Project cost prediction model and method

Abstract

The embodiment of the invention provides a project cost prediction model and method, and relates to the technical field of project cost, and the model comprises an obtaining module which is used for obtaining target data, and the target data comprises project amount data, quota data, labor, material and machine data, and / or information price data; the analysis module is used for analyzing target engineering characteristics according to the target data; the matching module is used for matching target indexes according to the target engineering features; and the prediction module is used for generating a cost prediction result according to the target index. In this way, the dependency degree of project cost prediction on manpower can be reduced, and project cost prediction precision and prediction efficiency can be improved.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of engineering cost technology, and in particular, to an engineering cost prediction model and method. Background Art

[0002] Project cost forecasting is mainly used to estimate project costs in order to improve the efficiency and accuracy of project cost control. In the related technologies, most of them rely on manual experience to estimate and forecast project costs. The above methods have the problems of high manual dependence on project cost forecasting, poor accuracy of project cost forecasting, and low efficiency.

[0003] Therefore, a new technical solution is urgently needed to solve the above technical problems. Summary of the invention

[0004] According to the embodiments of the present application, a project cost prediction model and method are provided, which can reduce the dependence of project cost prediction on manual labor and improve the accuracy and efficiency of project cost prediction.

[0005] In a first aspect of the present application, a project cost prediction model is provided, comprising: An acquisition module is used to acquire target data, wherein the target data includes: engineering quantity data, quota data, labor, material and machine data, and / or information price data; A parsing module, used to parse target engineering features based on target data; A matching module is used to match target indicators according to target engineering characteristics; The prediction module is used to generate cost prediction results based on target indicators.

[0006] In some feasible implementations, the acquisition module is provided with a variety of target interfaces for acquiring target data.

[0007] In some feasible implementations, the above-mentioned parsing module includes: The preprocessing unit is used to perform preprocessing operations on the target data.

[0008] In some feasible implementations, the above-mentioned parsing module further includes: A first parsing unit is used to determine a plurality of project units included in a target project according to target data; A second parsing unit is used to determine construction method characteristics, structural characteristics, and / or geological characteristics corresponding to the plurality of project units according to the target data; The generating unit is used to generate a target engineering feature set according to construction method features, structural features, and / or geological features.

[0009] In some feasible implementations, the matching module includes: The matching unit is used to determine the target indicator and the weight corresponding to the target indicator according to the similarity between the target engineering feature and the preset engineering feature.

[0010] In some feasible implementations, the above-mentioned target indicators include: necessary indicators, and / or reference indicators; Among them, the necessary indicators include: labor cost indicators, material cost indicators, and / or equipment cost indicators; Reference indicators include: tax payment indicators.

[0011] In some feasible implementations, the prediction module includes: A first prediction unit, configured to generate a cost prediction result according to necessary indicators and corresponding weights when it is determined that the accuracy requirement of the cost prediction result is less than or equal to the preset accuracy requirement and / or when it is determined that the feedback time of the cost prediction result is less than or equal to the preset time; The second prediction unit is used to generate a cost prediction result based on necessary indicators, reference indicators and corresponding weights when it is determined that the accuracy requirement of the cost prediction result is greater than the preset accuracy requirement and / or when it is determined that the feedback time of the cost prediction result is greater than the preset time.

[0012] In some feasible implementations, the project cost prediction model as described in any of the above items further includes: An evaluation module is used to generate an accuracy evaluation result of the engineering cost prediction model based on the square difference between the actual cost result and the cost prediction result; Among them, according to the square difference between the actual cost result and the cost prediction result, the accuracy evaluation result of the engineering cost prediction model is generated, including: according to the following formula, the accuracy evaluation result is generated: ; in, is the accuracy evaluation result, is the sample size, For the The predicted value of samples, For the The true value of the samples.

[0013] In some feasible implementations, the above-mentioned project cost prediction model further includes: An updating module, used for performing a target updating operation on the project cost prediction model based on a preset objective function when the accuracy evaluation result is less than or equal to a preset accuracy threshold; The preset objective function includes the following functions: ; in, For the The target parameters for the step update, For the The target parameters for the step update, is the target learning rate, is the preset constant, For the Step target first-order moment estimation momentum, No. Step target second moment estimation momentum.

[0014] In a second aspect of the present application, a method for predicting construction cost is provided, comprising: Acquire target data, wherein the target data includes: engineering quantity data, quota data, labor, material and machine data, and / or information price data; Analyze the target engineering features based on the target data; Match target indicators according to target project characteristics; Generate cost forecast results based on target indicators.

[0015] The embodiment of the present application provides a model and method for predicting the construction cost, wherein the model includes: an acquisition module for acquiring target data, wherein the target data includes: engineering quantity data, quota data, labor, material and machine data, and / or information price data; a parsing module for parsing target engineering features according to the target data; a matching module for matching target indicators according to the target engineering features; and a prediction module for generating cost prediction results according to the target indicators. The present application can reduce the dependence of construction cost prediction on manual labor and improve the accuracy and efficiency of construction cost prediction.

[0016] It should be understood that the contents described in the Summary of the Invention are not intended to limit the key or important features of the embodiments of the present application, nor are they intended to limit the scope of the present application. Other features of the present application will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The above and other features, advantages and aspects of the embodiments of the present application will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. In the accompanying drawings, the same or similar reference numerals represent the same or similar elements, wherein: Figure 1 A structural schematic diagram of a project cost prediction model provided in an embodiment of the present application; Figure 2 A schematic diagram of a process for predicting construction cost provided in an embodiment of the present application; Figure 3A structural schematic diagram of a terminal device or server provided in an embodiment of the present application. DETAILED DESCRIPTION

[0018] In order to make the purpose, technical solution and advantages of the embodiments of the present disclosure clearer, the technical solution in the embodiments of the present disclosure will be clearly and completely described below in conjunction with the drawings in the embodiments of the present disclosure. Obviously, the described embodiments are part of the embodiments of the present disclosure, not all of the embodiments. Based on the embodiments in the present disclosure, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present disclosure.

[0019] In addition, the term "and / or" in this article is only a description of the association relationship between the associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship.

[0020] According to a first aspect of the embodiments of the present application, a construction cost prediction model is proposed. Figure 1 A structural schematic diagram of a project cost prediction model 100 provided in an embodiment of the present application is shown as follows: Figure 1 As shown, the model 100 includes: The acquisition module 110 is used to acquire target data, wherein the target data includes: engineering quantity data, quota data, labor, material and machinery data, and / or information price data.

[0021] Exemplarily, the acquisition module 110 can determine the engineering quantity data, quota data, labor, material and machinery data, and / or information price data corresponding to the target project by traversing web page data, software data, official government websites of various regions, and / or manual input by target users in the engineering construction industry.

[0022] Exemplarily, the above-mentioned engineering quantity data may include: the scale of the target project, the volume of the target project, and / or the various work items included in the target project.

[0023] Exemplarily, the above-mentioned quota data may include: cost standards or quotas of unit engineering quantities of the target project, for example: construction costs per square meter.

[0024] Exemplarily, the above-mentioned labor, material and machinery data may include: labor data, material data, and / or mechanical equipment data, etc.

[0025] Exemplarily, the information price data can be determined based on the usage information and supply information of various engineering materials. Specifically, the information price information can be determined by calculating the weighted average based on the usage information and the supply information. It should be noted that the information price data can be the average market price of various engineering materials.

[0026] Specifically, the above-mentioned engineering quantity data, quota data, material and machinery data, and / or information price data may be stored as a target data table to facilitate the analysis module 120 to analyze and determine the target engineering features according to the above-mentioned target data.

[0027] The parsing module 120 is used to parse target engineering features according to target data.

[0028] Exemplarily, the target engineering features may include construction method features, structural features, environmental features, equipment features, labor features, and material features. The labor features may include labor cost features and / or labor efficiency features. The material features may include material cost features and / or material quality features.

[0029] It should be noted that the above-mentioned target engineering features may also include: associated features.

[0030] Exemplarily, the data associations between the above-mentioned engineering quantity data, quota data, labor, material and machinery data, and / or information price data can be determined by traversing the above-mentioned engineering quantity data, quota data, labor, material and machinery data, and / or information price data, and a target mapping relationship table can be established to determine the above-mentioned association characteristics.

[0031] Specifically, the above-mentioned target mapping relationship table can be established to determine the above-mentioned association characteristics by traversing the above-mentioned quota data and the above-mentioned labor, material and machinery data to determine the connection between the quota data and the labor, material and machinery data, traversing the above-mentioned labor, material and machinery data and the above-mentioned information price data to determine the connection between the above-mentioned labor, material and machinery data and the above-mentioned information price data.

[0032] Among them, the above-mentioned correlation characteristics may include: correlation characteristics between construction method characteristics and labor cost characteristics, correlation characteristics between construction method characteristics and material cost characteristics, correlation characteristics between structural characteristics and labor cost characteristics, correlation characteristics between structural characteristics and material cost characteristics, correlation characteristics between environmental characteristics and labor cost characteristics, correlation characteristics between environmental characteristics and labor cost characteristics, correlation characteristics between labor efficiency characteristics and labor cost characteristics, and / or correlation characteristics between material cost characteristics and material quality characteristics, etc.

[0033] The matching module 130 is used to match the target index according to the target engineering characteristics.

[0034] Exemplarily, the corresponding target indicators can be matched according to the construction method characteristics, structural characteristics, environmental characteristics, equipment characteristics, manual characteristics, material characteristics, and / or associated characteristics. It should be noted that the above-mentioned target indicators include: indicators whose contribution to the target project performance is greater than or equal to the preset contribution, such as: cost indicators, time indicators, quality indicators, efficiency indicators, and / or environmental protection indicators. Among them, the above-mentioned preset contribution can be determined according to the target user's accuracy requirements for the cost forecast results. The above-mentioned preset contribution is negatively correlated with the target user's accuracy requirements for the cost forecast results, that is, the higher the target user's accuracy requirements for the cost forecast results, the smaller the above-mentioned preset contribution.

[0035] The prediction module 140 is used to generate cost prediction results according to the target indicators.

[0036] In some feasible implementations, cost prediction results may be generated based on the above-mentioned cost indicators, time indicators, quality indicators, efficiency indicators, and / or environmental protection indicators.

[0037] Exemplarily, the cost prediction result may be generated based on regression analysis, or statistical methods such as Monte Carlo simulation, according to the cost indicators, the time indicators, quality indicators, efficiency indicators, and / or environmental indicators.

[0038] Specifically, the cost prediction result can be generated with the goal of minimizing the cost index and the minimum standards of the time index, quality index, efficiency index, and / or environmental protection index as constraints.

[0039] Based on this, the project cost prediction model provided by the present application includes: an acquisition module for acquiring target data, wherein the target data includes: engineering quantity data, quota data, material and machinery data, and / or information price data; a parsing module for parsing target engineering features according to the target data; a matching module for matching target indicators according to target engineering features; and a prediction module for generating cost prediction results according to target indicators. Based on this, the present application can accurately determine multiple target engineering features and the associated features between multiple target engineering features based on engineering quantity data, quota data, material and machinery data, and / or information price data, and accurately determine the target indicators based on the above multiple target engineering features and the associated features between the above multiple target engineering features, thereby accurately generating the above cost prediction results according to the above target indicators, which is conducive to improving the automatic determination accuracy and automatic determination efficiency of the target indicators, and thus improving the automatic generation accuracy and generation efficiency of the cost prediction results of the target project.

[0040] In some feasible implementations, the acquisition module is provided with a variety of target interfaces for acquiring target data.

[0041] Exemplarily, the above-mentioned multiple target interfaces may include: API interface, database interface, plug-in or extension interface, graphical user interface input control, etc., which are used to determine the above-mentioned target data by retrieving web page data, software data, and / or according to the data entered by the target user.

[0042] Based on this, the above-mentioned acquisition module can be equipped with multiple target interfaces to achieve parameter adjustment or variable adjustment, and access to a variety of construction projects, construction industry software or web pages, thereby expanding the acquisition methods and channels of the target data of this application, improving the richness of the target data, and thereby improving the generation accuracy of the cost forecast results of this application, and improving the scalability, adaptability and robustness of the engineering cost forecast model of this application.

[0043] In some feasible implementations, the above-mentioned parsing module includes: a preprocessing unit, which is used to perform preprocessing operations on the target data.

[0044] Exemplarily, the above-mentioned preprocessing operations may include: data cleaning operations, format unification operations, standardization and normalization operations, and / or dimensionality reduction operations.

[0045] The above data cleaning operations may include: deduplication operations, missing value compensation operations, and outlier removal operations.

[0046] The above-mentioned format unification operation may include: an encoding operation, such as: a one-hot encoding operation, or an integer encoding operation.

[0047] The standardization and normalization operations may be based on the unified standards of the construction industry, and the target data may be subjected to Z-Score standardization operations, Min-Max scaling operations, or RobustScaler scaling operations. The unified standards of the construction industry may be determined based on target files. The target files may include: regulatory documents and regulatory documents of the construction industry, such as the Methods for Preparing Design Estimates of Urban Rail Transit Projects.

[0048] Among them, the above-mentioned dimensionality reduction operation can be implemented based on principal component analysis, nonlinear dimensionality reduction method or linear discriminant analysis method.

[0049] Based on this, the above-mentioned analysis module can improve the data quality of the target data by configuring a preprocessing unit to perform preprocessing operations on the target data, thereby improving the analysis accuracy and efficiency of the analysis module for the target engineering features, and providing an accurate data basis for the matching module to accurately match the target indicators according to the target engineering features, thereby improving the prediction module and generating the accuracy and efficiency of the cost prediction results.

[0050] In some feasible implementations, the above-mentioned analysis module further includes: a first analysis unit, used to determine a plurality of project units included in the target project according to the target data.

[0051] Exemplarily, the first parsing unit may determine the multiple project units included in the target project based on the engineering quantity data, quota data, material and machine data, and / or information price data, based on the functional distinction, process flow, and / or time series of the target project. The functional distinction may include: foundation engineering, main structure, decoration and decoration, etc. The process flow may include: civil engineering, installation engineering, etc. The time series may include: early preparation, mid-term construction, late finishing, etc.

[0052] Specifically, when the target project is an urban rail project or a railway project, the above-mentioned multiple target project units may include: station project unit, elevated station project unit, ground station project unit, underground section project unit, elevated section project unit, ground section project unit, underground continuous wall project unit, bored pile project unit, earthwork project unit, steel support project unit, concrete support project unit, internal reinforced concrete structure project unit, etc.

[0053] In some feasible implementations, the above-mentioned analysis module also includes: a second analysis unit, which is used to determine the construction method characteristics, structural characteristics, and / or geological characteristics corresponding to the multiple project units based on the target data.

[0054] Exemplarily, the above-mentioned construction method characteristics may include: open excavation method, covered excavation method, concealed excavation method, etc.

[0055] It should be noted that the above construction method features can be encoded based on a one-hot encoding method to represent the above construction method features as a binary vector.

[0056] Exemplarily, the structural features may include: number of floors, floor height, column features, etc. The number of floors may include: 1 floor, 2 floors, 3 floors, 4 floors, 5 floors and above. The floor height feature is the vertical distance from the first floor to the upper floor inside the building. The floor height feature may include: 2 meters, 3 meters, 5 meters, etc. The column feature may include: single column, double columns, three columns, no column, etc. The geological features may include: rock sand, mud sand, red soil, granite, soft soil, etc.

[0057] It should be noted that the above structural features can be encoded based on integer encoding and one-hot encoding.

[0058] Exemplarily, the number of floors and the height of floors can be encoded based on integer encoding. Specifically, when the number of floors is 1, 2, 3, 4, and 5, the number of floors is represented as 1, 2, 3, 4, and 5. The column feature can be encoded based on a one-hot encoding method.

[0059] It should be noted that the above geological features can be encoded based on one-hot encoding.

[0060] Specifically, the above-mentioned construction method characteristics, structural characteristics, and / or geological characteristics can be determined based on multi-layer fully connected neural networks (Dense Layers).

[0061] Specifically, the above construction method characteristics, structural characteristics, and / or geological characteristics can be determined based on the following formula: (1); in, is the target engineering feature, The target vector The corresponding weight matrix is, The target data corresponds to the generated target vector, is the preset bias term. It should be noted that the target activation function ReLU (Rectified Linear Unit) can be used according to the above target engineering features. Get the target engineering feature output vector ,in, It can be determined based on the following formula: (2); It should be noted that based on the above formula (2), when determining the target engineering features If it is greater than 0, the above target engineering features are output , otherwise output 0 directly.

[0062] In some feasible implementations, the above-mentioned analysis module further includes: a generation unit, which is used to generate a target engineering feature set according to construction method characteristics, structural characteristics, and / or geological characteristics.

[0063] For example, when the target project unit is an underground continuous wall project unit, it can be determined that the target engineering feature set includes: construction method features, geological features, and column features. It should be noted that the preset number of experiments shows that according to the above target engineering feature set, matching the target indicators, and based on the above target indicators, the accuracy of the generated cost prediction results can reach 86.6%. Among them, the above preset number of experiments can be 100 times.

[0064] It should be noted that the target engineering features included in the target engineering feature set A1 in Table 1 are: construction method features, and the accuracy of the cost prediction results corresponding to the target engineering feature set A1 is 66.5%. The target engineering features included in the target engineering feature set A2 are: number of layers features, and the accuracy of the cost prediction results corresponding to the target engineering feature set A2 is 55.2%. The target engineering features included in the target engineering feature set A3 are: geological features, and the accuracy of the cost prediction results corresponding to the target engineering feature set A3 is 66.6%. The target engineering features included in the target engineering feature set A4 are: column features, and the accuracy of the cost prediction results corresponding to the target engineering feature set A4 is 66%. The target engineering features included in the target engineering feature set B1 are: construction method features and number of layers features, and the accuracy of the cost prediction results corresponding to the target engineering feature set B1 is 79.4%. The target engineering features included in the target engineering feature set B2 are: construction method features and geological features, and the accuracy of the cost prediction results corresponding to the target engineering feature set B2 is 77.9%. The target engineering features included in the target engineering feature set B3 are: construction method features and column features. The accuracy of the cost prediction results corresponding to the target engineering feature set B3 is 75.3%. The target engineering features included in the target engineering feature set B4 are: number of layers and geological features. The accuracy of the cost prediction results corresponding to the target engineering feature set B4 is 76.9%. The target engineering features included in the target engineering feature set B5 are: number of layers and column features. The accuracy of the cost prediction results corresponding to the target engineering feature set B5 is 83.5%. The target engineering features included in the target engineering feature set B6 are: geological features and column features. The accuracy of the cost prediction results corresponding to the target engineering feature set B6 is 83%. The target engineering features included in the target engineering feature set C1 are: construction method features, geological features, and number of layers. The accuracy of the cost prediction results corresponding to the target engineering feature set C1 is 81.4%. The target engineering features included in the target engineering feature set C2 are: construction method features, column features, and number of layers features. The accuracy of the cost prediction results corresponding to the target engineering feature set C2 is 82.8%. The target engineering features included in the target engineering feature set C3 are: construction method features, geological features, and column features. The accuracy of the cost prediction results corresponding to the target engineering feature set C3 is 85.1%. The target engineering features included in the target engineering feature set C4 are: construction method features, geological features, and column features. The accuracy of the cost prediction results corresponding to the target engineering feature set C4 is 86.6%. The target engineering features included in the target engineering feature set D are: construction method features, geological features, number of layers, and column features. The accuracy of the cost prediction results corresponding to the target engineering feature set D is 81.8%.

[0065] Table 1: ; For example, when the target project unit is a bored pile wall project unit, it can be determined that the target project engineering set includes: construction method characteristics and geological characteristics. It should be noted that the preset number of experiments shows that according to the above target engineering feature set, matching the target indicators, based on the above target indicators, the accuracy of the generated cost prediction results can reach 88.4%. Among them, the above preset number of experiments can be 100 times.

[0066] It should be noted that the target engineering features included in the target engineering feature set A1 in Table 2 are: construction method features, and the accuracy of the cost prediction results corresponding to the target engineering feature set A1 is 84.4%. The target engineering features included in the target engineering feature set A2 are: number of layers features, and the accuracy of the cost prediction results corresponding to the target engineering feature set A2 is 61.3%. The target engineering features included in the target engineering feature set A3 are: geological features, and the accuracy of the cost prediction results corresponding to the target engineering feature set A3 is 78.7%. The target engineering features included in the target engineering feature set A4 are: column features, and the accuracy of the cost prediction results corresponding to the target engineering feature set A4 is 64.7%. The target engineering features included in the target engineering feature set B1 are: construction method features and number of layers features, and the accuracy of the cost prediction results corresponding to the target engineering feature set B1 is 83.8%. The target engineering features included in the target engineering feature set B2 are: construction method features and geological features, and the accuracy of the cost prediction results corresponding to the target engineering feature set B2 is 88.4%. The target engineering features included in the target engineering feature set B3 are: construction method features and column features. The accuracy of the cost prediction results corresponding to the target engineering feature set B3 is 76.9%. The target engineering features included in the target engineering feature set B4 are: number of layers and geological features. The accuracy of the cost prediction results corresponding to the target engineering feature set B4 is 78.7%. The target engineering features included in the target engineering feature set B5 are: number of layers and column features. The accuracy of the cost prediction results corresponding to the target engineering feature set B5 is 62.3%. The target engineering features included in the target engineering feature set B6 are: geological features and column features. The accuracy of the cost prediction results corresponding to the target engineering feature set B6 is 77.1%. The target engineering features included in the target engineering feature set C1 are: construction method features, geological features, and number of layers. The accuracy of the cost prediction results corresponding to the target engineering feature set C1 is 85.4%. The target engineering features included in the target engineering feature set C2 are: construction method features, column features, and number of layers features. The accuracy of the cost prediction results corresponding to the target engineering feature set C2 is 76.2%. The target engineering features included in the target engineering feature set C3 are: construction method features, geological features, and column features. The accuracy of the cost prediction results corresponding to the target engineering feature set C3 is 74.7%. The target engineering features included in the target engineering feature set C4 are: construction method features, geological features, and column features. The accuracy of the cost prediction results corresponding to the target engineering feature set C4 is 85.7%. The target engineering features included in the target engineering feature set D are: construction method features, geological features, number of layers, and column features. The accuracy of the cost prediction results corresponding to the target engineering feature set D is 86.3%.

[0067] Table 2: ; For example, when the target project unit is an earthwork project unit, it can be determined that the target engineering feature set includes: construction method features, number of floors features, and column features. It should be noted that the preset number of experiments shows that according to the above target engineering feature set, matching the target indicators, and based on the above target indicators, the accuracy of the generated cost prediction results can reach 80.5%. Among them, the above preset number of experiments can be 100 times.

[0068] Alternatively, the target engineering feature set may also include: construction method features, number of layers features, column features, and geological features. It should be noted that the preset experiments show that according to the above target engineering feature set, matching the target indicators, and based on the above target indicators, the accuracy of the generated cost prediction results can reach 79.7%.

[0069] It should be noted that the target engineering features included in the target engineering feature set A1 in Table 3 are: construction method features, and the accuracy of the cost prediction results corresponding to the target engineering feature set A1 is 60.5%. The target engineering features included in the target engineering feature set A2 are: number of layers features, and the accuracy of the cost prediction results corresponding to the target engineering feature set A2 is 61.2%. The target engineering features included in the target engineering feature set A3 are: geological features, and the accuracy of the cost prediction results corresponding to the target engineering feature set A3 is 61.6%. The target engineering features included in the target engineering feature set A4 are: column features, and the accuracy of the cost prediction results corresponding to the target engineering feature set A4 is 62.2%. The target engineering features included in the target engineering feature set B1 are: construction method features and number of layers features, and the accuracy of the cost prediction results corresponding to the target engineering feature set B1 is 68.5%. The target engineering features included in the target engineering feature set B2 are: construction method features and geological features, and the accuracy of the cost prediction results corresponding to the target engineering feature set B2 is 70%. The target engineering features included in the target engineering feature set B3 are: construction method features and column features. The accuracy of the cost prediction results corresponding to the target engineering feature set B3 is 75.3%. The target engineering features included in the target engineering feature set B4 are: number of layers and geological features. The accuracy of the cost prediction results corresponding to the target engineering feature set B4 is 68.6%. The target engineering features included in the target engineering feature set B5 are: number of layers and column features. The accuracy of the cost prediction results corresponding to the target engineering feature set B5 is 70.1%. The target engineering features included in the target engineering feature set B6 are: geological features and column features. The accuracy of the cost prediction results corresponding to the target engineering feature set B6 is 67.8%. The target engineering features included in the target engineering feature set C1 are: construction method features, geological features, and number of layers. The accuracy of the cost prediction results corresponding to the target engineering feature set C1 is 70.9%. The target engineering features included in the target engineering feature set C2 are: construction method features, column features, and number of layers features. The accuracy of the cost prediction results corresponding to the target engineering feature set C2 is 80.5%. The target engineering features included in the target engineering feature set C3 are: construction method features, geological features, and column features. The accuracy of the cost prediction results corresponding to the target engineering feature set C3 is 71.4%. The target engineering features included in the target engineering feature set C4 are: construction method features, geological features, and column features. The accuracy of the cost prediction results corresponding to the target engineering feature set C4 is 78.1%. The target engineering features included in the target engineering feature set D are: construction method features, geological features, number of layers, and column features. The accuracy of the cost prediction results corresponding to the target engineering feature set D is 79.7%.

[0070] Table 3: ; For example, when the target project unit is a steel support project unit, it can be determined that the target engineering feature set includes: construction method features, geological features, and column features. It should be noted that the preset number of experiments shows that according to the above target engineering feature set, matching the target indicators, and based on the above target indicators, the accuracy of the generated cost prediction results can reach 89.5%. Among them, the above preset number of experiments can be 100 times.

[0071] Alternatively, the target engineering feature set may also include: construction method features, number of layers features, column features, and geological features. It should be noted that the preset experiments show that according to the above target engineering feature set, matching the target indicators, and based on the above target indicators, the accuracy of the cost prediction results generated can reach 88.9%.

[0072] It should be noted that the target engineering features included in the target engineering feature set A1 in Table 4 are: construction method features, and the accuracy of the cost prediction results corresponding to the target engineering feature set A1 is 75.2%. The target engineering features included in the target engineering feature set A2 are: number of layers features, and the accuracy of the cost prediction results corresponding to the target engineering feature set A2 is 66.1%. The target engineering features included in the target engineering feature set A3 are: geological features, and the accuracy of the cost prediction results corresponding to the target engineering feature set A3 is 72.5%. The target engineering features included in the target engineering feature set A4 are: column features, and the accuracy of the cost prediction results corresponding to the target engineering feature set A4 is 73.4%. The target engineering features included in the target engineering feature set B1 are: construction method features and number of layers features, and the accuracy of the cost prediction results corresponding to the target engineering feature set B1 is 71.9%. The target engineering features included in the target engineering feature set B2 are: construction method features and geological features, and the accuracy of the cost prediction results corresponding to the target engineering feature set B2 is 79.6%. The target engineering features included in the target engineering feature set B3 are: construction method features and column features. The accuracy of the cost prediction results corresponding to the target engineering feature set B3 is 73.9%. The target engineering features included in the target engineering feature set B4 are: number of layers and geological features. The accuracy of the cost prediction results corresponding to the target engineering feature set B4 is 72.8%. The target engineering features included in the target engineering feature set B5 are: number of layers and column features. The accuracy of the cost prediction results corresponding to the target engineering feature set B5 is 73.1%. The target engineering features included in the target engineering feature set B6 are: geological features and column features. The accuracy of the cost prediction results corresponding to the target engineering feature set B6 is 72.4%. The target engineering features included in the target engineering feature set C1 are: construction method features, geological features, and number of layers. The accuracy of the cost prediction results corresponding to the target engineering feature set C1 is 84.8%. The target engineering features included in the target engineering feature set C2 are: construction method features, column features, and number of layers features. The accuracy of the cost prediction results corresponding to the target engineering feature set C2 is 87.4%. The target engineering features included in the target engineering feature set C3 are: construction method features, geological features, and column features. The accuracy of the cost prediction results corresponding to the target engineering feature set C3 is 85.2%. The target engineering features included in the target engineering feature set C4 are: construction method features, geological features, and column features. The accuracy of the cost prediction results corresponding to the target engineering feature set C4 is 89.5%. The target engineering features included in the target engineering feature set D are: construction method features, geological features, number of layers, and column features. The accuracy of the cost prediction results corresponding to the target engineering feature set D is 88.9%.

[0073] Table 4: ; For example, when the target project unit is a concrete support project unit, it can be determined that the target engineering feature set includes: construction method features, geological features, and column features. It should be noted that the preset number of experiments shows that according to the above target engineering feature set, matching the target indicators, and based on the above target indicators, the accuracy of the generated cost prediction results can reach 86.4%. Among them, the above preset number of experiments can be 100 times.

[0074] Alternatively, the target engineering feature set may also include: construction method features, number of layers features, column features, and geological features. It should be noted that the preset experiments show that according to the above target engineering feature set, matching the target indicators, and based on the above target indicators, the accuracy of the generated cost prediction results can reach 89.8%.

[0075] It should be noted that the target engineering features included in the target engineering feature set A1 in Table 5 are: construction method features, and the accuracy of the cost prediction results corresponding to the target engineering feature set A1 is 77.8%. The target engineering features included in the target engineering feature set A2 are: number of layers features, and the accuracy of the cost prediction results corresponding to the target engineering feature set A2 is 73.9%. The target engineering features included in the target engineering feature set A3 are: geological features, and the accuracy of the cost prediction results corresponding to the target engineering feature set A3 is 80.5%. The target engineering features included in the target engineering feature set A4 are: column features, and the accuracy of the cost prediction results corresponding to the target engineering feature set A4 is 79.1%. The target engineering features included in the target engineering feature set B1 are: construction method features and number of layers features, and the accuracy of the cost prediction results corresponding to the target engineering feature set B1 is 82.6%. The target engineering features included in the target engineering feature set B2 are: construction method features and geological features, and the accuracy of the cost prediction results corresponding to the target engineering feature set B2 is 86.9%. The target engineering features included in the target engineering feature set B3 are: construction method features and column features. The accuracy of the cost prediction results corresponding to the target engineering feature set B3 is 85.7%. The target engineering features included in the target engineering feature set B4 are: number of layers and geological features. The accuracy of the cost prediction results corresponding to the target engineering feature set B4 is 85.7%. The target engineering features included in the target engineering feature set B5 are: number of layers and column features. The accuracy of the cost prediction results corresponding to the target engineering feature set B5 is 89%. The target engineering features included in the target engineering feature set B6 are: geological features and column features. The accuracy of the cost prediction results corresponding to the target engineering feature set B6 is 84.5%. The target engineering features included in the target engineering feature set C1 are: construction method features, geological features, and number of layers. The accuracy of the cost prediction results corresponding to the target engineering feature set C1 is 88.5%. The target engineering features included in the target engineering feature set C2 are: construction method features, column features, and number of layers features. The accuracy of the cost prediction results corresponding to the target engineering feature set C2 is 83.1%. The target engineering features included in the target engineering feature set C3 are: construction method features, geological features, and column features. The accuracy of the cost prediction results corresponding to the target engineering feature set C3 is 79.9%. The target engineering features included in the target engineering feature set C4 are: construction method features, geological features, and column features. The accuracy of the cost prediction results corresponding to the target engineering feature set C4 is 86.4%. The target engineering features included in the target engineering feature set D are: construction method features, geological features, number of layers, and column features. The accuracy of the cost prediction results corresponding to the target engineering feature set D is 89.8%.

[0076] Table 5: ; For example, when the target project unit is an internal reinforced concrete structure, the target engineering feature set can be determined to include: construction method features, geological features, and column features. It should be noted that the preset number of experiments shows that according to the above target engineering feature set, matching the target indicators, and based on the above target indicators, the accuracy of the generated cost prediction results can reach 88.1%. Among them, the above preset number of experiments can be 100 times.

[0077] It should be noted that the target engineering features included in the target engineering feature set A1 in Table 5 are: construction method features, and the accuracy of the cost prediction results corresponding to the target engineering feature set A1 is 82.5%. The target engineering features included in the target engineering feature set A2 are: number of layers features, and the accuracy of the cost prediction results corresponding to the target engineering feature set A2 is 76.9%. The target engineering features included in the target engineering feature set A3 are: geological features, and the accuracy of the cost prediction results corresponding to the target engineering feature set A3 is 77.3%. The target engineering features included in the target engineering feature set A4 are: column features, and the accuracy of the cost prediction results corresponding to the target engineering feature set A4 is 72%. The target engineering features included in the target engineering feature set B1 are: construction method features and number of layers features, and the accuracy of the cost prediction results corresponding to the target engineering feature set B1 is 83.6%. The target engineering features included in the target engineering feature set B2 are: construction method features and geological features, and the accuracy of the cost prediction results corresponding to the target engineering feature set B2 is 82.9%. The target engineering features included in the target engineering feature set B3 are: construction method features and column features. The accuracy of the cost prediction results corresponding to the target engineering feature set B3 is 86.2%. The target engineering features included in the target engineering feature set B4 are: number of layers and geological features. The accuracy of the cost prediction results corresponding to the target engineering feature set B4 is 77.3%. The target engineering features included in the target engineering feature set B5 are: number of layers and column features. The accuracy of the cost prediction results corresponding to the target engineering feature set B5 is 72.9%. The target engineering features included in the target engineering feature set B6 are: geological features and column features. The accuracy of the cost prediction results corresponding to the target engineering feature set B6 is 75%. The target engineering features included in the target engineering feature set C1 are: construction method features, geological features, and number of layers. The accuracy of the cost prediction results corresponding to the target engineering feature set C1 is 85.6%. The target engineering features included in the target engineering feature set C2 are: construction method features, column features, and number of layers features. The accuracy of the cost prediction results corresponding to the target engineering feature set C2 is 72.3%. The target engineering features included in the target engineering feature set C3 are: construction method features, geological features, and column features. The accuracy of the cost prediction results corresponding to the target engineering feature set C3 is 75.2%. The target engineering features included in the target engineering feature set C4 are: construction method features, geological features, and column features. The accuracy of the cost prediction results corresponding to the target engineering feature set C4 is 88.1%. The target engineering features included in the target engineering feature set D are: construction method features, geological features, number of layers, and column features. The accuracy of the cost prediction results corresponding to the target engineering feature set D is 84.3%.

[0078] Table 6: ; Based on this, the analysis module provided in the present application can realize accurate analysis of multiple project units included in the target project according to the above-mentioned target data by configuring the first analysis unit, and can realize accurate determination of the construction method characteristics, structural characteristics, and / or geological characteristics corresponding to multiple project units according to the above-mentioned target data by configuring the second analysis unit. By configuring the generation unit, it is possible to accurately generate a target project feature set according to the above-mentioned construction method characteristics, structural characteristics, and / or geological characteristics, thereby providing data support for the matching module to accurately match the target indicators according to the above-mentioned target project characteristics.

[0079] In some feasible implementations, the matching module includes: a matching unit, configured to determine a target indicator and a weight corresponding to the target indicator according to a similarity between the target engineering feature and the preset engineering feature.

[0080] Exemplarily, the preset engineering features can be obtained by using a multilayer perceptron (MLP) to learn historical data corresponding to historical projects within a preset time, wherein the historical data can include: engineering quantity data, quota data, labor, material and machinery data, information price data, and / or cost data, etc.

[0081] Exemplarily, the similarity between the target engineering feature and the preset engineering feature may be determined based on the cosine similarity between the target engineering feature and the preset engineering feature.

[0082] Specifically, the similarity between the above target engineering features and the preset engineering features can be determined based on the following formula: (3); in, Used to represent target engineering features and preset engineering features The similarity between Used to represent target engineering features, Used to represent preset engineering features.

[0083] It should be noted that, when the similarity between the target engineering feature and the preset engineering feature is greater than or equal to the preset similarity threshold, the target indicator corresponding to the target engineering feature can be determined according to the target indicator corresponding to the preset engineering feature. Among them, the above-mentioned preset similarity threshold can be determined based on the accuracy requirement of the cost prediction result of the target user. Specifically, the above-mentioned preset similarity threshold is positively correlated with the accuracy requirement of the cost prediction result of the target user, that is, the higher the accuracy requirement of the cost prediction result of the target user, the higher the preset similarity threshold.

[0084] It should be noted that the weight of the above target indicator can be determined based on the similarity between the preset engineering feature corresponding to the target indicator and the target engineering feature, and / or the similarity between the preset engineering corresponding to the preset engineering feature and the target engineering. The weight of the above target indicator is positively correlated with the similarity between the preset engineering feature and the target engineering feature, and / or the similarity between the preset engineering corresponding to the preset engineering feature and the target engineering, that is, the higher the similarity between the preset engineering feature and the target engineering feature, and / or the similarity between the preset engineering corresponding to the preset engineering feature and the target engineering, the higher the weight corresponding to the target indicator.

[0085] Specifically, the output results of the target indicator and the weight corresponding to the target indicator can be determined by neurons based on the following formulas: (4); in, It is used to represent the output result of the target indicator and the weight corresponding to the target indicator. For the The weight vector corresponding to the target indicator, Output vector for target engineering features, For the The bias term corresponding to the target indicator.

[0086] Based on this, the matching module provided in the present application can realize the accurate determination of the target indicators and the weights corresponding to the target indicators according to the similarity between the target engineering features and the preset engineering features by configuring the above-mentioned matching units, thereby providing data support for the subsequent accurate generation of cost prediction results according to the target indicators, thereby further improving the generation accuracy and efficiency of the cost prediction results.

[0087] In some feasible implementations, the above-mentioned target indicators include: necessary indicators, and / or reference indicators; wherein the necessary indicators include: labor cost indicators, material cost indicators, and / or equipment cost indicators; the reference indicators include: tax indicators.

[0088] It should be noted that the above-mentioned reference indicators may also include: regulatory fee cost indicators, safety measures cost indicators, lump sum fee cost indicators, and / or independent fee cost indicators, etc.

[0089] Specifically, the cost prediction result can be determined based on the above necessary indicators, reference indicators and corresponding weights.

[0090] In some feasible implementations, an indicator structure tree can be generated based on the above necessary indicators, reference indicators and corresponding weights. A line indicator structure tree can also be further generated based on the corresponding line positions of multiple target project units included in the target project and the corresponding necessary indicators, reference indicators and corresponding weights. It should be noted that by sending the above indicator structure tree or the above line indicator structure tree to the target user client, it is helpful to assist the target user to understand the various indicator data involved in the target project cost prediction results, so as to facilitate the target user to understand the detailed cost structure of the target project.

[0091] It should be noted that by limiting the above-mentioned target indicators to include necessary indicators and / or reference indicators, wherein the necessary indicators include labor cost indicators, material cost indicators, and / or equipment cost indicators, and the reference indicators include tax indicators, it is conducive to facilitating the further refinement of cost forecast results.

[0092] In some feasible embodiments, the above-mentioned prediction module includes: a first prediction unit, which is used to generate a cost prediction result based on necessary indicators and corresponding weights when it is determined that the accuracy requirement of the cost prediction result is less than or equal to the preset accuracy requirement, and / or when it is determined that the feedback time of the cost prediction result is less than or equal to the preset time.

[0093] It should be noted that the above cost prediction result accuracy requirement and / or the above prediction result feedback duration may be determined by manual input by the target user. The above preset accuracy requirement and / or the above preset duration may be set according to actual needs.

[0094] Based on this, when it is determined that the accuracy requirement of the cost prediction result is less than or equal to the preset accuracy requirement based on the data manually input by the target user, and / or the cost prediction result feedback time is less than or equal to the preset time, the cost prediction result is generated according to the labor cost indicator, material cost indicator, and / or equipment cost indicator and the corresponding weights, which is beneficial to improve the output efficiency of the cost prediction results corresponding to the target project with lower requirements for the accuracy of the cost prediction results and / or a shorter time limit for generating the cost prediction results, and shorten the response time of the engineering cost prediction model of this application.

[0095] In some feasible embodiments, the above-mentioned prediction module also includes: a second prediction unit, which is used to generate a cost prediction result based on necessary indicators, reference indicators and corresponding weights when it is determined that the accuracy requirement of the cost prediction result is greater than the preset accuracy requirement, and / or when it is determined that the feedback time of the cost prediction result is greater than the preset time.

[0096] It should be noted that the above cost prediction result accuracy requirement and / or the above prediction result feedback duration may be determined by manual input by the target user. The above preset accuracy requirement and / or the above preset duration may be set according to actual needs.

[0097] Based on this, when it is determined according to the data manually input by the target user that the accuracy requirement of the cost forecast result is greater than the preset accuracy requirement, and / or the cost forecast result feedback time is longer than the preset time, the cost forecast result is generated according to the labor cost indicator, material cost indicator, equipment cost indicator, tax indicator and corresponding weights, which is conducive to improving the output accuracy of the cost forecast result corresponding to the target project that meets the target user's high demand for the accuracy of the cost forecast result and / or has a long time limit for cost forecast result generation, thereby improving the reference value of the cost forecast result.

[0098] In some feasible implementations, the construction cost prediction model as described in any of the above items further includes: an evaluation module for generating an accuracy evaluation result of the construction cost prediction model based on the square difference between the actual cost result and the cost prediction result.

[0099] It should be noted that the accuracy evaluation operation can be performed on the engineering cost prediction model of the present application according to the square difference between the actual cost result of the target project fed back by the user and the recorded cost prediction result of the target project.

[0100] Specifically, the above accuracy evaluation result can be generated according to the following formula: (5); in, is the accuracy evaluation result, is the sample size, For the The predicted value of samples, For the The true value of the samples.

[0101] Based on this, the engineering cost prediction model provided by the present application can realize the square difference between the actual cost result and the cost prediction result by configuring the above-mentioned evaluation module. Based on the above-mentioned formula (5), the accuracy of the engineering cost prediction model can be periodically determined according to the mean square error, which is conducive to facilitating the subsequent update module to perform update operations on the above-mentioned engineering cost prediction model based on the accuracy evaluation results.

[0102] In some feasible implementations, the above-mentioned project cost prediction model further includes: an update module, which is used to perform a target update operation on the project cost prediction model based on a preset objective function when the accuracy evaluation result is less than or equal to a preset accuracy threshold.

[0103] The above-mentioned preset accuracy threshold is manually input and set by the target user.

[0104] It should be noted that, when the accuracy result of the engineering cost prediction model determined according to the above formula (5) is less than or equal to the above preset accuracy threshold, the target update operation can be performed based on the following preset objective function: (6); in, For the The target parameters for the step update, For the The target parameters for the step update, is the target learning rate, is the preset constant, For the Step target first-order moment estimation momentum, No. Step target second moment estimation momentum.

[0105] It should be noted that the above preset constants Can be set to .

[0106] Among them, the above Step target first-order moment estimation momentum It can be determined based on the following formula: (7); in, For the Step target first-order moment estimation momentum, For the The first-order moment estimate of momentum, For the Step 1 deviation correction threshold.

[0107] Among them, Step-first moment estimation of momentum It can be determined based on the following formula: (8); in, For the The first-order moment estimate of momentum, For the The first-order moment estimate of momentum, is the momentum coefficient, For the The target gradient of the step.

[0108] It should be noted that the momentum coefficient Can be set to 0.9.

[0109] Among them, the above Step target second moment estimation momentum It can be determined based on the following formula: (9); Among them, the above For the Step target second-order moment estimation momentum, For the Step second moment estimation momentum, For the Step 2 deviation correction threshold.

[0110] Among them, the above Step-by-step second-order moment estimation of momentum It can be determined based on the following formula: (10); in, For the Step second moment estimation momentum, is the variance coefficient, For the Step second moment estimation momentum, For the The target gradient of the step.

[0111] It should be noted that the above variance coefficient Can be set to 0.999.

[0112] Based on this, the present application can configure the above-mentioned update module to implement a back-propagation-based method to determine the gradient of each parameter step by step, so as to perform a target update operation on each parameter through a step-by-step gradient clearing operation, so as to perform a target update operation on the above-mentioned engineering cost prediction model, thereby further improving the accuracy of the engineering cost prediction model.

[0113] It should be noted that the engineering cost prediction model described in this application can be implemented based on a neural network model. The engineering cost prediction model described in this application can be provided with an input layer, a hidden layer and an output layer to achieve the above-mentioned beneficial technical effects.

[0114] The second aspect of the embodiment of the present application provides a method for predicting the construction cost. For example, Figure 2 A schematic diagram of a process of predicting a construction cost 200 provided in an embodiment of the present application is shown in FIG. Figure 2 As shown, the method 200 includes: Step S210, obtaining target data, wherein the target data includes: engineering quantity data, quota data, labor, material and machine data, and / or information price data; Step S220, analyzing target engineering features according to target data; Step S230, matching target indicators according to target engineering features; Step S240, generating a cost prediction result according to the target indicator.

[0115] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the described method can refer to the corresponding process in the aforementioned model embodiment, and will not be repeated here.

[0116] Figure 3 The structure diagram of a terminal device or server suitable for implementing the embodiment of the present application is shown. Figure 3 As shown, the electronic device 300 includes a central processing unit (CPU) 301, which can perform various appropriate actions and processes according to the program stored in the read-only memory (ROM) 302 or the program loaded from the storage part 308 into the random access memory (RAM) 303. In the RAM 303, various programs and data required for the operation of the terminal device or the server are also stored. The CPU 301, the ROM 302, and the RAM 303 are connected to each other via a bus 304. An input / output (I / O) interface 305 is also connected to the bus 304.

[0117] The following components are connected to the I / O interface 305: an input section 306 including a keyboard, a mouse, etc.; an output section 307 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 308 including a hard disk, etc.; and a communication section 309 including a network interface card such as a LAN card, a modem, etc. The communication section 309 performs communication processing via a network such as the Internet. A drive 310 is also connected to the I / O interface 305 as needed. A removable medium 311, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 310 as needed, so that a computer program read therefrom is installed into the storage section 308 as needed.

[0118] In particular, according to an embodiment of the present application, the above method flow steps can be implemented as a computer software program. For example, an embodiment of the present application includes a computer program product, which includes a computer program carried on a machine-readable medium, and the computer program includes a program code for executing the method shown in the flow chart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 309, and / or installed from the removable medium 311. When the computer program is executed by the central processing unit (CPU) 301, the above-mentioned functions defined in the model of the present application are executed.

[0119] It should be noted that the computer-readable medium described in the present application may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor model, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present application, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution model, device or device. In the present application, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution model, apparatus or device. The program code contained on the computer-readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, optical cable, RF, etc., or any suitable combination of the above.

[0120] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the model, method and computer program product according to various embodiments of the present application. In this regard, each box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the aforementioned module, a program segment or a part of a code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a sequence different from that marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flow chart, and the combination of the boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based model that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0121] The units or modules involved in the embodiments described in the present application may be implemented by software or hardware. The units or modules described may also be arranged in a processor. The names of these units or modules do not, in some cases, constitute limitations on the units or modules themselves.

[0122] The above description is only a preferred embodiment of the present application and an explanation of the technical principles used. Those skilled in the art should understand that the scope of application involved in the present application is not limited to the technical solution formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the aforementioned application concept. For example, the above features are replaced with (but not limited to) technical features with similar functions applied in the present application.

Claims

1. A project cost prediction model, characterized in that: include: An acquisition module, used for acquiring target data, wherein the target data includes: engineering quantity data, quota data, labor, material and machine data, and / or information price data; A parsing module, used for parsing target engineering features according to the target data; A matching module, used to match target indicators according to the target engineering characteristics; The prediction module is used to generate cost prediction results according to the target indicators.

2. The project cost prediction model according to claim 1 is characterized in that: The acquisition module is provided with a plurality of target interfaces for acquiring the target data.

3. The project cost prediction model according to claim 1 is characterized in that: The parsing module comprises: A preprocessing unit is used to perform a preprocessing operation on the target data.

4. The project cost prediction model according to claim 1 is characterized in that: The parsing module also includes: A first parsing unit, configured to determine a plurality of project units included in a target project according to the target data; A second parsing unit is used to determine the construction method characteristics, structural characteristics, and / or geological characteristics corresponding to the multiple project units according to the target data; A generating unit is used to generate a target engineering feature set according to the construction method characteristics, structural characteristics, and / or geological characteristics.

5. The project cost prediction model according to claim 1 is characterized in that: The matching module comprises: The matching unit is used to determine the target indicator and the weight corresponding to the target indicator according to the similarity between the target engineering feature and the preset engineering feature.

6. The project cost prediction model according to claim 5 is characterized in that: The target indicators include: necessary indicators, and / or reference indicators; Wherein, the necessary indicators include: labor cost indicators, material cost indicators, and / or equipment cost indicators; The reference indicators include: tax payment indicators.

7. The project cost prediction model according to claim 6 is characterized in that: The prediction module comprises: A first prediction unit, configured to generate the cost prediction result according to the necessary indicators and corresponding weights when it is determined that the accuracy requirement of the cost prediction result is less than or equal to the preset accuracy requirement and / or when it is determined that the feedback time of the cost prediction result is less than or equal to the preset time; The second prediction unit is used to generate the cost prediction result according to the necessary indicators, the reference indicators and the corresponding weights when it is determined that the accuracy requirement of the cost prediction result is greater than the preset accuracy requirement and / or when it is determined that the feedback time of the cost prediction result is greater than the preset time.

8. The project cost prediction model according to any one of claims 1 to 7, characterized in that: Also includes: An evaluation module, used to generate an accuracy evaluation result of the project cost prediction model according to the square difference between the actual cost result and the cost prediction result; Wherein, generating the accuracy evaluation result of the engineering cost prediction model according to the square difference between the actual cost result and the cost prediction result includes: generating the accuracy evaluation result according to the following formula: ； in, is the accuracy evaluation result, is the sample size, For the The predicted value of samples, For the The true value of the samples.

9. The project cost prediction model according to claim 8, characterized in that: Also includes: An updating module, configured to perform a target updating operation on the project cost prediction model based on a preset objective function when the accuracy evaluation result is less than or equal to a preset accuracy threshold; The preset objective function includes the following functions: ； in, For the The target parameters for the step update, For the The target parameters for the step update, is the target learning rate, is the preset constant, For the Step target first-order moment estimation momentum, No. Step target second moment estimation momentum.

10. A method for predicting construction cost, characterized in that: include: Acquiring target data, wherein the target data includes: engineering quantity data, quota data, labor, material and machine data, and / or information price data; Analyzing target engineering features according to the target data; Matching target indicators according to the target engineering characteristics; According to the target indicator, a cost prediction result is generated.

Images