Engineering data index analysis method and device, equipment and storage medium

Abstract

The embodiment of the invention discloses an engineering data index analysis method and device, equipment and a storage medium, which are used for carrying out index classification on engineering data through a preset rule, predicting the condition that the preset rule cannot be identified by using a preset prediction model under the condition that an identification result is unsuccessful, and assisting with manual checking, so that the engineering data index analysis efficiency is improved. And the accuracy of engineering data classification and identification is improved. The method provided by the embodiment of the invention comprises the steps of obtaining engineering data; performing classification and identification on the engineering data by using a preset rule to obtain an identification result; under the condition that the identification result is successful, displaying the index classification corresponding to the identification result; or, under the condition that the identification result is unsuccessful, inputting the engineering data features into a first preset prediction model, outputting a prediction result, and under the condition that the precision value of the prediction result is greater than a preset threshold value, displaying the index classification corresponding to the prediction result.

Description

Technical Field

[0001] This application relates to the field of engineering construction, and particularly to a method, device, equipment and storage medium for analyzing engineering data indicators. Background Art

[0002] In construction projects, there are usually manual compilation mode, expert rule mode and machine learning mode for analyzing engineering data indicators. However, the manual compilation mode is time-consuming and laborious, relying on the experience and ability of the compiler, and ordinary people cannot directly complete the classification work. The expert rule mode is limited by the expert experience of the compilation rules and often cannot cover various project feature situations. The machine learning mode depends on data quality: incorrect classified data will affect the accuracy of model prediction. Summary of the Invention

[0003] Embodiments of this application provide a method, device, equipment and storage medium for analyzing engineering data indicators, which are used to classify the engineering data by preset rules, and in the case of unsuccessful recognition results, use a preset prediction model to predict the situations that cannot be recognized by the preset rules, and assist with manual verification to improve the accuracy of engineering data classification and recognition.

[0004] The first aspect of this application provides a method for analyzing engineering data indicators, which may include:

[0005] Obtain engineering data;

[0006] Classify and recognize the engineering data using preset rules to obtain a recognition result;

[0007] In the case where the recognition result is successful, display the index classification corresponding to the recognition result; or,

[0008] In the case where the recognition result is unsuccessful, input the engineering data into a first preset prediction model, output a prediction result, and in the case where the accuracy value of the prediction result is greater than a preset threshold, display the index classification corresponding to the prediction result.

[0009] In some possible implementation manners of this application, the classifying and recognizing the engineering data using preset rules to obtain a recognition result may include: extracting features from the engineering data to obtain engineering feature data; in response to the selection of a target index analysis rule template, using the preset rule corresponding to the target index analysis rule template to classify and recognize the engineering feature data to obtain a recognition result.

[0010] In some possible implementation manners of this application, the method may further include:

[0011] In response to the confirmation operation on the recognition result, the current classification process ends; or,

[0012] Respond to the modification operation on the recognition result to obtain a modified recognition result, and display the modified recognition result.

[0013] In some possible implementation manners of the present application, the method may further include:

[0014] Respond to the confirmation operation on the prediction result, and the current classification process ends; or,

[0015] Respond to the modification operation on the prediction result to obtain a modified prediction result, and display the modified prediction result.

[0016] In some possible implementation manners of the present application, the method may further include: when the accuracy value of the prediction result is less than or equal to the preset threshold, display an unclassified indication message; respond to the classification operation on the engineering data to obtain a classification result.

[0017] In some possible implementation manners of the present application, the method may further include: obtaining classified engineering data; performing feature extraction on the classified engineering data to obtain classified engineering feature data; performing model training on the classified engineering feature data to obtain a second preset prediction model.

[0018] In some possible implementation manners of the present application, the method may further include: determining verification data according to the classified engineering feature data; inputting the verification data into the first preset prediction model to obtain a first output classification result; determining a first error between the first output classification result and the classification result of the verification data; inputting the verification data into the second preset prediction model to obtain a second output classification result; determining a second error between the second output classification result and the classification result of the verification data; when the first error is greater than the second error, updating the first preset prediction model to the second preset prediction model; when the first error is less than or equal to the second error, not performing an update process.

[0019] In a third aspect of the present application, a terminal device is provided, including a memory, one or more processors, and a display. The memory is coupled to the processor. The memory is used to store computer program code, and the computer program code includes computer instructions. When the processor executes the computer instructions, the terminal device executes the method according to any one of the first aspect.

[0020] In the fourth aspect of the present application, a chip is provided. The chip includes a processor, a memory, and a display. The processor can be a logic circuit, an integrated circuit, or a general-purpose processor, etc. Instructions are stored in the memory. The processor can implement the method in the first aspect by reading the software code stored in the memory. The memory can be integrated in the processor or can exist independently outside the processor.

[0021] In the fifth aspect of the present application, a chip system is provided. The chip system is applied to a terminal device. The chip system includes one or more interface circuits, one or more processors, and a display. The interface circuits and the processors are interconnected by lines. The interface circuits are used to receive signals from the memory of the terminal device and send the signals to the processors. The signals include computer instructions stored in the memory. When the processors execute the computer instructions, the terminal device executes the method according to any one of the first aspect.

[0022] In the sixth aspect of the present application, a computer program product is provided. The computer program product includes a computer program (which can also be called code or instructions). When the computer program is run, the computer is caused to execute the method in the first aspect.

[0023] In the seventh aspect of the present application, a computer-readable storage medium is provided. The computer-readable storage medium stores a computer program (which can also be called code or instructions). When it runs on a computer, the computer is caused to execute the method in the first aspect.

[0024] As can be seen from the above technical solutions, the embodiments of the present application have the following advantages:

[0025] In this technical solution, engineering data is obtained; the engineering data is classified and identified using preset rules to obtain an identification result; in the case where the identification result is successful, the index classification corresponding to the identification result is displayed; or, in the case where the identification result is unsuccessful, the engineering data is input into a first preset prediction model to output a prediction result. In the case where the accuracy value of the prediction result is greater than a preset threshold, the index classification corresponding to the prediction result is displayed. It is used to classify the indicators of engineering data by preset rules. In the case where the identification result is successful, the index classification corresponding to the identification result is displayed. In the case where the identification result is unsuccessful, a preset prediction model is used to predict the situation that cannot be identified by the preset rules, and assisted by manual verification to improve the accuracy of the classification and identification of engineering data. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] To more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings required for the description of the embodiments and the prior art. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained according to these drawings.

[0027] Figure 1 Schematic diagram of an embodiment of the method for analyzing engineering data indicators in an embodiment of the present application;

[0028] Figure 2 Schematic flow diagram of the method for analyzing engineering data indicators in an embodiment of the present application;

[0029] Figure 3 Another schematic diagram of an embodiment of the method for analyzing engineering data indicators in an embodiment of the present application;

[0030] Figure 4A Schematic flow diagram of a process for training a second preset prediction model in an embodiment of the present application;

[0031] Figure 4B Schematic diagram of a classified engineering data in an embodiment of the present application;

[0032] Figure 4C Another schematic diagram of a classified engineering data in an embodiment of the present application;

[0033] Figure 4D Another schematic diagram of a classified engineering data in an embodiment of the present application;

[0034] Figure 5 Schematic diagram of an embodiment of the device for analyzing engineering data indicators in an embodiment of the present application;

[0035] Figure 6 Schematic diagram of an embodiment of an electronic device in an embodiment of the present application;

[0036] Figure 7 Another schematic diagram of an embodiment of an electronic device in an embodiment of the present application. Detailed implementation manners

[0037] The embodiments of the present application provide a method, device, equipment and storage medium for analyzing engineering data indicators, which are used to classify the indicators of engineering data through preset rules. In the case where the recognition result is unsuccessful, a preset prediction model is used to predict the situation that cannot be recognized by the preset rules, and manual verification is assisted to improve the accuracy of engineering data classification and recognition.

[0038] To enable those skilled in the art to better understand the solution of this application, the technical solutions in the embodiments of this application will be described below in conjunction with the accompanying drawings in the embodiments of this application. Obviously, the described embodiments are only a part of the embodiments of this application, rather than all the embodiments. Based on the embodiments in this application, they should all fall within the scope protected by this application.

[0039] First, a brief explanation of some related terms involved in this application is given as follows:

[0040] 1) Machine learning is a technology that enables a computer to automatically complete tasks and optimize algorithms by learning data and patterns.

[0041] 2) Expert rules are rules formed by collecting and summarizing relevant field experts, and are organized into a computer-readable configuration file in a custom format, and then the computer is instructed to automatically identify and process according to the set configuration.

[0042] 3) Project cost indicators refer to a series of indicators used to evaluate and control project costs in construction projects. It includes cost indicators, quantity indicators, unit price indicators, etc., and mainly reflects the average cost level per unit (such as building area, road length, etc.).

[0043] Single project: In a construction project, a project with independent design documents, capable of independent construction organization, and that can independently play a production capacity or use function after completion.

[0044] Unit project: A unit project refers to a smaller project unit within a single project that can be independently organized for construction and can be independently used or form a production capacity after completion. A unit project is usually a component of a single project.

[0045] Sub - project: A sub - project refers to a smaller project part divided according to professional nature or construction location within a unit project. It is a further subdivision of a unit project.

[0046] Bill of quantities: A detailed list of the names and corresponding quantities of sub - items, measure items, and other items in a construction project, as well as items such as fees and taxes. The bill of quantities generally includes project codes, project names, project characteristics, measurement units, and quantities. It is generally abbreviated as "the list".

[0047] Quota: It refers to the quantity standard of resources that must be consumed to complete a unit of qualified products under normal construction conditions, reasonable labor organization, reasonable use of materials and machinery.

[0048] Consumption: Under specified working conditions, it is the quantity standard of labor, materials, machinery, equipment, and related expenses required to complete a qualified unit of building and installation products.

[0049] Composition of engineering projects: In descending order, it includes project, single project, unit project, sub - project, bill of quantities, quota, consumption.

[0050] Cost index: According to business requirements, classify each cost - component element in the engineering project, then conduct numerical statistics by classification, and finally divide the statistical result by the construction scale (commonly floor area, road length, etc.) to obtain average data with reference significance.

[0051] 1. Manual compilation mode

[0052] Traditional index analysis usually classifies in a manual way. After receiving the project data, cost estimators classify manually through information such as the project hierarchy, descriptions in the bill of quantities, and names and specifications of materials and machinery, and extract the quantities and cost information under each classification. Then, they statistically calculate the total data of the classification by accumulation. Finally, divide the total value in the total data by the construction scale (such as floor area, road length) to obtain the average cost index. The calculated average cost index also needs to be manually filled into the index analysis form to form a visual result.

[0053] Main disadvantages: Time - consuming and labor - intensive. Usually, it takes 5 working days to complete the analysis of one project. Relying on the experience and ability of the compiler, ordinary people cannot directly complete the classification work.

[0054] 2. Expert rule mode

[0055] Relying on the experience of some experts in the cost field with rich experience, form the form of index analysis rules, extract keywords in various descriptions, convert them into a rule - description language that can be processed by a specific computer, and then through a dedicated rule engine, realize the automatic extraction and classification of engineering indicators, and further obtain the index values of various types through statistical calculation.

[0056] Generally, experts describe rules in a natural - language way. For example, for the index item of "earthwork project", the bills with list codes within 010101002 - 010101006 and 010102001 - 010102004 should be selected for statistics.

[0057] Natural language usually needs to be further converted into a machine - recognizable structured data form, such as:

[0058] Index [earthwork project]: {type [list]; code [010101002 - 010101006; 010102001 - 010102004];}

[0059] In order to handle more complex logic, more operation judgments need to be introduced, such as greater than, greater than or equal to, less than, less than or equal to, equal to, not equal to, contain, not contain, etc.

[0060] At the same time, logical judgments need to be introduced, such as AND and OR.

[0061] The disadvantage of the rule model is that it is limited by the expert experience of the rule compilation and often cannot cover the characteristics of various projects: in different engineering cost projects, the keyword descriptions of the compilers are not standardized and unified, and it is impossible to enumerate all the keywords. There are specific professional terms, such as synonyms such as "concrete" and "concrete", which are difficult to enumerate completely. Some elements will have different indicator classifications in different business environments. For example, "stainless steel pipe" can be classified as "water supply pipe" and "fire water pipe" at the same time.

[0062] 3. Machine Learning Model

[0063] Text classification is an important field in machine learning. By sorting out the classified indicator data, preprocessing the classified data through the word segmentation algorithm, extracting the preprocessed text features, and forming an indicator classification model. When using it, load the corresponding model and input any element. The algorithm will predict the most likely classification through calculation, thereby realizing automatic classification of indicators.

[0064] Main disadvantages: Requires data for training: Users have basically no ready-made indicator classification data and cannot effectively train the model; Relies on a large amount of data: The training sample data needs to be sufficient to make the classification calculation more accurate; Relies on data quality: Incorrect classification data will affect the accuracy of model predictions.

[0065] In the embodiment of the present application, the electronic device may be a computer or a terminal device. The terminal device may be a mobile phone, a tablet computer, a computer with wireless transceiver function, a virtual reality (VR) terminal device, an augmented reality (AR) terminal device, a wireless terminal device in industrial control, a wireless terminal device in self-driving, a wireless terminal device in remote medical, a wireless terminal device in smart grid, a wireless terminal device in transportation safety, a wireless terminal device in smart city, or a wireless terminal device in smart home, etc.

[0066] By way of example and not limitation, in the embodiments of the present application, the terminal device may also be a wearable device with a display interface. A wearable device, also known as a wearable intelligent device, is a general term for devices developed by applying wearable technology to the intelligent design of daily wear, such as glasses, gloves, watches, clothing, and shoes. A wearable device is a portable device that is directly worn on the body or integrated into the user's clothes or accessories. A wearable device is not just a hardware device, but also realizes powerful functions through software support, data interaction, and cloud interaction. Broadly speaking, wearable intelligent devices include those with complete functions and large sizes that can achieve complete or partial functions without relying on a smart phone, such as smart watches or smart glasses, as well as those that only focus on a certain type of application function and need to cooperate with other devices such as smart phones, such as various smart bracelets and smart jewelry for physical sign monitoring.

[0067] Next, the technical solution of the present application will be further described by way of embodiments, as Figure 1 shown, which is a schematic diagram of an embodiment of the method for analyzing engineering data indicators in the embodiments of the present application, and may include:

[0068] 101. Obtain engineering data.

[0069] In some possible implementation manners, the obtaining of the engineering data may include: obtaining an engineering file; performing file parsing on the engineering file to extract the engineering data.

[0070] In some possible implementation manners, the obtaining of the engineering file may include: responding to an upload operation of the engineering file to obtain the engineering file.

[0071] In some possible implementation manners, the method may further include: storing the engineering data in a database.

[0072] Exemplarily, the engineering file may be referred to as a cost file, and the engineering data may also be referred to as cost data. That is, the user uploads the cost file, and the electronic device performs file parsing on the cost file to extract the cost data.

[0073] In the present technical solution, the engineering data can be extracted by parsing the engineering file, which is convenient for subsequent classification and recognition of the engineering data using preset rules to obtain a recognition result.

[0074] 102. Use preset rules to classify and recognize the engineering data to obtain a recognition result.

[0075] In some possible implementation manners, classifying and identifying the project data by using a preset rule to obtain an identification result may include: extracting features of the project data to obtain project feature data; in response to the selection of a target index analysis rule template, using the preset rule corresponding to the target index analysis rule template to classify and identify the project feature data to obtain an identification result.

[0076] Exemplarily, the preset rule here can be understood as an expert rule. That is, it is an index analysis rule formed by organizing the experiences of some experts with rich experience in the cost field. The electronic device analyzes the project data, extracts the feature data in the project data to obtain project feature data; the user can also select an index analysis rule template, and in response to the selection of the target index analysis rule template, the electronic device uses the preset rule corresponding to the target index analysis rule template to classify and identify the project feature data to obtain an identification result. Among them, the index analysis rule template can also be abbreviated as the index rule template.

[0077] In the technical solution, in the index classification automatic identification process, the identification result is first generated by an expert rule with relatively high accuracy, which improves the accuracy of algorithm identification.

[0078] For the extraction of the cost data features of this application, different eigenvalue contents are set according to different types of cost data, as follows:

[0079] Single project: name;

[0080] Unit project: name, professional type, attribution information;

[0081] Sub-project: name, attribution information;

[0082] Bill of quantities: code, name, features, unit, attribution information;

[0083] Quota: code, name, unit, specialty, attribution information;

[0084] Material: name, specification, unit, attribution information.

[0085] Among them, the attribution information is constructed in the order of the cost data type hierarchy (single project > unit project > sub-project > bill of quantities > quota > material), and the format is: single project name@unit project name professional type@sub-project name@bill of quantities code_bill of quantities name_bill of quantities features@quota specialty_quota code.

[0086] The content of the un-reached level does not need to be described, as follows:

[0087] The attribution information of the bill of quantities needs to be described: single project name@unit project name_professional type@sub-project name;

[0088] The attribution information of the quota needs to be described: name of the individual project @ name of the unit project _ professional type @ name of the sub - project @ list code _ list name _ list characteristics.

[0089] In this application, the expert rules of the indicators can be defined in JSON format as follows:

[0090]

[0091]

[0092] Format definition description:

[0093] Basic definition of the indicator rule template: name, version, description, publisher, release time.

[0094] Indicator item definition: indicator item name, unit, description, calculation caliber, source, etc.

[0095] Source definition: data type, statistical field, statistical method and statistical conditions.

[0096] Statistical condition definition: condition logical relationship, conditional formula, grouping.

[0097] Conditional formula definition: field, field conditional value, operator.

[0098] 103. When the recognition result is successful, display the indicator classification corresponding to the recognition result.

[0099] 104. When the recognition result is unsuccessful, input the project data into the first preset prediction model, output the prediction result, and when the accuracy value of the prediction result is greater than the preset threshold, display the indicator classification corresponding to the prediction result.

[0100] Exemplarily, the recognition result is the result of the classification recognition of project data.

[0101] When the recognition result is successful, display the indicator classification corresponding to the recognition result on the display interface of the electronic device, that is, set the corresponding indicator classification according to the recognition result. Optionally, when the recognition result is successful, the recognition result is used to update the first preset prediction model.

[0102] When the recognition result is unsuccessful, input the project data into the first preset prediction model for model prediction, output the prediction result. If the accuracy value in the prediction result, that is, the prediction accuracy, is greater than the preset threshold, display the indicator classification corresponding to the prediction result on the display interface of the electronic interface, that is, set the corresponding indicator classification according to the prediction result.

[0103] In some possible implementations, the method further includes:

[0104] In response to the confirmation operation on the metric classification corresponding to the recognition result, the current classification process ends; or,

[0105] In response to the modification operation on the metric classification corresponding to the recognition result, a modified recognition result is obtained, and the modified recognition result is displayed.

[0106] Exemplarily, after the metric classification corresponding to the recognition result is displayed on the display interface of the electronic interface, if it meets the expectation, that is, the user performs a confirmation operation on the recognition result. If the metric classification corresponding to the recognition result is confirmed to be correct, the current classification process ends. If it does not meet the expectation, that is, the user modifies the metric classification corresponding to the recognition result, the electronic device responds to the modification operation on the metric classification corresponding to the recognition result, obtains a modified recognition result, and displays the modified recognition result. Optionally, in the case where the recognition result is unsuccessful, the modified recognition result is used to update the first preset prediction model.

[0107] In the technical solution of the present application, in the case where the recognition result is successful, after the metric classification corresponding to the recognition result is displayed, in response to the confirmation operation on the metric classification corresponding to the recognition result, the current classification process ends; or, in response to the modification operation on the metric classification corresponding to the recognition result, a modified recognition result is obtained, and the modified recognition result is displayed, facilitating the user to confirm whether the metric classification is accurate.

[0108] In some possible implementations, the method further includes:

[0109] In response to the confirmation operation on the metric classification corresponding to the prediction result, the current classification process ends; or,

[0110] In response to the modification operation on the metric classification corresponding to the prediction result, a modified prediction result is obtained, and the modified prediction result is displayed.

[0111] Exemplarily, after the metric classification corresponding to the prediction result is displayed on the display interface of the electronic interface, if it meets the expectation, that is, the user performs a confirmation operation on the prediction result. If the metric classification corresponding to the prediction result is confirmed to be correct, the current classification process ends. If it does not meet the expectation, that is, the user modifies the metric classification corresponding to the prediction result, the electronic device responds to the modification operation on the metric classification corresponding to the prediction result, obtains a modified prediction result, and displays the modified prediction result. Optionally, the modified prediction result is used to update the first preset prediction model.

[0112] In the present technical solution, when the prediction result is successful, after presenting the metric classification corresponding to the prediction result, in response to an operation to confirm the metric classification corresponding to the prediction result, the current classification process ends; or, in response to an operation to modify the metric classification corresponding to the prediction result, a modified prediction result is obtained, and the modified prediction result is presented to facilitate the user to confirm whether the metric classification is accurate.

[0113] In some possible implementation manners, the method may further include: when the accuracy value of the prediction result is less than or equal to the preset threshold, presenting an indication information of unclassified; in response to an operation to classify the engineering data, a classification result is obtained.

[0114] Exemplarily, when the accuracy value of the prediction result, that is, the prediction accuracy, is less than or equal to the preset threshold, it is marked as unrecognized, or marked as unclassified, and the indication information of unclassified is presented; the user can classify the engineering data according to the presented indication information of unclassified, and the electronic device obtains a classification result in response to the operation to classify the engineering data. Optionally, the classification result is used to update a first preset prediction model.

[0115] In the present technical solution, when the accuracy value of the prediction result is less than the preset threshold, it indicates that the reliability is relatively low. By presenting the indication information of unclassified, the user can manually classify to ensure the accuracy rate of the classification result of the engineering data.

[0116] As Figure 2 shown, it is a schematic flowchart of a method for analyzing engineering data metrics in an embodiment of the present application. 1) After the user uploads a cost file, an index classification template can be selected. 2) The system of the electronic device analyzes and extracts the data in the cost file to obtain engineering cost data, and extracts the characteristic data in the engineering cost data. 3) The system of the electronic device loads the configured expert rule file to automatically classify and identify the engineering cost data. 4) The unrecognized classification is handed over to a pre-trained recognition model for prediction. When the accuracy value of the prediction result is greater than the preset threshold, the classification by machine recognition is adopted; when the accuracy value of the prediction result is less than or equal to the preset threshold, it is marked as unrecognized. 5) The user checks the recognized classification and can make manual adjustments if it does not meet the expectation. 6) The system records the relevant adjustments for later model training.

[0117] In an embodiment of the present application, engineering data is obtained; the engineering data is classified and identified using a preset rule to obtain an identification result; in the case where the identification result is successful, the index classification corresponding to the identification result is displayed; or, in the case where the identification result is unsuccessful, the engineering data is input into a first preset prediction model to output a prediction result, and in the case where the accuracy value of the prediction result is greater than a preset threshold, the index classification corresponding to the prediction result is displayed. It is used to classify the engineering data by a preset rule, display the index classification corresponding to the identification result in the case where the identification result is successful, predict the situation that cannot be identified by the preset rule with a preset prediction model in the case where the identification result is unsuccessful, and assist with manual verification to improve the accuracy of the classification and identification of engineering data.

[0118] As Figure 3 shown, it is a schematic diagram of another embodiment of the method for analyzing engineering data indicators in an embodiment of the present application, which may include:

[0119] 301. Obtain engineering data.

[0120] 302. Classify and identify the engineering data using a preset rule to obtain an identification result.

[0121] 303. In the case where the identification result is successful, display the index classification corresponding to the identification result.

[0122] 304. In the case where the identification result is unsuccessful, input the engineering data into a first preset prediction model to output a prediction result, and in the case where the accuracy value of the prediction result is greater than a preset threshold, display the index classification corresponding to the prediction result.

[0123] It should be noted that steps 301-304 in the embodiment of the present application are similar to steps 101-104 in the Figure 1 shown embodiment, and will not be elaborated here.

[0124] 305. Obtain the classified engineering data.

[0125] After a period of time, when the electronic device accumulates a certain amount of classified engineering data, the model training process will be automatically triggered to obtain a more accurate preset prediction model.

[0126] In some possible implementation manners, in the case where the identification result is successful, the identification result is used to update the first preset prediction model. In the case where the identification result is unsuccessful, the modified identification result is used to update the first preset prediction model. The modified prediction result is used to update the first preset prediction model. The classification result is used to update the first preset prediction model.

[0127] Therefore, the classified engineering data may include: the recognition result in the case of successful recognition result; or, in the case of unsuccessful recognition result, the modified recognition result; or, the modified prediction result; or, the classification result.

[0128] 306. Extract features from the classified engineering data to obtain classified engineering feature data.

[0129] Exemplarily, extract features from the classified engineering data, where features are extracted from a sufficient amount of classified engineering data because the larger the amount of data for model training, the higher the reliability.

[0130] For example: extract features from the recognition result to obtain classified engineering feature data; or, extract features from the modified recognition result to obtain classified engineering feature data; or, extract features from the modified prediction result to obtain classified engineering feature data; or, extract features from the classification result to obtain classified engineering feature data.

[0131] 307. Train a model with the classified engineering feature data to obtain a second preset prediction model.

[0132] Exemplarily, after extracting features from the classified engineering data to obtain classified engineering feature data, the classified engineering feature data can be used for model training to obtain a second preset prediction model.

[0133] 308. Determine verification data based on the classified engineering feature data.

[0134] Exemplarily, the classified engineering feature data can be split into training data and verification data, and the proportion of training data is greater than that of verification data. Or, all the classified engineering feature data can be directly used as training data, and part of the classified engineering feature data can be used as verification data.

[0135] 309. Input the verification data into the first preset prediction model to obtain a first output classification result.

[0136] 310. Determine a first error between the first output classification result and the classification result of the verification data.

[0137] Exemplarily, input the verification data into the first preset prediction model to obtain a first output classification result, and then the first output classification result can be compared with the corresponding classification result of the verification data to determine the first error between the first output classification result and the classification result of the verification data.

[0138] 311. Input the verification data into the second preset prediction model to obtain the second output classification result.

[0139] 312. Determine the second error between the second output classification result and the classification result of the verification data.

[0140] Exemplarily, input the verification data into the second preset prediction model to obtain the second output classification result, and then the second output classification result can be compared with the corresponding classification result of the verification data to determine the second error between the second output classification result and the classification result of the verification data.

[0141] 313. When the first error is greater than the second error, update the first preset prediction model to the second preset prediction model.

[0142] Exemplarily, when the first error between the first output classification result and the classification result of the verification data is greater than the second error between the second output classification result and the classification result of the verification data, it indicates that the accuracy of the newly trained second preset prediction model is greater than that of the first preset prediction model, and update the first preset prediction model to the second preset prediction model.

[0143] 314. When the first error is less than or equal to the second error, do not execute the update process.

[0144] Exemplarily, when the first error between the first output classification result and the classification result of the verification data is less than or equal to the second error between the second output classification result and the classification result of the verification data, it indicates that the accuracy of the newly trained second preset prediction model is less than or equal to that of the first preset prediction model, and there is no need to update the first preset prediction model.

[0145] As Figure 4A shown, it is a schematic flowchart of training the second preset prediction model in an embodiment of the present application. The system of the electronic device generates classification data through feature extraction from the classified cost data. For example: split it into training data and verification data according to a ratio of 8:2; the training data is used for model training through a text classification model to generate a new model; the new model and the original model are evaluated and compared for the same verification data. When the new model is better than the original model, update and replace the original model; otherwise, retain the original model.

[0146] It should be noted that for the text classification model, that is, the text classification algorithm and the word segmentation algorithm, there are multiple algorithms that can be implemented in computer implementation, and the present application does not make specific limitations.

[0147] As Figure 4B shown, it is a schematic diagram of the classified engineering data in an embodiment of the present application. As Figure 4CAs shown, it is another schematic diagram of the classified engineering data in the embodiment of the present application. For example, Figure 4D As shown, it is another schematic diagram of the classified engineering data in the embodiment of the present application.

[0148] In the embodiment of the present application, engineering data is obtained; when the recognition result is successful, the index classification corresponding to the recognition result is displayed; when the recognition result is unsuccessful, the engineering data is input into the first preset prediction model, and a prediction result is output. When the accuracy value of the prediction result is greater than the preset threshold, the index classification corresponding to the prediction result is displayed; the classified engineering data is obtained; feature extraction is performed on the classified engineering data to obtain classified engineering feature data; feature extraction is performed on the classified engineering data to obtain classified engineering feature data; model training is performed on the classified engineering feature data to obtain a second preset prediction model; verification data is determined according to the classified engineering feature data; the verification data is input into the first preset prediction model to obtain a first output classification result; the first error between the first output classification result and the classification result of the verification data is determined; the verification data is input into the second preset prediction model to obtain a second output classification result; the second error between the second output classification result and the classification result of the verification data is determined; when the first error is greater than the second error, the first preset prediction model is updated to the second preset prediction model; when the first error is less than or equal to the second error, the update process is not executed.

[0149] In the present application, an automatic index classification recognition process is provided. Data generated by expert rules with higher accuracy is used to train a model using machine learning algorithms, and then the model is used to predict and handle situations that cannot be recognized by expert rules, supplemented by manual verification. The efficiency of algorithm recognition is provided. The expert recognition rule configuration is defined, and the rules can be updated quickly by modifying the configuration file. The feature extraction generation rules for cost data are defined, which can better generate training data for machine learning.

[0150] This solution combines the advantages of manual classification, expert rule classification, and machine learning classification:

[0151] Through the method of expert rules and manual adjustment, a generation mechanism for basic data is formed, providing basic data for machine learning. The expert rule configuration can be modified and upgraded, providing room for optimization and upgrading. By expanding the synonym library, problems such as synonyms, near-synonyms, and non-standard descriptions are solved, and model feature acquisition is optimized. The classification model can be iteratively updated regularly to gradually improve the recognition rate of the model.

[0152] This technical solution forms basic expert rules by collecting expert opinions, which are used for automatic classification of some indicators and assisted by manual correction. After a certain amount of classification data is formed, a classification model is formed through the text classification algorithm of machine learning to predict the classification that cannot be recognized by the rules, so as to improve the recognition rate of indicator classification and solve various disadvantages in the single mode:

[0153] 1. Based on expert opinions, benchmark rules are formed and assisted by manual work to solve the problem of the source of training data and ensure the quality of data.

[0154] 2. The file structure format of expert rules is defined and assisted by relevant parsing engines to achieve rule-based automatic classification. At the same time, it is convenient to adjust the leave rules configuration.

[0155] 3. The word segmentation algorithm and the thesaurus are introduced to expand the description of the generated feature text, improve the recognition of text classification features, and solve the problems of near synonyms and non-standard words.

[0156] 4. The classification is predicted through the calculation of the classification algorithm to handle the classification of items that cannot be recognized by expert rules and solve the deficiencies of expert rules.

[0157] 5. Optimize the classification data feature extraction algorithm and introduce more cost information, such as the attribution information of cost objects, to solve the distinction of the same feature word in different uses.

[0158] Such as Figure 5 As shown, it is a schematic diagram of an embodiment of the device for analyzing engineering data indicators in the embodiment of the present application, which may include:

[0159] An acquisition module 501, configured to acquire engineering data;

[0160] A processing module 502, configured to classify and identify the engineering data using preset rules to obtain an identification result;

[0161] A display module 503, configured to display the indicator classification corresponding to the identification result in the case where the identification result is successful; or,

[0162] The processing module 502 is further configured to input the engineering data into a first preset prediction model and output a prediction result in the case where the identification result is unsuccessful;

[0163] The display module 503 is further configured to display the indicator classification corresponding to the prediction result in the case where the accuracy value of the prediction result is greater than a preset threshold.

[0164] In some possible implementation manners, the processing module 502 is specifically configured to extract features from the project data to obtain project feature data; in response to the selection of a target index analysis rule template, use the preset rule corresponding to the target index analysis rule template to classify and identify the project feature data to obtain an identification result.

[0165] In some possible implementation manners, the processing module 502 is further configured to respond to a confirmation operation on the index classification corresponding to the identification result, and end the current classification process; or,

[0166] The processing module 502 is further configured to respond to a modification operation on the index classification corresponding to the identification result to obtain a modified identification result, and the display module 503 is further configured to display the modified identification result.

[0167] In some possible implementation manners, the processing module 502 is further configured to respond to a confirmation operation on the index classification corresponding to the prediction result, and end the current classification process; or,

[0168] The processing module 502 is further configured to respond to a modification operation on the index classification corresponding to the prediction result to obtain a modified prediction result, and the display module 503 is further configured to display the modified prediction result.

[0169] In some possible implementation manners, the display module 503 is further configured to display an unclassified indication message when the accuracy value of the prediction result is less than or equal to the preset threshold;

[0170] The processing module 502 is further configured to respond to a classification operation on the project data to obtain a classification result.

[0171] In some possible implementation manners, the processing module 502 is further configured to obtain classified project data; extract features from the classified project data to obtain classified project feature data; perform model training on the classified project feature data to obtain a second preset prediction model.

[0172] In some possible implementation manners, the processing module 502 is further configured to determine verification data according to the classified project feature data; input the verification data into the first preset prediction model to obtain a first output classification result; determine a first error between the first output classification result and the classification result of the verification data; input the verification data into the second preset prediction model to obtain a second output classification result; determine a second error between the second output classification result and the classification result of the verification data; in the case where the first error is greater than the second error, update the first preset prediction model to the second preset prediction model; in the case where the first error is less than or equal to the second error, do not execute the update process.

[0173] As Figure 6 shown, it is a schematic diagram of an embodiment of an electronic device in an embodiment of the present application, which may include a device for analyzing engineering data indicators as Figure 5 shown.

[0174] As Figure 7 shown, it is a schematic diagram of another embodiment of an electronic device in an embodiment of the present application, which may include:

[0175] A memory 701, a processor 702, and a display 703. The memory 701 stores a computer program that can run on the processor 702;

[0176] The processor 702 is configured to obtain engineering data; classify and identify the engineering data using a preset rule to obtain an identification result;

[0177] The display 703 is configured to display the index classification corresponding to the identification result when the identification result is successful; or,

[0178] The processor 702 is further configured to input the engineering data into a first preset prediction model and output a prediction result when the identification result is unsuccessful;

[0179] The display 703 is further configured to display the index classification corresponding to the prediction result when the accuracy value of the prediction result is greater than a preset threshold.

[0180] In some possible implementation manners, the processor 702 is specifically configured to extract features from the engineering data to obtain engineering feature data; in response to the selection of a target index analysis rule template, classify and identify the engineering feature data using the preset rule corresponding to the target index analysis rule template to obtain an identification result.

[0181] In some possible implementation manners, the processor 702 is further configured to end the current classification process in response to a confirmation operation on the index classification corresponding to the identification result; or,

[0182] The processor 702 is further configured to obtain a modified identification result in response to a modification operation on the index classification corresponding to the identification result, and the display 703 is further configured to display the modified identification result.

[0183] In some possible implementation manners, the processor 702 is further configured to end the current classification process in response to a confirmation operation on the index classification corresponding to the prediction result; or,

[0184] The processor 702 is further configured to obtain a modified prediction result in response to a modification operation on the metric classification corresponding to the prediction result, and the display 703 is further configured to display the modified prediction result.

[0185] In some possible implementation manners, the display 703 is further configured to display unclassified indication information when the accuracy value of the prediction result is less than or equal to the preset threshold;

[0186] The processor 702 is further configured to obtain a classification result in response to a classification operation on engineering data.

[0187] In some possible implementation manners, the processor 702 is further configured to obtain classified engineering data; extract features from the classified engineering data to obtain classified engineering feature data; perform model training on the classified engineering feature data to obtain a second preset prediction model.

[0188] In some possible implementation manners, the processor 702 is further configured to determine verification data according to the classified engineering feature data; input the verification data into the first preset prediction model to obtain a first output classification result; determine a first error between the first output classification result and the classification result of the verification data; input the verification data into the second preset prediction model to obtain a second output classification result; determine a second error between the second output classification result and the classification result of the verification data; in the case where the first error is greater than the second error, update the first preset prediction model to the second preset prediction model; in the case where the first error is less than or equal to the second error, do not execute the update process.

[0189] The embodiment of the present application further provides a computer-readable storage medium, where the computer-readable storage medium stores computer instructions, and when the computer instructions run on an electronic device, the electronic device is enabled to execute the method embodiment described above.

[0190] The above computer-readable storage medium may adopt any combination of one or more computer-readable media. The computer-readable media may be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples (non-exhaustive list) of the computer-readable storage medium include: an electrical connection having 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 a 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 this document, the computer-readable storage medium may be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0191] The computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries the computer-readable program code. Such a propagated data signal may take various forms, including - but not limited to - an electromagnetic signal, an optical signal, or any suitable combination of the above. The computer-readable signal medium may also be any computer-readable medium other than the computer-readable storage medium, which can send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0192] The program code contained on the computer-readable medium may be transmitted using any appropriate medium, including - but not limited to - wireless, wire, optical fiber, radio frequency (RF), etc., or any suitable combination of the above.

[0193] Computer program code for performing the operations of this specification may be written in one or more programming languages or combinations thereof, including object-oriented programming languages such as Java, Smalltalk, C++, and also including conventional procedural programming languages such as the "C" language or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, executed as a stand-alone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., connected through the Internet using an Internet service provider).

[0194] An embodiment of this application also provides a computer program product. When the computer program product runs on a computer, it causes the computer to execute some or all of the steps in the foregoing method embodiment.

[0195] An embodiment of this application provides a chip system. The chip system includes a processor and may also include a memory for implementing the functions of the electronic device in the foregoing method. The chip system may be composed of chips or may include chips and other discrete devices.

[0196] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the systems, devices, and units described above can refer to the corresponding processes in the foregoing method embodiments and will not be elaborated here.

[0197] In several embodiments provided in this application, it should be understood that the disclosed systems, devices, and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units is only a logical function division, and there may be other division methods in actual implementation. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections to each other may be through some interfaces, and the indirect couplings or communication connections of the devices or units may be in electrical, mechanical, or other forms.

[0198] The unit described as a separation component may or may not be physically separated. The component shown as a unit may or may not be a physical unit, that is, it may be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0199] In addition, each functional unit in various embodiments of the present application can be integrated into a processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of a software functional unit.

[0200] If the above-mentioned integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art, or all or part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present application. The aforementioned storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks, or optical discs that can store program codes.

[0201] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of various embodiments of the present application.

Claims

1. A method for analyzing engineering data indicators, characterized in that, Including: Obtain engineering data; Classify and identify the engineering data using preset rules to obtain an identification result; When the identification result is successful, display the index classification corresponding to the identification result; or, When the identification result is unsuccessful, input the engineering data into a first preset prediction model to output a prediction result, and when the accuracy value of the prediction result is greater than a preset threshold, display the index classification corresponding to the prediction result.

2. The method according to claim 1, wherein The classifying and identifying the engineering data using preset rules to obtain an identification result includes: Extract features from the engineering data to obtain engineering feature data; In response to the selection of a target index analysis rule template, classify and identify the engineering feature data using the preset rules corresponding to the target index analysis rule template to obtain an identification result.

3. The method according to claim 1 or 2, characterized in that, The method further includes: In response to a confirmation operation for the index classification corresponding to the identification result, the current classification process ends; or, In response to a modification operation for the index classification corresponding to the identification result, obtain a modified identification result and display the modified identification result.

4. The method according to claim 1 or 2, characterized in that, The method further includes: In response to a confirmation operation for the index classification corresponding to the prediction result, the current classification process ends; or, In response to a modification operation for the index classification corresponding to the prediction result, obtain a modified prediction result and display the modified prediction result.

5. The method according to claim 1 or 2, characterized in that, The method further includes: When the accuracy value of the prediction result is less than or equal to the preset threshold, display an unclassified indication message; In response to a classification operation on the engineering data, obtain a classification result.

6. The method according to claim 1 or 2, characterized in that, The method further includes: Obtain classified engineering data; Extract features from the classified engineering data to obtain classified engineering feature data; Train a model on the classified engineering feature data to obtain a second preset prediction model.

7. The method according to claim 6, wherein The method further includes: Determine verification data based on the classified engineering feature data; Input the verification data into the first preset prediction model to obtain a first output classification result; Determine a first error between the first output classification result and the classification result of the verification data; Input the verification data into the second preset prediction model to obtain a second output classification result; Determine a second error between the second output classification result and the classification result of the verification data; When the first error is greater than the second error, update the first preset prediction model to the second preset prediction model; When the first error is less than or equal to the second error, do not execute the update process.

8. An apparatus for analyzing engineering data indicators, characterized in that, Including: An acquisition module for acquiring engineering data; A processing module for classifying and identifying the engineering data using preset rules to obtain an identification result; When the identification result is successful, display the index classification corresponding to the identification result; or, when the identification result is unsuccessful, input the engineering data into a first preset prediction model to output a prediction result, and when the accuracy value of the prediction result is greater than a preset threshold, display the index classification corresponding to the prediction result.

9. An electronic device, characterized in that, Including: A memory, a processor, and a display, wherein the memory stores a computer program that can run on the processor, and the electronic device implements the method according to any one of claims 1-7 when executing the program.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, the method according to any one of claims 1-7 is implemented.

Images