Project evaluation method and device based on large model and electronic equipment

Abstract

The invention provides a project evaluation method and device based on a large model and electronic equipment, and relates to the technical field of computers, in particular to the application fields of software and hardware project development, software and hardware project evaluation, machine learning, large models and the like. The specific implementation scheme is as follows: acquiring an evaluation intention for a target project; wherein the evaluation intention is used for requesting to evaluate the target project under a specific evaluation index; obtaining an initial evaluation result for the target project under the specific evaluation index by using the large model based on the evaluation intention; and obtaining a target evaluation result for the target project based on the initial evaluation result.

Description

Technical Field

[0001] The present disclosure relates to the field of computer technology, and in particular to application fields such as software and hardware project development, software and hardware project evaluation, machine learning, and large models, and specifically to a project evaluation method, device, and electronic equipment based on large models. Background Art

[0002] With the advancement of digitalization in the project delivery process (e.g., software and hardware projects), project implementers hope to easily grasp the progress of each stage in the software project delivery process and evaluate relevant indicators in subsequent stages (e.g., project health, project risks, estimated completion time, project benefits, etc.) to ensure the efficiency and controllability of the overall delivery. Summary of the invention

[0003] The present invention provides a project evaluation method, device and electronic device based on a large model.

[0004] According to a first aspect of the present disclosure, a project evaluation method based on a large model is provided, comprising:

[0005] Obtain the evaluation intention for the target project; wherein the evaluation intention is used to request the target project to be evaluated under a specific evaluation indicator;

[0006] Using the big model, based on the evaluation intention, we can obtain the initial evaluation results for the target project under specific evaluation indicators;

[0007] Based on the initial assessment results, target assessment results for the target projects are obtained.

[0008] According to a second aspect of the present disclosure, there is provided a project evaluation device based on a large model, comprising:

[0009] An intention acquisition unit, used to acquire an evaluation intention for a target project; wherein the evaluation intention is used to request an evaluation of the target project under a specific evaluation indicator;

[0010] A first result acquisition unit is used to obtain an initial evaluation result for a target project under a specific evaluation indicator based on the evaluation intention by using the large model;

[0011] The second result obtaining unit is used to obtain a target evaluation result for the target project based on the initial evaluation result.

[0012] According to a third aspect of the present disclosure, there is provided an electronic device, including:

[0013] at least one processor;

[0014] a memory communicatively coupled to the at least one processor;

[0015] The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the method provided in the first aspect of the present disclosure.

[0016] According to a fourth aspect of the present disclosure, a non-transitory computer-readable storage medium storing computer instructions is provided, wherein the computer instructions are used to enable the computer to execute the method provided according to the first aspect of the present disclosure.

[0017] According to a fifth aspect of the present disclosure, a computer program product is provided, comprising a computer program, which implements the method provided according to the first aspect of the present disclosure when executed by a processor.

[0018] The use of the present disclosure can improve the efficiency and reliability of obtaining target evaluation results for target projects.

[0019] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present disclosure, nor is it intended to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The accompanying drawings are used to better understand the present solution and do not constitute a limitation of the present disclosure.

[0021] Figure 1 A flowchart of a project evaluation method based on a large model provided in an embodiment of the present disclosure;

[0022] Figure 2 A schematic diagram of a workflow of a data search model provided in an embodiment of the present disclosure;

[0023] Figure 3 A schematic diagram of a workflow of a data processing model provided in an embodiment of the present disclosure;

[0024] Figure 4 A schematic diagram of a workflow of a prediction and reasoning model provided in an embodiment of the present disclosure;

[0025] Figure 5 A complete flow chart of a project evaluation method based on a large model provided in an embodiment of the present disclosure;

[0026] Figure 6 A schematic diagram of an application scenario of a project evaluation method based on a large model provided in an embodiment of the present disclosure;

[0027] Figure 7 A schematic structural block diagram of a project evaluation device based on a large model provided in an embodiment of the present disclosure;

[0028] Figure 8 A schematic structural block diagram of an electronic device provided in an embodiment of the present disclosure. DETAILED DESCRIPTION

[0029] The following is a description of exemplary embodiments of the present disclosure in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope of the present disclosure. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0030] As described in the background technology, with the advancement of digitalization of the project delivery process, the project implementers hope to easily grasp the progress of each stage in the software project delivery process and evaluate the relevant indicators of the subsequent stages, aiming to ensure the efficiency and controllability of the overall delivery. At present, the project evaluation method adopted is usually: manually collect project data related to the target project, and obtain the evaluation results for the target project based on these project data through manual evaluation. However, manual evaluation not only consumes a lot of manpower, but also reduces the efficiency and reliability of obtaining the evaluation results due to personnel experience, subjective bias, work involvement, etc., thereby having a negative impact on the efficiency and controllability of the overall delivery.

[0031] In view of the above problems, the embodiments of the present disclosure provide a project evaluation method based on a large model, which can be applied to electronic devices. The electronic device can be a server or a terminal device. Here, the terminal device can be a workbench, a mainframe computer, a conventional computer (for example, a desktop computer, a laptop computer, a car computer, etc.), a personal digital processing unit or other similar computing devices. Figure 1 The flowchart shown in the figure illustrates a project evaluation method based on a large model provided by an embodiment of the present disclosure. It should be noted that although a logical order is shown in the flowchart, in some cases, the steps shown or described in the flowchart may also be performed in other orders.

[0032] Step S101, obtaining the evaluation intention for the target project.

[0033] Among them, the target project can be a software or hardware project that is about to be developed or is being developed, such as a functional software development project, an application development project, a website development project, a system integration project, an infrastructure construction project, etc.; the evaluation intention can be issued by a user (for example, a project member) to request an evaluation of the target project under a specific evaluation indicator. Here, the specific evaluation indicator can be project health, project risk, estimated completion time, project revenue, etc., which is not limited in the embodiments of the present disclosure. Exemplarily, the evaluation intention is: please evaluate the XX software project to obtain the evaluation result for the XX software project under the evaluation indicator of project revenue. Among them, the project revenue is a specific evaluation indicator.

[0034] Step S102, using the big model and based on the evaluation intention, obtain the initial evaluation result for the target project under specific evaluation indicators.

[0035] The large model may be a large language model. Here, the large language model may be a pre-trained neural network model that has general and reliable language knowledge, world knowledge, and expertise in various fields (e.g., expertise in the field of software and hardware project evaluation). In one example, the large model may be an autoregressive generative model of a Transformer architecture.

[0036] Since the big model has universal and reliable language knowledge, world knowledge, and professional knowledge in various fields, after obtaining the evaluation intention for requesting the evaluation of the target project under specific evaluation indicators, the evaluation intention can be directly input into the big model, so as to use the big model to obtain the initial evaluation results for the target project under specific evaluation indicators based on the evaluation intention. For example, if the specific evaluation indicator is project health, the initial evaluation results may include at least one of the project progress, budget control performance, resource quality, risk management capabilities, and health score of the target project, as well as the comprehensive health of the target project. For another example, if the specific evaluation indicator is project revenue, the initial evaluation results may include at least one of the actual revenue, investment cost, and return on investment of the target project.

[0037] Step S103: obtaining a target evaluation result for the target project based on the initial evaluation result.

[0038] In the embodiments of the present disclosure, after obtaining the initial evaluation results, the initial evaluation results may be processed to obtain the target evaluation results for the target project. For example, the initial evaluation results may be processed according to the preset format requirements and / or the preset expression requirements to obtain the target evaluation results for the target project. The preset format requirements and the preset expression requirements may be set according to actual business needs or application scenarios, and the embodiments of the present disclosure do not limit this.

[0039] The project evaluation method based on the big model provided by the embodiment of the present disclosure can obtain the evaluation intention for requesting the evaluation of the target project under specific evaluation indicators, and use the big model, based on the evaluation intention, to obtain the initial evaluation result for the target project under the specific evaluation indicator, and then based on the initial evaluation result, obtain the target evaluation result for the target project. On the one hand, the project evaluation method based on the big model provided by the embodiment of the present disclosure does not have a substantial human participation part. Therefore, compared with the prior art, it can not only save human resources, but also put aside the influence of subjective factors (for example, personnel experience, subjective bias, work involvement, etc.) on the target evaluation result, and improve the efficiency and reliability of obtaining the target evaluation result; on the other hand, in the embodiment of the present disclosure, the main evaluation steps are implemented based on the big model, and the big model has universal and reliable language knowledge, world knowledge, and professional knowledge in various fields. Therefore, the reliability of the target evaluation result can be further improved. Finally, the project evaluation method based on the big model provided by the embodiment of the present disclosure can have a positive impact on the efficiency and controllability of the overall delivery.

[0040] In some optional implementations, step S101, i.e., “obtaining evaluation intention for the target project” may include:

[0041] Get the initial intent for the target project;

[0042] Generate inquiry information when it is determined that there are clarity issues with the initial intent;

[0043] Report the inquiry information;

[0044] In response to an intent optimization request received after the query information is broadcast, the initial intent is optimized to obtain an evaluated intent.

[0045] It can be understood that in the embodiments of the present disclosure, when it is determined that the initial intent does not have clarity issues, the initial intent can be used as the evaluation intent.

[0046] In addition, it should be noted that in the embodiments of the present disclosure, when it is determined that the initial intent has a clarity problem, query information is generated and the query information is broadcasted, and then the initial intent is optimized in response to the intent optimization request. After the optimized initial intent is obtained, if the optimized initial intent does not have a clarity problem, the optimized initial intent can be used as the evaluation intent; if the optimized initial intent has a clarity problem, the optimized initial intent is used as the new initial intent, and the steps of "when it is determined that the initial intent does not have a clarity problem, the initial intent is used as the evaluation intent; when it is determined that the initial intent has a clarity problem, a query information is generated; the query information is broadcasted; and the initial intent is optimized in response to the intent optimization request" are executed again until the optimized initial intent does not have a clarity problem, and the optimized initial intent is used as the evaluation intent.

[0047] It should also be noted that, in the embodiment of the present disclosure, step S101 may be performed using an intention recognition model, wherein the intention recognition model may also be a large language model.

[0048] Specifically, in the disclosed embodiment, the intent recognition model can be used to determine multiple preset word slots, so that when there are multiple intent keywords corresponding to the multiple preset word slots in the initial intent, a clarity evaluation result is obtained to characterize that the initial intent does not have a clarity problem, otherwise a clarity evaluation result is obtained to characterize that the initial intent has a clarity problem. Thereafter, when the clarity evaluation result characterizes that the initial intent does not have a clarity problem, the initial intent can be used as the evaluation intent; when the clarity evaluation result characterizes that the initial intent has a clarity problem, the word slots to be supplemented are determined, and based on the word slots to be supplemented, query information is generated, and the query information is broadcasted to prompt the user that an intent optimization request needs to be issued, and after receiving the intent optimization request, the initial intent is optimized in response to the intent optimization request. Among them, the word slot to be supplemented can be a preset word slot in which the intent keyword captured from the initial intent is empty; the broadcasting of the query information can be voice broadcasting and / or text display.

[0049] Exemplarily, the multiple preset word slots include: [project name] and [specific evaluation indicator], and the initial intention is: Please evaluate the XX software project to obtain the evaluation result of the XX software project under the evaluation indicator of project revenue. Among them, the intention keyword captured by the preset word slot [project name] is XX software project, and the intention keyword captured by the preset word slot [specific evaluation indicator] is project revenue. Therefore, a clarity evaluation result can be obtained to characterize that the initial intention does not have clarity problems, and the initial intention is used as the evaluation intention.

[0050] Exemplarily, multiple preset word slots include: [project name] and [specific evaluation indicators], and the initial intention is: please evaluate the XX software project to obtain the evaluation result of the XX software project. Among them, the intent keyword captured by the preset word slot [project name] is the XX software project, and the intent keyword captured by the preset word slot [specific evaluation indicators] is empty. Therefore, a clarity evaluation result for characterizing that the initial intention has a clarity problem can be obtained, and [specific evaluation indicators] is determined as a word slot to be supplemented. Thereafter, based on the word slot to be supplemented, an inquiry information can be generated, and the inquiry information can be broadcasted to prompt the user that an intent optimization request needs to be issued, and after receiving the intent optimization request, the initial intent is optimized in response to the intent optimization request. Assume that there is an inquiry information generated based on the word slot to be supplemented: Under what evaluation indicators do you need to evaluate the XX software project? Here, the candidate evaluation indicators include project health, project risk, estimated completion time, and project benefits. Assume again that the received intent optimization request is: please evaluate the XX software project under the evaluation indicator of project benefits. Then, after receiving the intent optimization request, the initial intent can be optimized in response to the intent optimization request to obtain an optimized initial intent: Please evaluate the XX software project to obtain the evaluation result of the XX software project under the evaluation indicator of project revenue. Among them, the optimized initial intent does not have clarity problems, so the optimized initial intent can be used as the evaluation intent.

[0051] Through the above method, in the disclosed embodiment, the initial intention for the target project can be obtained, and when it is determined that the initial intention has a clarity problem, query information is generated, the query information is broadcast, and then in response to the intent optimization request received after the query information is broadcast, the initial intention is optimized to obtain the evaluation intention. In this way, the clarity of the evaluation intention can be ensured, that is, the large model can clearly understand the evaluation intention, thereby further improving the reliability of the target evaluation result.

[0052] In some optional implementations, the step S102 of “obtaining an initial evaluation result for the target project under a specific evaluation indicator based on the evaluation intention” may include:

[0053] Scheduling the data search model to obtain preliminary project data related to the target project and targeting specific evaluation indicators based on the evaluation intent using the data search model;

[0054] Schedule the data processing model to optimize the preliminary project data based on the evaluation intention and obtain the target project data by using the data processing model;

[0055] The predictive reasoning model is scheduled to utilize the predictive reasoning model to process the target project data based on the evaluation intention and obtain the initial evaluation results for the target project under specific evaluation indicators.

[0056] Among them, the data search model, data processing model and predictive reasoning model can all be fine-tuned large language models; the preliminary project data can be the project data generated and recorded during the current period that is related to the target project and is preliminarily selected for specific evaluation indicators; correspondingly, the target project data can be the project data generated and recorded during the current period that is related to the target project and is finally selected for specific evaluation indicators.

[0057] In one example, “obtaining preliminary selected project data related to the target project and targeting specific evaluation indicators based on the evaluation intention” may include:

[0058] Based on the evaluation intention, determining at least one target data source from a plurality of candidate data sources;

[0059] Acquire original project data related to the target project and for a specific evaluation indicator from each of the at least one target data source to obtain at least one original project data corresponding one-to-one to the at least one target data source;

[0060] Based on at least one original project data, preliminary selected project data are obtained.

[0061] The multiple candidate data sources may be databases, document repositories, and designated servers related to a target organization (eg, a company responsible for developing a target project).

[0062] Please combine Figure 2 In a specific example, after obtaining the evaluation intention for requesting the evaluation of the target project under a specific evaluation indicator, the data search model can be used to determine at least one candidate data source corresponding to the specific evaluation indicator from multiple candidate data sources (including databases, document libraries, and designated servers) as the target data source according to its built-in data source correspondence relationship, so as to obtain at least one target data source. For example, when the specific evaluation indicator is the project revenue, the candidate data sources corresponding to the specific evaluation indicator may include databases, document libraries, and designated servers.

[0063] After at least one target data source is determined from multiple candidate data sources based on the evaluation intention, the data search model can be used to obtain original project data related to the target project and for specific evaluation indicators from each of the at least one target data source according to its built-in data search strategy, so as to obtain at least one original project data corresponding to the at least one target data source. Exemplarily, the specific evaluation indicator is the project revenue situation, then the original project data related to the target project and for the specific evaluation indicator obtained from the database according to the data search strategy may include the project contract, asset consumption situation, etc. of the target project; the original project data related to the target project and for the specific evaluation indicator obtained from the document library may include the professional terminology of the target project; the original project data related to the target project and for the specific evaluation indicator obtained from the designated server may include the salary of the project members, reimbursement expenses, etc.

[0064] After obtaining at least one original project data corresponding one-to-one to at least one target data source, the at least one original project data can be used together as preliminary selected project data, that is, the at least one original project data can be merged to obtain preliminary selected project data.

[0065] In the above example, the data search model can be used to determine at least one target data source from multiple candidate data sources based on the evaluation intention, and obtain original project data related to the target project and for specific evaluation indicators from each of the at least one target data source to obtain at least one original project data corresponding to the at least one target data source, and then obtain preliminary project data based on the at least one original project data. In this process, there is also no substantial manual participation, so the efficiency and reliability of obtaining the target evaluation results can be further improved. Moreover, the preliminary project data is obtained by using the data search model, and does not involve complex calculation logic, so the efficiency and reliability of obtaining the target evaluation results can also be further improved.

[0066] In addition, as mentioned above, in the embodiment of the present disclosure, the data search model can be a fine-tuned large language model (hereinafter referred to as the first large language model for the convenience of distinction). The following will describe the fine-tuning process of the first large language model:

[0067] (1) Obtain a first evaluation intention sample for a first target item sample.

[0068] The first target project sample may be a completed software or hardware project, such as a functional software development project, an application development project, a website development project, a system integration project, an infrastructure construction project, etc. The first evaluation intention sample may be used to request an evaluation of the first target project sample under a first specific evaluation indicator. Here, the first specific evaluation indicator may be project health, project risk, estimated completion time, project revenue, etc.

[0069] (2) By manual search, based on the first evaluation intention sample, a preliminary project data sample related to the first target project sample and targeting the first specific evaluation indicator is obtained as a preliminary project data label.

[0070] In one example, at least one first target data source can be determined from multiple candidate data sources based on a first evaluation intention sample through manual search, and a preliminary project data sample related to the first target project sample and targeting a first specific evaluation indicator can be obtained from the at least one first target data source as a preliminary project data label.

[0071] (3) Using the first language model and based on the first evaluation intention sample, a preliminary selection project data sample related to the first target project sample and targeting the first specific evaluation indicator is obtained as predictive preliminary selection project data.

[0072] It can be understood that in the embodiments of the present disclosure, the first large language model can be used to determine at least one second target data source from multiple candidate data sources based on the first evaluation intention sample, and obtain original project data samples related to the first target project sample and for the first specific evaluation indicator from each second target data source of the at least one second target data source to obtain at least one original project data sample corresponding one-to-one to the at least one second target data source, and then obtain preliminary project data based on the at least one original project data sample as predictive preliminary project data.

[0073] In one example, after obtaining a first evaluation intention sample for requesting an evaluation of a first target project sample under a first specific evaluation indicator, the first large language model can be used to determine, according to its built-in data source correspondence, at least one first candidate data source corresponding to the first specific evaluation indicator from multiple candidate data sources (including databases, document libraries, and designated servers) as a second target data source to obtain at least one second target data source, and then the first large language model is used to obtain, according to its built-in data search strategy, original project data samples related to the first target project sample and for the first specific evaluation indicator from each second target data source in the at least one second target data source to obtain at least one original project data sample corresponding one-to-one to the at least one second target data source, and based on the at least one original project data sample, a preliminary project data sample is obtained as predictive preliminary project data.

[0074] (4) Calculate the first data loss of the predictive preliminary item data and the preliminary item data labels.

[0075] When the first data loss meets the first loss requirement, the first large language model is used as the data search model; when the first data loss does not meet the first loss requirement, the first large language model is fine-tuned to obtain a fine-tuned first large language model, and the fine-tuned first large language model is used as a new first large language model to re-enter the fine-tuning process of the first large language model. The first loss requirement can be set according to actual business needs or application scenarios, and the embodiments of the present disclosure are not limited to this.

[0076] It can be understood that, in the embodiment of the present disclosure, fine-tuning the first language model can affect at least one of the data source correspondence and the data search strategy.

[0077] In one example, “optimizing the preliminary selected project data based on the evaluation intention to obtain the target project data” may include:

[0078] Based on the evaluation intention, the preliminary project data is optimized from multiple optimization dimensions to obtain the target project data.

[0079] Please combine Figure 3 In a specific example, the data processing model can be used to optimize the preliminary project data from multiple optimization dimensions based on the evaluation intention according to its built-in data optimization strategy to obtain the target project data. The multiple optimization dimensions can include at least one of data deduplication, redundant data removal, and merging of data with the same attributes.

[0080] Here, data deduplication can be performed on multiple project data with the same meaning. For example, in the preliminary project data, there is the first project data "Project member A's salary in May 2024 is 30,000 yuan", and the second project data "Project member A's salary in May 2024 is 30,000 yuan". Since these two project data have the same meaning, they can be deduplicated. Specifically, only the first project data or only the second project data can be retained. Redundant data removal can be performed on the redundant data in the preliminary project data, and the redundant data can be determined by the data processing model and is consistent with the original data. Project data that has no substantial relevance to the target project, or that has substantial relevance to the target project but does not target specific evaluation indicators; the merging of data with the same attributes can be the merging of data with the same attributes. For example, in the preliminary project data, there is a third project data "Project member A's salary in June 2024 is RMB 30,000", and a fourth project data "Project member A's project bonus in June 2024 is RMB 10,000". Since these two project data have the same attributes, they can be merged to obtain new project data "Project member A's salary and project bonus in June 2024 are RMB 40,000 in total".

[0081] In the above example, the data processing model can be used to optimize the preliminary project data from multiple optimization dimensions based on the evaluation intention to obtain the target project data. In this process, there is no substantial human participation, so the efficiency and reliability of obtaining the target evaluation results can be further improved. Moreover, the target project data is obtained by using the data processing model, and does not involve complex calculation logic. Therefore, the efficiency and reliability of obtaining the target evaluation results can also be further improved.

[0082] In addition, as mentioned above, in the embodiment of the present disclosure, the data processing model can be a fine-tuned large language model (hereinafter referred to as the second largest language model for the convenience of distinction). The following will describe the fine-tuning process of the second largest language model:

[0083] (1) Obtain a second evaluation intention sample for a second target project sample, and a preliminary selected project data sample related to the second target project sample and for a second specific evaluation indicator.

[0084] The second target project sample may be a completed software or hardware project, such as a functional software development project, an application development project, a website development project, a system integration project, an infrastructure construction project, etc. The second evaluation intention sample may be used to request an evaluation of the second target project sample under a second specific evaluation indicator. Here, the second specific evaluation indicator may be project health, project risk, estimated completion time, project revenue, etc.

[0085] (2) Through manual processing, based on the second evaluation intention sample, the preliminary project data sample is optimized from multiple optimization dimensions to obtain the target project data sample as the target project data label.

[0086] (3) Using the second largest language model, based on the second evaluation intention sample, the preliminary project data samples are optimized from multiple optimization dimensions to obtain the target project data samples as predictive target project data.

[0087] It can be understood that in the embodiment of the present disclosure, the second largest language model can be used to optimize the preliminary project data samples from multiple optimization dimensions according to its built-in data optimization strategy based on the second evaluation intention sample to obtain the target project data samples as predictive target project data.

[0088] (4) Calculate the second data loss of the predictive target item data and the target item data label.

[0089] When the second data loss meets the second loss requirement, the second largest language model is used as the data processing model; when the second data loss does not meet the second loss requirement, the second largest language model is fine-tuned to obtain a fine-tuned second largest language model, and the fine-tuned second largest language model is used as a new second largest language model to re-enter the fine-tuning process of the second largest language model. The second loss requirement can be set according to actual business needs or application scenarios, and the embodiments of the present disclosure are not limited to this.

[0090] It is understandable that, in the embodiments of the present disclosure, fine-tuning the second largest language model may affect the data optimization strategy.

[0091] In one example, “processing the target project data based on the evaluation intention to obtain an initial evaluation result for the target project under a specific evaluation indicator” may include:

[0092] Based on the evaluation intention, the target evaluation strategy is obtained;

[0093] Obtain historical project data related to the target project and targeting specific evaluation indicators;

[0094] According to the target evaluation strategy, the target project data and historical project data are processed to obtain the initial evaluation results for the target project under specific evaluation indicators.

[0095] The historical project data may be the project data generated and recorded in the historical period, which is related to the target project and is finally selected for a specific evaluation indicator. Here, the historical period is another period that is continuous with the current period and is located before the current period.

[0096] Please combine Figure 4 In a specific example, after obtaining the evaluation intention for requesting the evaluation of the target project under a specific evaluation indicator, the predictive reasoning model can be used to select a target evaluation strategy corresponding to the specific evaluation indicator from multiple candidate evaluation strategies according to its built-in evaluation strategy correspondence, and obtain historical project data related to the target project and for the specific evaluation indicator, and then process the target project data and the historical project data according to the target evaluation strategy to obtain an initial evaluation result for the target project under the specific evaluation indicator.

[0097] Specifically, when obtaining historical project data related to the target project and targeting specific evaluation indicators, the data search model can be scheduled again to use the data search model to obtain preliminary historical project data related to the target project and targeting specific evaluation indicators based on the evaluation intent, and the data processing model can be scheduled again to use the data processing model to optimize the preliminary historical project data based on the evaluation intent to obtain the historical project data that is finally selected. This process can be specifically referred to the aforementioned description of "scheduling the data search model to use the data search model to obtain preliminary project data related to the target project and targeting specific evaluation indicators based on the evaluation intent; scheduling the data processing model to use the data processing model to optimize the preliminary project data based on the evaluation intent to obtain the target project data", which will not be repeated here.

[0098] In another specific example, “processing the target project data and historical project data according to the target evaluation strategy to obtain an initial evaluation result for the target project under a specific evaluation indicator” may include:

[0099] Based on the target project data and historical project data, predict subsequent project data related to the target project and targeting specific evaluation indicators;

[0100] According to the target evaluation strategy, based on the target project data, historical project data and subsequent project data, the initial evaluation results for the target project under specific evaluation indicators are obtained.

[0101] The subsequent project data may be project data related to the target project and for specific evaluation indicators that may be generated and recorded in a future period. Here, the future period is another period that is continuous with the current period and is located after the current period.

[0102] In addition, it should be noted that in the disclosed embodiment, after obtaining the target project data and the historical project data, the predictive reasoning model can be used to predict the subsequent project data related to the target project and for specific evaluation indicators based on the target project data and the historical project data according to its built-in data prediction algorithm. Among them, the data prediction algorithm can be a time series analysis method, a regression analysis method, etc. After obtaining the target project data, the historical project data, and the subsequent project data, the predictive reasoning model can be used to obtain the initial evaluation results for the target project under specific evaluation indicators based on the target project data, the historical project data, and the subsequent project data according to the target evaluation strategy.

[0103] In the above example, the predictive reasoning model can be used to obtain the target evaluation strategy based on the evaluation intention, and obtain the historical project data related to the target project and for the specific evaluation index, and then process the target project data and the historical project data according to the target evaluation strategy to obtain the initial evaluation result for the target project under the specific evaluation index. Specifically, the subsequent project data related to the target project and for the specific evaluation index can be predicted based on the target project data and the historical project data, and the initial evaluation result for the target project under the specific evaluation index can be obtained based on the target project data, the historical project data and the subsequent project data according to the target evaluation strategy. In this process, there is also no substantial human participation part, so the efficiency and reliability of obtaining the target evaluation result can be further improved. Moreover, the initial evaluation result is obtained by using the predictive reasoning model, and does not involve complex calculation logic, so the efficiency and reliability of obtaining the target evaluation result can also be further improved.

[0104] In addition, as mentioned above, in the embodiment of the present disclosure, the predictive reasoning model can be a fine-tuned large language model (hereinafter referred to as the third language model for the convenience of distinction). The fine-tuning process of the third language model will be described below:

[0105] (1) Obtain a third evaluation intention sample for a third target project sample, and a target project data sample related to the third target project sample and for a third specific evaluation indicator.

[0106] The third target project sample can be a completed software and hardware project, such as a functional software development project, an application development project, a website development project, a system integration project, an infrastructure construction project, etc. The third evaluation intention sample can be used to request an evaluation of the third target project sample under a third specific evaluation indicator. Here, the third specific evaluation indicator can be project health, project risk, estimated completion time, project revenue, etc.

[0107] (2) Through manual evaluation, based on the third evaluation intention sample, the target project data sample is processed to obtain a true evaluation result for the third target project sample under the third specific evaluation indicator as an evaluation result label.

[0108] In one example, the data search model can be scheduled to utilize the data search model to obtain preliminary historical project data samples related to the third target project sample and for a third specific evaluation indicator based on the third evaluation intent sample, and the data processing model can be scheduled to utilize the data processing model to optimize the preliminary historical project data samples based on the third evaluation intent sample to obtain the historical project data samples that are finally selected. The specific process can refer to the aforementioned description of "scheduling the data search model to utilize the data search model to obtain preliminary project data related to the target project and for a specific evaluation indicator based on the evaluation intent; scheduling the data processing model to utilize the data processing model to optimize the preliminary project data based on the evaluation intent to obtain the target project data", which will not be repeated here.

[0109] In addition, since the third target project sample is a software and hardware project that has been developed, the data search model can also be scheduled to utilize the data search model to obtain preliminary selected subsequent project data samples related to the third target project sample and for the third specific evaluation indicator based on the third evaluation intention sample, and the data processing model can be scheduled to utilize the data processing model to optimize the preliminary selected subsequent project data samples based on the third evaluation intention sample to obtain the subsequent project data samples that are finally selected. This process can be specifically referred to the aforementioned description of "scheduling the data search model to utilize the data search model to obtain preliminary selected project data related to the target project and for specific evaluation indicators based on the evaluation intention; scheduling the data processing model to utilize the data processing model to optimize the preliminary selected project data based on the evaluation intention to obtain the target project data", which will not be repeated here.

[0110] Thereafter, a true evaluation result for the third target project sample under the third specific evaluation indicator can be obtained as an evaluation result label through manual evaluation based on the target project data sample, the historical project data sample and the subsequent project data sample.

[0111] (3) Using the third language model, based on the third evaluation intention sample, the target project data sample is processed to obtain a predictive initial evaluation result for the third target project sample under the third specific evaluation indicator.

[0112] It can be understood that in the embodiments of the present disclosure, a target evaluation strategy sample can be obtained based on the third evaluation intention sample, and a historical project data sample related to the third target project sample and for the third specific evaluation indicator can be obtained. Then, according to the target evaluation strategy sample, the target project data sample and the historical project data sample are processed to obtain an initial evaluation result for the third target project sample under the third specific evaluation indicator as a predictive evaluation result.

[0113] In a specific example, after obtaining a third evaluation intention sample for requesting an evaluation of a third target project sample under a third specific evaluation indicator, the third largest language model can be used to select a target evaluation strategy sample corresponding to the third specific evaluation indicator from multiple candidate evaluation strategies according to its built-in evaluation strategy correspondence, and obtain a historical project data sample related to the third target project sample and for the third specific evaluation indicator, and then process the target project data sample and the historical project data sample according to the target evaluation strategy sample to obtain an initial evaluation result for the third target project sample under the third specific evaluation indicator as a predictive evaluation result. Specifically, after obtaining the target project data sample and the historical project data sample, the third largest language model can be used to predict subsequent project data samples related to the third target project sample and for the third specific evaluation indicator based on the target project data sample and the historical project data sample according to its built-in data prediction algorithm. Among them, the data prediction algorithm can be a time series analysis method, a regression analysis method, etc. After obtaining the target project data samples, historical project data samples and subsequent project data samples, the third language model can be used to obtain an initial evaluation result for the third target project sample under a third specific evaluation indicator based on the target project data samples, historical project data samples and subsequent project data samples according to the target evaluation strategy sample as a predictive evaluation result.

[0114] (4) Calculate the third data loss of the predictive evaluation results and the evaluation result labels.

[0115] When the third data loss meets the third loss requirement, the third language model is used as the inference prediction model; when the third data loss does not meet the third loss requirement, the third language model is fine-tuned to obtain a fine-tuned third language model, and the fine-tuned third language model is used as a new third language model to re-enter the fine-tuning process of the third language model. The third loss requirement can be set according to actual business needs or application scenarios, and the embodiments of the present disclosure are not limited to this.

[0116] It can be understood that, in the embodiment of the present disclosure, fine-tuning the third language model can affect at least one of the evaluation strategy correspondence, the candidate evaluation strategies, and the data prediction algorithm.

[0117] Through the above methods, in the disclosed embodiment, the data search model can be scheduled to use the data search model to obtain preliminary project data related to the target project and for specific evaluation indicators based on the evaluation intention, and the data processing model can be scheduled to use the data processing model to optimize the preliminary project data based on the evaluation intention to obtain the target project data, and then the prediction and reasoning model can be scheduled to use the prediction and reasoning model to process the target project data based on the evaluation intention to obtain the initial evaluation result for the target project under the specific evaluation indicator. On the one hand, in the disclosed embodiment, three serially set models (data search model, data processing model and reasoning prediction model) are used to obtain the initial evaluation result for the target project under the specific evaluation indicator based on the evaluation intention. Therefore, the three models can be fine-tuned, tested and optimized independently, thereby improving the flexibility and scalability of the project evaluation method; on the other hand, the three serially set models can respectively perform preliminary selection, optimization and fusion processing on the data, thereby further improving the reliability of the target evaluation result.

[0118] The following, combined with Figure 5 , the complete process of the large model-based project evaluation method provided in the embodiment of the present disclosure is explained.

[0119] First, an initial intent issued by a user may be received.

[0120] After receiving the initial intent sent by the user, the initial intent can be identified using the intent recognition model to determine whether the initial intent has a clarity problem, and if it is determined that the initial intent does not have a clarity problem, the initial intent is used as an evaluation intent, and if it is determined that the initial intent has a clarity problem, the intent clarification process is entered, that is, query information is generated, the query information is broadcasted, and then the initial intent is optimized in response to the intent optimization request. After obtaining the optimized initial intent, if the optimized initial intent does not have a clarity problem, the optimized initial intent can be used as the evaluation intent; if the optimized initial intent has a clarity problem, the optimized initial intent is used as the new initial intent, and the steps of "if it is determined that the initial intent does not have a clarity problem, the initial intent is used as the evaluation intent; if it is determined that the initial intent has a clarity problem, the query information is generated; the query information is broadcasted; and the initial intent is optimized in response to the intent optimization request" are performed again until the optimized initial intent does not have a clarity problem, and the optimized initial intent is used as the evaluation intent. Among them, the intent recognition model can be a large language model.

[0121] After that, the large model can be used to obtain the initial evaluation results for the target project under specific evaluation indicators based on the evaluation intention. The large model can also be a large language model. Specifically, the large model can be used to:

[0122] Scheduling the data search model to obtain preliminary project data related to the target project and targeting specific evaluation indicators based on the evaluation intent using the data search model;

[0123] Schedule the data processing model to optimize the preliminary project data based on the evaluation intention and obtain the target project data by using the data processing model;

[0124] The predictive reasoning model is scheduled to utilize the predictive reasoning model to process the target project data based on the evaluation intention and obtain the initial evaluation results for the target project under specific evaluation indicators.

[0125] Finally, the result processing flow may be entered, that is, the initial evaluation result may be processed to obtain the target evaluation result for the target project. For example, the initial evaluation result may be processed according to the preset format requirements and / or the preset expression requirements to obtain the target evaluation result for the target project. The preset format requirements and the preset expression requirements may be set according to actual business needs or application scenarios, and the embodiments of the present disclosure do not limit this.

[0126] For the specific functions and examples of each step above, please refer to the relevant descriptions of the corresponding steps in the aforementioned project evaluation method embodiment based on the big model, which will not be repeated here.

[0127] See also Figure 6 , which is a schematic diagram of an application scenario of a large model-based project evaluation method provided in an embodiment of the present disclosure.

[0128] The project evaluation method based on a large model provided by the embodiment of the present disclosure is applied to an electronic device. The electronic device may be a server or a terminal device. Here, the terminal device may be a workbench, a mainframe computer, a conventional computer (e.g., a desktop computer, a laptop computer, a vehicle-mounted computer, etc.), a personal digital processing or other similar computing device.

[0129] Here, electronic devices are used to:

[0130] Obtain the evaluation intention for the target project; wherein the evaluation intention is used to request the target project to be evaluated under a specific evaluation indicator;

[0131] Using the big model, based on the evaluation intention, we can obtain the initial evaluation results for the target project under specific evaluation indicators;

[0132] Based on the initial assessment results, target assessment results for the target projects are obtained.

[0133] It should be noted that in the embodiments of the present disclosure, Figure 6 The application scenario diagram shown is only illustrative and not restrictive. Those skilled in the art can Figure 6 Various obvious changes and / or substitutions are made to the examples, and the obtained technical solutions still fall within the scope of the disclosure of the embodiments of the present disclosure.

[0134] In order to better implement the aforementioned project evaluation method based on a large model, the embodiment of the present disclosure also provides a project evaluation device based on a large model, which can be integrated into an electronic device. The electronic device can be a server or a terminal device. Here, the terminal device can be a workbench, a mainframe computer, a conventional computer (for example, a desktop computer, a laptop computer, a car computer, etc.), a personal digital processing or other similar computing device. Figure 7 The schematic structural block diagram shown illustrates a large model-based project evaluation device 700 provided in the disclosed embodiment.

[0135] The project evaluation device 700 based on the large model comprises:

[0136] The intention acquisition unit 701 is used to acquire the evaluation intention for the target project; wherein the evaluation intention is used to request the target project to be evaluated under a specific evaluation indicator;

[0137] The first result acquisition unit 702 is used to obtain an initial evaluation result for a target project under a specific evaluation indicator based on the evaluation intention by using the large model;

[0138] The second result obtaining unit 703 is used to obtain a target evaluation result for the target project based on the initial evaluation result.

[0139] In some optional implementations, the first result obtaining unit 702 is used to:

[0140] Scheduling the data search model to obtain preliminary project data related to the target project and targeting specific evaluation indicators based on the evaluation intent using the data search model;

[0141] Schedule the data processing model to optimize the preliminary project data based on the evaluation intention and obtain the target project data by using the data processing model;

[0142] The predictive reasoning model is scheduled to utilize the predictive reasoning model to process the target project data based on the evaluation intention and obtain the initial evaluation results for the target project under specific evaluation indicators.

[0143] In some optional implementations, the first result obtaining unit 702 is used to:

[0144] Based on the evaluation intention, determining at least one target data source from a plurality of candidate data sources;

[0145] Acquire original project data related to the target project and for a specific evaluation indicator from each of the at least one target data source to obtain at least one original project data corresponding one-to-one to the at least one target data source;

[0146] Based on at least one original project data, preliminary selected project data are obtained.

[0147] In some optional implementations, the first result obtaining unit 702 is used to:

[0148] Based on the evaluation intention, the preliminary project data is optimized from multiple optimization dimensions to obtain the target project data; wherein the multiple optimization dimensions include at least one of data deduplication, redundant data removal and merging of data with the same attributes.

[0149] In some optional implementations, the first result obtaining unit 702 is used to:

[0150] Based on the evaluation intention, the target evaluation strategy is obtained;

[0151] Obtain historical project data related to the target project and targeting specific evaluation indicators;

[0152] According to the target evaluation strategy, the target project data and historical project data are processed to obtain the initial evaluation results for the target project under specific evaluation indicators.

[0153] In some optional implementations, the first result obtaining unit 702 is used to:

[0154] Based on the target project data and historical project data, predict subsequent project data related to the target project and targeting specific evaluation indicators;

[0155] According to the target evaluation strategy, based on the target project data, historical project data and subsequent project data, the initial evaluation results for the target project under specific evaluation indicators are obtained.

[0156] In some optional implementations, the second result obtaining unit 703 is used to:

[0157] Get the initial intent for the target project;

[0158] Generate inquiry information when it is determined that there are clarity issues with the initial intent;

[0159] Report the inquiry information;

[0160] In response to an intent optimization request received after the query information is broadcast, the initial intent is optimized to obtain an evaluated intent.

[0161] In the embodiment of the present disclosure, the specific functions and examples of each unit in the large model-based project evaluation device 700 can be found in the relevant descriptions of the corresponding steps in the aforementioned large model-based project evaluation method embodiment, and will not be repeated here.

[0162] In the technical solution disclosed herein, the acquisition, storage and application of user personal information involved are in compliance with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0163] According to an embodiment of the present disclosure, the present disclosure also provides an electronic device, a readable storage medium, and a computer program product.

[0164] Figure 8 A schematic block diagram of an example electronic device 800 that can be used to implement an embodiment of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as vehicle-mounted computing devices, laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present disclosure described and / or required herein.

[0165] like Figure 8 As shown, the electronic device 800 includes a computing unit 801, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 802 or a computer program loaded from a storage unit 808 to a random access memory (RAM) 803. In RAM 803, various programs and data required for the operation of the electronic device 800 can also be stored. The computing unit 801, ROM 802, and RAM 803 are connected to each other via a bus 804. An input / output (I / O) interface 805 is also connected to the bus 804.

[0166] A number of components in the electronic device 800 are connected to the I / O interface 805, including: an input unit 806, such as a keyboard, a mouse, etc.; an output unit 807, such as various types of renderers, speakers, etc.; a storage unit 808, such as a disk, an optical disk, etc.; and a communication unit 809, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 809 allows the electronic device 800 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0167] The computing unit 801 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the computing unit 801 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, digital signal processors (DSP), and any appropriate processors, controllers, microcontrollers, etc. The computing unit 801 performs the various methods and processes described above, for example, a project evaluation method based on a large model. For example, in some embodiments, the project evaluation method based on a large model may be implemented as a computer software program, which is tangibly included in a machine-readable medium, for example, a storage unit 808. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 800 via ROM 802 and / or a communication unit 809. When the computer program is loaded into RAM 803 and executed by the computing unit 801, one or more steps in the project evaluation method based on the large model described above may be executed. Alternatively, in other embodiments, the computing unit 801 may be configured as a large model-based project evaluation method in any other appropriate manner (eg, by means of firmware).

[0168] Various embodiments of the systems and techniques described above herein may be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0169] The program code for implementing the method of the present disclosure can be written in any combination of one or more programming languages. These program codes can be provided to a processor or controller of a general-purpose computer, a special-purpose computer or other programmable data optimization device so that the program code, when executed by the processor or controller, enables the functions / operations specified in the flow chart and / or block diagram to be implemented. The program code can be executed completely on the machine, partially on the machine, partially on the machine as an independent software package and partially on a remote machine or completely on a remote machine or server.

[0170] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a RAM, a 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 foregoing.

[0171] To provide interaction with a user, the systems and techniques described herein may be implemented on a computer having: a rendering device (e.g., a cathode ray tube (CRT) renderer or a liquid crystal display (LCD)) for rendering information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices are also used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0172] The systems and techniques described herein can be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: Local Area Network (LAN), Wide Area Network (WAN), and the Internet.

[0173] The computer system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The relationship between the client and the server is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server can be a cloud server, a server of a distributed system, or a server combined with a blockchain.

[0174] The embodiment of the present disclosure also provides a non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to enable a computer to execute a project evaluation method based on a large model.

[0175] The embodiment of the present disclosure also provides a computer program product, including a computer program, which implements a large model-based project evaluation method when executed by a processor.

[0176] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps recorded in the present disclosure can be executed in parallel, sequentially or in a different order, as long as the desired results of the technical solutions disclosed in the present disclosure can be achieved, and this document does not limit this. In addition, in the present disclosure, relational terms such as "first", "second", "third", etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply the existence of any such actual relationship or order between these entities or operations. In addition, "multiple" in the present disclosure can be understood as at least two.

[0177] The above specific implementations do not constitute a limitation on the protection scope of the present disclosure. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the principles of the present disclosure shall be included in the protection scope of the present disclosure.

Claims

1. A project evaluation method based on a large model, comprising: Acquire an evaluation intention for a target project; wherein the evaluation intention is used to request that the target project be evaluated under a specific evaluation indicator; Using the large model, based on the evaluation intention, an initial evaluation result for the target project under the specific evaluation indicator is obtained; Based on the initial evaluation result, a target evaluation result for the target project is obtained.

2. The method according to claim 1, wherein: The obtaining of the initial evaluation result for the target project under the specific evaluation indicator based on the evaluation intention includes: Scheduling a data search model to obtain preliminary selected project data related to the target project and targeting the specific evaluation indicator based on the evaluation intention by using the data search model; Scheduling a data processing model to optimize the preliminary selected project data based on the evaluation intention using the data processing model to obtain target project data; The predictive reasoning model is scheduled to utilize the predictive reasoning model to process the target project data based on the evaluation intention to obtain an initial evaluation result for the target project under the specific evaluation indicator.

3. The method according to claim 2, wherein: The step of obtaining preliminary selected project data related to the target project and targeting the specific evaluation indicator based on the evaluation intention includes: Based on the evaluation intention, determining at least one target data source from a plurality of candidate data sources; Acquire original project data related to the target project and for the specific evaluation indicator from each of the at least one target data source to obtain at least one original project data corresponding to the at least one target data source one by one; The preliminary selected project data is obtained based on the at least one original project data.

4. The method according to claim 2, wherein: The step of optimizing the preliminary selected project data based on the evaluation intention to obtain target project data includes: Based on the evaluation intention, the preliminary project data is optimized from multiple optimization dimensions to obtain the target project data; wherein the multiple optimization dimensions include at least one of data deduplication, redundant data removal and merging of data with the same attributes.

5. The method according to claim 2, wherein: The step of processing the target project data based on the evaluation intention to obtain an initial evaluation result for the target project under the specific evaluation indicator includes: Based on the evaluation intention, a target evaluation strategy is obtained; Acquire historical project data related to the target project and targeting the specific evaluation indicator; According to the target evaluation strategy, the target project data and the historical project data are processed to obtain an initial evaluation result for the target project under the specific evaluation indicator.

6. The method according to claim 5, wherein: The step of processing the target project data and the historical project data according to the target evaluation strategy to obtain an initial evaluation result for the target project under the specific evaluation indicator includes: Based on the target project data and the historical project data, predicting subsequent project data related to the target project and targeting the specific evaluation indicator; According to the target evaluation strategy, based on the target project data, the historical project data and the subsequent project data, an initial evaluation result for the target project under the specific evaluation indicator is obtained.

7. The method according to any one of claims 1 to 6, wherein: The obtaining of the evaluation intention for the target project includes: Get the initial intent for the target project; If it is determined that the initial intent has a clarity problem, generating an inquiry message; reporting the inquiry information; In response to an intent optimization request received after the query information is broadcast, the initial intent is optimized to obtain the evaluated intent.

8. A project evaluation device based on a large model, comprising: An intention acquisition unit, used to acquire an evaluation intention for a target project; wherein the evaluation intention is used to request that the target project be evaluated under a specific evaluation indicator; A first result acquisition unit is used to obtain an initial evaluation result for the target project under the specific evaluation indicator based on the evaluation intention by using the large model; The second result obtaining unit is used to obtain a target evaluation result for the target project based on the initial evaluation result.

9. The device according to claim 8, wherein: The first result obtaining unit is used for: Scheduling a data search model to obtain preliminary selected project data related to the target project and targeting the specific evaluation indicator based on the evaluation intention by using the data search model; Scheduling a data processing model to optimize the preliminary selected project data based on the evaluation intention using the data processing model to obtain target project data; The predictive reasoning model is scheduled to utilize the predictive reasoning model to process the target project data based on the evaluation intention to obtain an initial evaluation result for the target project under the specific evaluation indicator.

10. The device according to claim 9, wherein: The first result obtaining unit is used for: Based on the evaluation intention, determining at least one target data source from a plurality of candidate data sources; Acquire original project data related to the target project and for the specific evaluation indicator from each of the at least one target data source to obtain at least one original project data corresponding to the at least one target data source one by one; The preliminary selected project data is obtained based on the at least one original project data.

11. The device according to claim 9, wherein: The first result obtaining unit is used for: Based on the evaluation intention, the preliminary project data is optimized from multiple optimization dimensions to obtain the target project data; wherein the multiple optimization dimensions include at least one of data deduplication, redundant data removal and merging of data with the same attributes.

12. The device according to claim 9, wherein: The first result obtaining unit is used for: Based on the evaluation intention, a target evaluation strategy is obtained; Acquire historical project data related to the target project and targeting the specific evaluation indicator; According to the target evaluation strategy, the target project data and the historical project data are processed to obtain an initial evaluation result for the target project under the specific evaluation indicator.

13. The device according to claim 12, wherein: The first result obtaining unit is used for: Based on the target project data and the historical project data, predicting subsequent project data related to the target project and targeting the specific evaluation indicator; According to the target evaluation strategy, based on the target project data, the historical project data and the subsequent project data, an initial evaluation result for the target project under the specific evaluation indicator is obtained.

14. The device according to any one of claims 8 to 13, wherein: The second result obtaining unit is used for: Get the initial intent for the target project; If it is determined that the initial intent has a clarity problem, generating an inquiry message; reporting the inquiry information; In response to an intent optimization request received after the query information is broadcast, the initial intent is optimized to obtain the evaluated intent.

15. An electronic device, comprising: at least one processor; a memory communicatively coupled to the at least one processor; The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can perform the method according to any one of claims 1 to 7.

16. A non-transitory computer-readable storage medium storing computer instructions, wherein: The computer instructions are used to cause the computer to execute the method according to any one of claims 1 to 7.

17. A computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the computer program implements the method according to any one of claims 1 to 7.

Images