, helping to improve the relevance and policy adaptability of the assessment system. Summary of the Invention

[0005] In view of this, the present invention provides a method, system, computer equipment and storage medium for evaluating the effectiveness of land consolidation based on the fusion of LASSO regression and expert knowledge, which can solve the problems of unstable evaluation results, distorted variable selection and poor interpretability of pure data-driven models.

[0006] The first objective of this invention is to provide a method for evaluating the effectiveness of land consolidation based on the fusion of LASSO regression and expert knowledge.

[0007] The second objective of this invention is to provide a land consolidation effectiveness evaluation system based on the fusion of LASSO regression and expert knowledge.

[0008] A third objective of this invention is to provide a computer device.

[0009] A fourth objective of this invention is to provide a storage medium.

[0010] The first objective of this invention can be achieved by adopting the following technical solution:

[0011] A land consolidation effectiveness evaluation method based on LASSO regression and expert knowledge fusion, the method comprising:

[0012] Based on the assessment requirements of land consolidation projects, an indicator evaluation system covering multiple dimensions is constructed;

[0013] Collect regulatory data on land consolidation projects in each region to be evaluated, and construct an indicator observation dataset for each region to be evaluated based on the indicator evaluation system;

[0014] The comprehensive expert score for each land consolidation project in the region to be evaluated is used as a monitoring and learning label.

[0015] A LASSO regression model with dimensional structure constraints is used. The model is trained using an indicator observation dataset and supervised learning labels, and the training results are output.

[0016] Based on the expert review of the model training results, it is determined whether the indicator evaluation system needs to be adjusted. If so, the indicator evaluation system is adjusted and the model is retrained. If not, the weight ratio of each dimension of the model is adjusted, and the final weight of all indicators under each dimension is uniformly scaled or enhanced proportionally to achieve the evaluation of the land consolidation effectiveness of each region to be evaluated.

[0017] Furthermore, the collection of monitoring data on land consolidation projects in each region to be evaluated, and the construction of an indicator observation dataset for each region to be evaluated based on the indicator evaluation system, includes:

[0018] We collected regulatory data on land consolidation projects in each region to be evaluated. Combined with policy implementation reports and project summary documents for each region, we identified and analyzed unstructured data tables and text descriptions in the policy implementation reports, extracted key numerical indicators, and constructed an indicator observation dataset for each region to be evaluated by semantic matching through a large language model, based on the indicator evaluation system.

[0019] Furthermore, the LASSO regression model with dimensional structure constraints is based on the LASSO sparsity mechanism, introducing dimensional-level structural constraints to limit the sum of the weights of each dimensional indicator to no less than a threshold set by experts.

[0020] Furthermore, the objective function of the LASSO regression model is as follows:

[0021] ;

[0022] in, , represents the comprehensive score vector of experts for land consolidation projects in n areas to be evaluated, where each element is the comprehensive score of experts for an area to be evaluated, and serves as a label for supervised learning; , represents the indicator data matrix, where each row corresponds to a region to be evaluated and each column corresponds to a specific indicator, containing a total of p indicators; , is the vector of indicator weight coefficients to be learned. Each element in the vector represents the weight of the corresponding indicator in the expert comprehensive score, and there are a total of p weights. This is the optimal index weight vector obtained from the final training. A regularization coefficient greater than 0 controls the strength of the sparsity penalty term; , which is the set of indicator indices corresponding to the k-th dimension; This represents the lower limit of the minimum total weight of the indicators in the k-th dimension, as set by the experts.

[0023] Furthermore, the adjustment of the indicator evaluation system includes: eliminating incorrectly selected indicators, supplementing missing variables, and / or correcting the dimensional structure.

[0024] Furthermore, the various dimensions of the indicator evaluation system include driving force dimension, pressure dimension, state dimension, influence dimension, and response dimension.

[0025] Furthermore, the model training results include the optimal weight of each indicator, the key indicators retained in each dimension, and the importance retention status of each dimension.

[0026] The second objective of this invention can be achieved by adopting the following technical solution:

[0027] A land consolidation effectiveness evaluation system based on LASSO regression and expert knowledge fusion, the system comprising:

[0028] The first construction module is used to build an indicator evaluation system covering multiple dimensions based on the assessment needs of land consolidation projects;

[0029] The second construction module is used to collect regulatory data on land consolidation projects in each region to be evaluated, and to construct an indicator observation dataset for each region to be evaluated based on the indicator evaluation system.

[0030] The acquisition module is used to obtain the comprehensive expert scores for each land consolidation project in the region to be evaluated as labels for supervised learning.

[0031] The training module is used to train the LASSO regression model with dimensional structure constraints using the indicator observation dataset and supervised learning labels, and output the model training results, which include the optimal weight of each indicator, the key indicators retained in each dimension, and the importance retention of each dimension.

[0032] The adjustment module is used to determine whether to adjust the indicator evaluation system based on the expert review of the model training results. If so, the indicator evaluation system is adjusted and the model is trained again. If not, the weight ratio of each dimension of the model is adjusted, and the final weight of all indicators under each dimension is uniformly scaled or enhanced in proportion to achieve the evaluation of the land consolidation effectiveness of each region to be evaluated.

[0033] The third objective of this invention can be achieved by adopting the following technical solution:

[0034] A computer device includes a processor and a memory for storing processor-executable programs, wherein when the processor executes the programs stored in the memory, it implements the above-described land consolidation effectiveness evaluation method.

[0035] The fourth objective of this invention can be achieved by adopting the following technical solution:

[0036] A storage medium storing a program that, when executed by a processor, implements the aforementioned method for evaluating the effectiveness of land consolidation.

[0037] The present invention has the following advantages over the prior art:

[0038] 1. Based on the traditional data-driven model, this invention introduces dimensional structural constraints set by experts, which enables the model to retain the policy significance and practical importance of each dimension while automatically identifying key indicators. This solves the problem that relying solely on data in the prior art may lead to dimensional omissions or poor interpretability, and realizes the integration of expert knowledge and learning models.

[0039] 2. The model trained by this invention can not only be used for comprehensive scoring and ranking of land consolidation projects, serving as an evaluation tool for actual management decisions; it can also output key indicators and their weights within the dimensions, assisting experts in optimizing and reducing the indicator system, serving as an indicator selection tool for the early stage of system construction, and has strong flexibility and practicality.

[0040] 3. This invention supports flexible adjustment of the weight ratios of each dimension based on the policy priorities and rectification objectives of different regions, achieving a differentiated evaluation method tailored to local conditions. While maintaining the model structure, it is applicable to evaluation tasks in different provinces and cities and for different types of projects, thus improving the model's applicability and scalability. Attached Figure Description

[0041] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the structures shown in these drawings without creative effort.

[0042] Figure 1 This is a simplified flowchart of the land consolidation effectiveness evaluation method based on semantic change detection method according to Embodiment 1 of the present invention.

[0043] Figure 2 This is a detailed flowchart of the land consolidation effectiveness evaluation method based on semantic change detection method according to Embodiment 1 of the present invention.

[0044] Figure 3 This is a structural block diagram of the land consolidation effectiveness evaluation system based on semantic change detection method according to Embodiment 2 of the present invention.

[0045] Figure 4 This is a structural block diagram of the computer device according to Embodiment 3 of the present invention. Detailed Implementation

[0046] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are some embodiments of the present invention, but not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0047] Example 1:

[0048] like Figure 1 and Figure 2 As shown in the figure, this embodiment provides a land consolidation effectiveness evaluation method based on LASSO regression and expert knowledge fusion. The method includes the following steps:

[0049] S201. Based on the assessment requirements of land consolidation projects, construct an indicator evaluation system covering multiple dimensions.

[0050] This embodiment uses the DPSIR framework (Drivers–Pressures–State–Impact–Response) as the theoretical basis for constructing the indicator system. The DPSIR model is an environmental and social systems analysis framework proposed by the European Environment Agency to describe the causal relationship between human activities and environmental systems. This framework includes the following five components: Drivers: Social, economic, and political factors that drive human activities and economic development; Pressures: The pressures caused by the drivers, such as land reclamation and resource overexploitation; State: The current state of the ecological or land system, such as land degradation rate and vegetation cover; Impact: The impact of state changes on ecosystems or human society, such as a decline in ecosystem service functions; Response: Policy or management measures to address environmental problems, such as financial investment. The advantage of using this framework is that it can systematically construct an indicator system covering multiple dimensions, including ecology, economy, and society, starting from the causal chain, ensuring that the selected indicators have logical integrity and policy interpretability.

[0051] Guided by the DPSIR framework, this embodiment invited multiple domain experts to form an expert group. Based on relevant national policy documents, local governance experience and academic literature, and combined with actual survey results, a preliminary selection of candidate indicators covering multiple dimensions was made. This indicator system will be organized into tables, listing some indicator names under each dimension as the input basis for subsequent model training, as shown in Tables 1 to 5 below.

[0052] Table 1. Driving Force Dimensions in the Indicator Evaluation System

[0053]

[0054] Table 2. Stress Dimension in the Indicator Evaluation System

[0055]

[0056] Table 3. State Dimension in the Indicator Evaluation System

[0057]

[0058] Table 4. Influence Dimensions in the Indicator Evaluation System

[0059]

[0060] Table 5 Response Dimensions in the Indicator Evaluation System

[0061]

[0062] S202. Collect regulatory data on land consolidation projects in each region to be evaluated, and construct an indicator observation dataset for each region to be evaluated based on the indicator evaluation system.

[0063] Specifically, regulatory data on land consolidation projects in each region to be evaluated are collected. Combined with policy implementation reports and project summary documents (as supplementary data sources) of each region to be evaluated, this embodiment uses artificial intelligence technology to identify and parse the unstructured data tables and text descriptions in the policy implementation reports, extract key numerical indicators, and combine them with the indicator definitions and interpretations provided by experts in the indicator evaluation system. Semantic matching is performed through a large language model to construct an indicator observation dataset for each region to be evaluated.

[0064] S203. Obtain the expert comprehensive score for each land consolidation project in the region to be evaluated as a monitoring and learning label.

[0065] In this embodiment, experts are invited to conduct a comprehensive evaluation of land consolidation projects in multiple regions to be evaluated, and the comprehensive expert score for each region to be evaluated is used as a supervised learning label.

[0066] S204. Using a LASSO regression model with dimensional structure constraints, train the LASSO regression model using the indicator observation dataset and supervised learning labels, and output the model training results.

[0067] In this embodiment, a LASSO regression model with dimensional structure constraints is used during the model training phase. The LASSO regression model with dimensional structure constraints introduces dimensional-level structural constraints on the basis of the traditional LASSO sparsity mechanism, which limits the total weight of each dimension indicator to no less than the threshold set by experts. This ensures that while the model automatically selects key indicators, it retains the policy significance and practical importance of each dimension, and prevents the model from removing certain important dimensions as a whole due to insufficient representativeness of the training samples.

[0068] Specifically, experts provided scores for the importance of each dimension based on policy guidance. Subsequently, the analytic hierarchy process (AHP) was used to normalize the expert scores, construct a judgment matrix, calculate the importance weights of each dimension, and set corresponding retention lower limits based on the weights of different dimensions. This means that the sum of the weights of all indicators within a dimension must not be lower than this threshold. The higher the weight of a dimension, the higher its corresponding setting should be.

[0069] The objective function of the LASSO regression model in this embodiment is as follows:

[0070] ;

[0071] in, , represents the comprehensive score vector of experts for land consolidation projects in n areas to be evaluated, where each element is the comprehensive score of experts for an area to be evaluated, and serves as a label for supervised learning; , represents the indicator data matrix, where each row corresponds to a region to be evaluated and each column corresponds to a specific indicator, containing a total of p indicators; , is the vector of indicator weight coefficients to be learned. Each element in the vector represents the weight of the corresponding indicator in the expert comprehensive score, and there are a total of p weights. This is the optimal index weight vector obtained from the final training. A regularization coefficient greater than 0 controls the strength of the sparsity penalty term; , is the set of indexes corresponding to the k-th dimension, and in this embodiment w=5; This represents the lower limit of the minimum total weight of the indicators in the k-th dimension, as set by the experts.

[0072] The constraint means that the sum of the absolute weights of all indicators in the nth dimension is at least 1. This is used to preserve the importance of the dimension and prevent it from being completely removed during training.

[0073] After the model training is completed, this embodiment outputs the model training results: the optimal weight of each indicator, the key indicators retained in each dimension, and the importance retention status of each dimension (whether the constraints are met).

[0074] S205. Based on the expert review of the model training results, determine whether to adjust the indicator evaluation system.

[0075] In this embodiment, the model training results are fed back to the expert group, which reviews the indicators based on its research experience and determines the following: whether the key indicators selected by the model have policy or practical significance; whether important variables (such as indicators specific to certain regions) have been omitted; and whether the ranking of the importance of dimensions needs to be adjusted or refined. If the indicator evaluation system needs to be adjusted, appropriate adjustments are made according to the expert opinions, such as removing incorrectly selected indicators, supplementing omitted variables, and correcting the dimensional structure. Subsequently, the corrected indicator data is re-inputted into the model for training, forming a closed-loop optimization process of "model training - expert feedback - indicator correction". If the indicator evaluation system does not need to be adjusted, the process proceeds to step S206.

[0076] S206. Adjust the weight ratio of each dimension of the model, and uniformly scale or enhance the final weight of all indicators under each dimension in proportion to achieve the evaluation of the land consolidation effectiveness of each region to be evaluated.

[0077] Since the focus of land reclamation may differ in different regions—for example, some regions prioritize ecological restoration while others place greater emphasis on economic benefits or social impact—this embodiment supports adjusting the weight ratio of each dimension in practical applications without changing the model structure and parameters. The final weight of all indicators under that dimension can be uniformly scaled or enhanced proportionally to achieve the evaluation of land reclamation effectiveness in each region to be evaluated, thereby enabling flexible and site-specific assessments.

[0078] It should be noted that although the method operations of the above embodiments are described in a specific order, this does not require or imply that these operations must be performed in that specific order, or that all the operations shown must be performed to achieve the desired result. On the contrary, the described steps may be performed in a different order. Additionally or alternatively, certain steps may be omitted, multiple steps may be combined into one step, and / or one step may be broken down into multiple steps.

[0079] Example 2:

[0080] like Figure 3 As shown, this embodiment provides a land consolidation effectiveness evaluation system based on LASSO regression and expert knowledge fusion. The system includes a first construction module 301, a second construction module 302, an acquisition module 303, a training module 304, and an adjustment module 305. The specific functions of each module are as follows:

[0081] The first construction module 301 is used to construct an indicator evaluation system covering multiple dimensions based on the assessment needs of land consolidation projects;

[0082] The second construction module 302 is used to collect regulatory data on land consolidation projects in each region to be evaluated, and to construct an indicator observation dataset for each region to be evaluated based on the indicator evaluation system.

[0083] Module 303 is used to obtain the expert comprehensive score of each land consolidation project in the region to be evaluated as a label for supervised learning.

[0084] Training module 304 is used to train the LASSO regression model with dimensional structure constraints using the indicator observation dataset and supervised learning labels, and output the model training results, which include the optimal weight of each indicator, the key indicators retained in each dimension, and the importance retention of each dimension.

[0085] The adjustment module 305 is used to determine whether to adjust the indicator evaluation system based on the expert review of the model training results. If so, the indicator evaluation system is adjusted and the model is trained again. If not, the weight ratio of each dimension of the model is adjusted, and the final weight of all indicators under each dimension is uniformly scaled or enhanced in proportion to achieve the evaluation of the land consolidation effectiveness of each region to be evaluated.

[0086] It should be noted that the system provided in this embodiment is only an example of the above-described division of functional units. In practical applications, the above functions can be assigned to different functional units as needed, that is, the internal structure can be divided into different functional units to complete all or part of the functions described above.

[0087] Example 3:

[0088] This embodiment provides a computer device, such as... Figure 4 As shown, it includes a processor 402, a memory, an input device 403, a display device 404, and a network interface 405 connected via a system bus 401. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium 306 and internal memory 407. The non-volatile storage medium 406 stores an operating system, computer programs, and a database. The internal memory 407 provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. When the processor 402 executes the computer programs stored in the memory, it implements the land consolidation effectiveness evaluation method of Embodiment 1 described above, as follows:

[0089] Based on the assessment requirements of land consolidation projects, a multi-dimensional indicator evaluation system is constructed. Regulatory data on land consolidation projects in each region to be evaluated are collected, and an indicator observation dataset for each region is constructed according to the indicator evaluation system. Expert comprehensive scores for land consolidation projects in each region are obtained as supervised learning labels. A LASSO regression model with dimensional structure constraints is used, and the model is trained using the indicator observation dataset and supervised learning labels, outputting the model training results. Based on expert review of the model training results, it is determined whether the indicator evaluation system needs adjustment. If so, the indicator evaluation system is adjusted, and the model is retrained. If not, the weight proportions of each dimension of the model are adjusted, and the final weights of all indicators under each dimension are uniformly scaled or enhanced proportionally to achieve the assessment of the land consolidation effectiveness in each region to be evaluated.

[0090] Example 4:

[0091] This embodiment provides a storage medium, which is a computer-readable storage medium, storing a computer program. When the computer program is executed by a processor, it implements the land consolidation effectiveness evaluation method of Embodiment 1 above, as follows:

[0092] Based on the assessment requirements of land consolidation projects, a multi-dimensional indicator evaluation system is constructed. Regulatory data on land consolidation projects in each region to be evaluated are collected, and an indicator observation dataset for each region is constructed according to the indicator evaluation system. Expert comprehensive scores for land consolidation projects in each region are obtained as supervised learning labels. A LASSO regression model with dimensional structure constraints is used, and the model is trained using the indicator observation dataset and supervised learning labels, outputting the model training results. Based on expert review of the model training results, it is determined whether the indicator evaluation system needs adjustment. If so, the indicator evaluation system is adjusted, and the model is retrained. If not, the weight proportions of each dimension of the model are adjusted, and the final weights of all indicators under each dimension are uniformly scaled or enhanced proportionally to achieve the assessment of the land consolidation effectiveness in each region to be evaluated.

[0093] It should be noted that the computer-readable storage medium in this embodiment can be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. The computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof.

[0094] In this embodiment, the computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in connection with an instruction execution system, apparatus, or device. In this embodiment, the computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, carrying a computer-readable program. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. The computer-readable signal medium can also be any computer-readable storage medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The computer program contained on the computer-readable storage medium can be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination thereof.

[0095] The computer-readable storage medium described above can be used to write computer programs for executing this embodiment in one or more programming languages ​​or combinations thereof. These programming languages ​​include object-oriented programming languages—such as Java, Python, and C++—and conventional procedural programming languages—such as C or similar programming languages. The program can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0096] In summary, this invention introduces expert-defined dimensional structural constraints on the basis of traditional data-driven models, enabling the model to retain the policy significance and practical importance of each dimension while automatically identifying key indicators. This solves the problem that relying solely on data in existing technologies may lead to dimensional omissions or poor interpretability, and achieves the integration of expert knowledge and learning models.

[0097] The above description is merely a preferred embodiment of the present invention. It will be apparent to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects. The scope of the present invention is defined by the appended claims rather than the foregoing description, and thus all variations falling within the meaning and scope of the equivalents of the claims are intended to be included within the present invention.

Claims

1. A land consolidation effectiveness evaluation method based on LASSO regression and expert knowledge fusion, characterized in that, The method includes: Based on the assessment requirements of land consolidation projects, an indicator evaluation system covering multiple dimensions is constructed; Collect regulatory data on land consolidation projects in each region to be evaluated, and construct an indicator observation dataset for each region to be evaluated based on the indicator evaluation system; The comprehensive expert score for each land consolidation project in the region to be evaluated is used as a monitoring and learning label. A LASSO regression model with dimensional structure constraints is used. The model is trained using an indicator observation dataset and supervised learning labels, and the training results are output. Based on the expert review of the model training results, it is determined whether the indicator evaluation system needs to be adjusted. If so, the indicator evaluation system is adjusted and the model is retrained. If not, the weight ratio of each dimension of the model is adjusted, and the final weight of all indicators under each dimension is uniformly scaled or enhanced proportionally to achieve the evaluation of the land consolidation effectiveness of each region to be evaluated.

2. The land consolidation effectiveness evaluation method according to claim 1, characterized in that, The process involves collecting monitoring data on land consolidation projects in each region to be evaluated, and constructing an indicator observation dataset for each region based on the indicator evaluation system, including: We collected regulatory data on land consolidation projects in each region to be evaluated. Combined with policy implementation reports and project summary documents for each region, we identified and analyzed unstructured data tables and text descriptions in the policy implementation reports, extracted key numerical indicators, and constructed an indicator observation dataset for each region to be evaluated by semantic matching through a large language model, based on the indicator evaluation system.

3. The land consolidation effectiveness evaluation method according to claim 1, characterized in that, The LASSO regression model with dimensional structure constraints is based on the LASSO sparsity mechanism, which introduces dimensional-level structural constraints to limit the total weight of each dimensional indicator to no less than a threshold set by experts.

4. The land consolidation effectiveness evaluation method according to claim 1, characterized in that, The objective function of the LASSO regression model is as follows: ； in, , represents the comprehensive score vector of experts for land consolidation projects in n areas to be evaluated, where each element is the comprehensive score of experts for an area to be evaluated, and serves as a label for supervised learning; , represents the indicator data matrix, where each row corresponds to a region to be evaluated and each column corresponds to a specific indicator, containing a total of p indicators; , is the vector of indicator weight coefficients to be learned. Each element in the vector represents the weight of the corresponding indicator in the expert comprehensive score, and there are a total of p weights. This is the optimal index weight vector obtained from the final training. A regularization coefficient greater than 0 controls the strength of the sparsity penalty term; , which is the set of indicator indices corresponding to the k-th dimension; This represents the lower limit of the minimum total weight of the indicators in the k-th dimension, as set by the experts.

5. The land consolidation effectiveness evaluation method according to claim 1, characterized in that, The adjustment of the indicator evaluation system includes: eliminating incorrectly selected indicators, supplementing missing variables, and / or correcting the dimensional structure.

6. The land consolidation effectiveness evaluation method according to any one of claims 1-5, characterized in that, The evaluation system includes four dimensions: driving force, pressure, state, influence, and response.

7. The land consolidation effectiveness evaluation method according to any one of claims 1-5, characterized in that, The model training results include the optimal weights for each indicator, the key indicators retained in each dimension, and the importance retention status of each dimension.

8. A land consolidation effectiveness evaluation system based on LASSO regression and expert knowledge fusion, characterized in that, The system includes: The first construction module is used to build an indicator evaluation system covering multiple dimensions based on the assessment needs of land consolidation projects; The second construction module is used to collect regulatory data on land consolidation projects in each region to be evaluated, and to construct an indicator observation dataset for each region to be evaluated based on the indicator evaluation system. The acquisition module is used to obtain the comprehensive expert scores for each land consolidation project in the region to be evaluated as labels for supervised learning. The training module is used to train the LASSO regression model with dimensional structure constraints using the indicator observation dataset and supervised learning labels, and output the model training results, which include the optimal weight of each indicator, the key indicators retained in each dimension, and the importance retention of each dimension. The adjustment module is used to determine whether to adjust the indicator evaluation system based on the expert review of the model training results. If so, the indicator evaluation system is adjusted and the model is trained again. If not, the weight ratio of each dimension of the model is adjusted, and the final weight of all indicators under each dimension is uniformly scaled or enhanced in proportion to achieve the evaluation of the land consolidation effectiveness of each region to be evaluated.

9. A computer device comprising a processor and a memory for storing a processor-executable program, characterized in that, When the processor executes the program stored in the memory, it implements the land consolidation effectiveness evaluation method according to any one of claims 1-7.

10. A storage medium storing a program, characterized in that, When the program is executed by the processor, it implements the land consolidation effectiveness evaluation method according to any one of claims 1-7.

Images