Matching object determination method and device and electronic equipment

Abstract

The invention provides a matching object determination method and device and electronic equipment, and the method comprises the steps: inputting innovation operation data of a target enterprise into a pre-trained multi-task model, and obtaining a current target innovation stage of the target enterprise and a capability index of an optimal matching object corresponding to the target innovation stage of the target enterprise; according to the target innovation stage, obtaining a target capability index model corresponding to the target innovation stage; inputting the auxiliary innovation capability data, corresponding to the target innovation stage, of the plurality of candidate matching objects into a target capability index model to obtain a plurality of capability indexes corresponding to the plurality of candidate matching objects; respectively matching the plurality of capability indexes with the capability index of the optimal matching object to obtain a first matching result; and screening out a target matching object of the target enterprise from the plurality of candidate matching objects according to the first matching result.

Description

Technical Field

[0001] The present disclosure relates to the field of computer technology, and in particular to a matching object determination method, device and electronic device. Background Art

[0002] The development process of enterprises will be accompanied by various innovative activities. Scientific and effective innovation activities can also promote the vigorous development of the enterprise itself. It can be said that the development of an enterprise is inseparable from the development of innovation activities. Therefore, in order to achieve better development, enterprises are also beginning to pay attention to their own innovation.

[0003] Today's innovation is no longer an individual behavior, but an ecological behavior, and the innovation chain has been divided into multiple innovation stages. At each innovation stage, enterprises need the intervention of other partners to better achieve the goals of each innovation stage, and then be able to fully open up the entire innovation chain.

[0004] Innovation chain is still a relatively new concept for most enterprises. Therefore, it is difficult for enterprises to analyze their own innovation stage based on their own corporate data. Moreover, they may not be able to accurately determine their own innovation stage based only on a small amount of their own data, and they will not be able to find partners suitable for their current innovation stage. Summary of the invention

[0005] The purpose of the present disclosure is to provide a matching object determination method, device and electronic device.

[0006] According to one aspect of the present disclosure, a matching object determination method is provided, comprising:

[0007] Inputting the target enterprise's innovation operation data into a pre-trained multi-task model to obtain the target innovation stage to which the target enterprise currently belongs and the capability index of the best matching object corresponding to the target innovation stage of the target enterprise;

[0008] Obtaining a target capability indicator model corresponding to the target innovation stage;

[0009] Inputting auxiliary innovation capability data of a plurality of candidate matching objects corresponding to the target innovation stage into the target capability indicator model to obtain a plurality of capability indicators corresponding to the plurality of candidate matching objects respectively;

[0010] Using the plurality of capability indicators to match the capability indicators of the best matching object respectively to obtain a first matching result;

[0011] A target matching object of the target enterprise is screened out from the plurality of candidate matching objects according to the first matching result.

[0012] According to a second aspect of the present application, a matching object determination device is disclosed, comprising:

[0013] A multi-task model calling module is used to input the innovation operation data of the target enterprise into the pre-trained multi-task model to obtain the target innovation stage to which the target enterprise currently belongs and the capability index of the best matching object corresponding to the target innovation stage of the target enterprise;

[0014] An indicator model acquisition module, used to acquire a target capability indicator model corresponding to the target innovation stage;

[0015] A capability indicator determination module, used to input auxiliary innovation capability data of a plurality of candidate matching objects corresponding to the target innovation stage into the target capability indicator model, and obtain a plurality of capability indicators corresponding to the plurality of candidate matching objects respectively;

[0016] A matching result acquisition module, used to respectively match the several capability indicators with the capability indicators of the best matching object to obtain a first matching result;

[0017] The matching object determination module is used to screen out a target matching object of the target enterprise from the plurality of candidate matching objects according to the first matching result.

[0018] According to a third aspect of the present disclosure, an electronic device is provided, including a processor and a memory, wherein the memory stores a computer program, and the processor executes the computer program stored in the memory to implement the steps in the method described in the first aspect above.

[0019] According to a fourth aspect of the present disclosure, a computer program is provided, comprising computer instructions, which, when executed by a processor, implement the steps in the method described in the first aspect above.

[0020] According to a fifth aspect of the present disclosure, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the method described in the first aspect are implemented. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 A flowchart of a matching object determination method provided by an embodiment of the present disclosure;

[0022] Figure 2 A flowchart of another matching object determination method provided by an embodiment of the present disclosure;

[0023] Figure 3 A schematic diagram flow for obtaining a capability indicator model provided by an embodiment of the present disclosure;

[0024] Figure 4 A flowchart of another matching object determination method provided by an embodiment of the present disclosure;

[0025] Figure 5 A schematic diagram of the structure of a matching object determination device provided by an embodiment of the present disclosure;

[0026] Figure 6 A schematic diagram of the structure of an electronic device provided by an embodiment of the present disclosure. DETAILED DESCRIPTION

[0027] Before introducing the embodiments of the present disclosure, it should be noted that:

[0028] Some embodiments of the present disclosure are described as processing flows. Although the various operation steps of the flow may be given sequential step numbers, the operation steps therein may be implemented in parallel, concurrently or simultaneously.

[0029] In the embodiments of the present disclosure, the terms "first", "second", etc. may be used to describe various features, but these features should not be limited by these terms. These terms are used only to distinguish one feature from another.

[0030] The term “and / or” may be used in embodiments of the present disclosure. “And / or” includes any and all combinations of one or more of the associated features listed.

[0031] It should be understood that when describing the connection or communication relationship between two components, unless it is explicitly stated that the two components are directly connected or directly communicating, the connection or communication between the two components can be understood as direct connection or communication, or as indirect connection or communication through an intermediate component.

[0032] In order to make the technical solutions and advantages of the embodiments of the present disclosure more clearly understood, the exemplary embodiments of the present disclosure are further described in detail below in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present disclosure, rather than an exhaustive list of all the embodiments. It should be noted that the embodiments of the present disclosure and the features in the embodiments can be combined with each other without conflict.

[0033] Scientific and effective innovation activities can promote the vigorous development of the enterprise itself. While opening up the innovation chain, it usually also means that the enterprise has achieved better development. Therefore, currently all small and medium-sized enterprises are paying close attention to their own innovation activities.

[0034] The innovation chain includes multiple innovation stages. Innovation includes different specific operations for different types of enterprises, but in general, the innovation chain includes five stages (refer to ISO 56005 international standard): identifying opportunities, creating concepts, verifying concepts, developing solutions, and deploying solutions. Among them, the opportunity identification stage aims to identify and define innovation opportunities based on the internal and external environment of the organization and its innovation, innovation intentions, the scope of the innovation plan, and the experience and lessons of previous innovation plans; the concept creation stage aims to select appropriate ideas and possible solutions based on the identified and defined innovation opportunities and established criteria to form a concept with a preliminary value realization model; the concept verification stage aims to conduct verification in a variety of ways based on the created concepts to evaluate the feasibility of the created concepts and deal with uncertainties, assumptions or assumptions that still need to be resolved; the solution development stage aims to develop feasible solutions based on the verified concepts and determine the required activities, resources, relationships and time plans to meet the needs of comprehensive or phased deployment of solutions; the deployment solution stage aims to provide solutions to users, customers, partners and other stakeholders based on the developed solutions, and monitor feedback results to improve solutions.

[0035] Taking common technology companies as an example, identifying opportunities includes: innovation generation, feasibility analysis and project establishment. The source of innovation may come from its own scientific research results, or from the scientific research results of cooperating universities and scientific research institutions. The source of innovation may also come from customer needs, technology updates, etc. Feasibility analysis includes technical feasibility analysis, business opportunity feasibility analysis, etc. Creating concepts includes: product design and research and development (such as making samples); verifying concepts includes: conducting verification tests of different scales on products; development plans include: putting the tested products into production and mass production; the final deployment plan includes: continuous operation of the product after mass production, including product iteration, market development, etc. The specific behaviors of the overall innovation process of different types of enterprises may be different, but the overall innovation process is similar, and this application will not go into details here.

[0036] Today's innovation is no longer an individual behavior, but an ecological behavior. At each innovation stage of the above-mentioned innovation process, enterprises need the intervention of other partners to better achieve the goals of each innovation stage. For example, in the opportunity identification stage, the participation of universities or research institutions may be needed to achieve better scientific research results. In the concept creation stage, the participation of design companies, product manufacturers and other stakeholders may be needed. In the concept verification stage, the participation of testing agencies and other stakeholders is needed. In the development plan stage, the participation of factories and other stakeholders is needed. The deployment plan requires the participation of marketing companies, investors and other stakeholders, etc.

[0037] Innovation chain is still a relatively new concept for most enterprises. Therefore, it is difficult for enterprises to analyze their own innovation stage based on their own corporate data. Moreover, they may not be able to accurately determine their own innovation stage based only on a small amount of their own data, and they will not be able to find partners suitable for their current innovation stage.

[0038] like Figure 1 As shown, in response to the above problem, the present application proposes a matching object determination method, including:

[0039] S101, inputting the innovation operation data of the target enterprise into a pre-trained multi-task model to obtain the target innovation stage to which the target enterprise currently belongs and the capability index of the best matching object corresponding to the target innovation stage of the target enterprise;

[0040] S102, obtaining a target capability indicator model corresponding to the target innovation stage;

[0041] S103, inputting auxiliary innovation capability data of a plurality of candidate matching objects corresponding to the target innovation stage into the target capability indicator model, and obtaining a plurality of capability indicators corresponding to the plurality of candidate matching objects respectively;

[0042] S104, using the acquired capability indicators to match the capability indicators of the best matching object respectively to obtain a first matching result;

[0043] S105: Filter out a target matching object of a target enterprise from a plurality of candidate matching objects according to the first matching result.

[0044] like Figure 2As shown, it is a schematic diagram of the above overall process. With the above method, the target enterprise that needs to find a partner in the innovation process can input its own innovation operation data, and the background can call the multi-task model and the capability index model to screen out the matching objects that meet the current innovation stage of the target enterprise from the massive candidate matching objects, thereby assisting the target enterprise to complete its innovation activities. It can be understood that the matching object determination method of the present application can be deployed in the server in the form of a software program, and provided to the target enterprise in the form of an application program or a web page. The relevant personnel of the target enterprise log in to the web page or application program, and as long as they enter their own innovation operation data, they can quickly know their own innovation stage, and at the same time, they can know the matching objects that meet their own needs. The candidate matching objects mentioned above can be other units that are pre-registered and can assist various enterprises in their innovation activities. For example, the candidate matching objects can be factories that can mass produce products, scientific research institutions that can provide scientific research capabilities, etc. Since the multi-task model and the capability index model are obtained based on massive big data training, compared with the analysis of the enterprise based on its own small amount of data, it can ensure that the innovation stage of the enterprise currently determined is more accurate, and further ensure that the target matching objects found are more in line with their own needs.

[0045] The following is an introduction to the training process of the multi-task model mentioned above.

[0046] According to the above, the multi-task model is used to input the innovation and operation data of the target enterprise, obtain the target innovation stage to which the target enterprise currently belongs, and the capability index of the best matching object corresponding to the target innovation stage of the target enterprise. That is, the two tasks of the multi-task model are: predicting the target innovation stage to which the enterprise currently belongs based on the innovation and operation data of the enterprise, and predicting the capability index of the best matching object corresponding to the target enterprise in the target innovation stage based on the innovation and operation data of the enterprise. Since the two tasks are highly correlated, the overall features required by the tasks are the same but the labels are different. Therefore, in order to make the model accuracy of the two tasks better, this application proposes to adopt a multi-task learning method to train the multi-task model using these two more related tasks. The training data set of the multi-task model uses all the training sample sets of the two tasks, which is equivalent to an implicit data enhancement, which can improve the generalization ability of the model. Therefore, the trained multi-task model has better prediction effect than the model of each single task.

[0047] Specifically, the training process of the multi-task model includes the following operations.

[0048] A first training sample set is obtained for the first task, wherein the first task is to predict the innovation stage of the enterprise based on the innovation operation data of the enterprise; each training sample in the first training sample set uses the innovation operation data of the enterprise as a feature value and the innovation stage of the enterprise as a label value.

[0049] A second training sample set is obtained for the second task, and the second task is to predict the capability index of the best matching object of the enterprise at its innovation stage based on the innovation operation data of the enterprise; each training sample in the second training sample set uses the innovation operation data of the enterprise as a feature value, and uses the capability index of the best matching object of the enterprise at its innovation stage as a label value.

[0050] As shown in Table 1, these are the contents of the two training sample sets.

[0051]

[0052]

[0053] Table 1

[0054] The characteristic values ​​in the first training sample set and the second training sample set, that is, the innovation operation data, can be obtained through the websites of various enterprises and the data provided by the enterprises themselves when participating in various selections. The innovation operation data is specifically used to describe some key data of the enterprise in its innovation process, such as the type of enterprise, business scope, whether there are scientific research results to be productized, the type of scientific research results, whether there are existing products, product testing status, product mass production status, product market size, product update status and other data.

[0055] The label value in the first training sample set, that is, the innovation stage to which the enterprise belongs, can be manually annotated, and the manually annotated label value can be directly obtained when it is acquired.

[0056] The label value in the second training sample set, i.e., the capability index of the best matching object in the innovation stage to which the enterprise belongs, is relatively abstract. It is a kind of data that needs to be abstracted from the data of the matching object itself and can evaluate the matching object's ability to assist the target enterprise in innovating at a certain innovation stage, rather than actual existing data. Therefore, it is difficult to label manually. Based on this, the present disclosure proposes a specific method for obtaining the matching object capability index, which can be specifically referred to below and will not be described in detail here.

[0057] After obtaining the first training sample set and the second training sample set, the first training sample set and the second training sample set may be used to perform supervised training on the initial multi-task model to obtain a trained multi-task model.

[0058] In one embodiment, the initial multi-task model is supervised trained using the first training sample set and the second training sample set to obtain the trained multi-task model. The multi-task model can be first supervised trained using the training samples in the first training sample set, and then supervised trained using the training samples in the second training sample set after the training is completed, so that the trained multi-task model can process two tasks.

[0059] In another embodiment, the initial multi-task model is supervisedly trained using the first training sample set and the second training sample set to obtain a trained multi-task model, which may also be:

[0060] The first training sample set and the second training sample set are divided into several groups, each group containing several training samples; any group of training samples in the first training sample set is selected to be combined with any group of training samples in the second training sample set to obtain several groups of combined training samples; each group of combined training samples is used as the training sample for each iteration to perform supervised training on the multi-task model, wherein the loss function of the multi-task model includes the losses of the first task and the second task.

[0061] That is, the training sample set of each group includes both the training samples in the first training sample set and the training samples in the second training sample set, so that the parameters of the model can be adjusted towards the training direction of the two tasks in each iteration.

[0062] The multi-task model in this application can be an existing multi-task model such as the Share-Bottom model. This application does not limit the specific type of the multi-task model, as long as it can predict two output results based on the input data.

[0063] Considering that in different innovation stages, the target enterprise needs different capabilities of matching objects (partners). As mentioned above, in the opportunity identification stage, more attention may be paid to the partner's scientific research capabilities, while in the solution development stage, more attention may be paid to the partner's production and manufacturing capabilities. Therefore, in different innovation stages, the innovation capability indicators of matching objects (partners) are different. For this reason, the present disclosure proposes that in one embodiment, different capability indicator models can be pre-trained according to different innovation stages, and the trained capability indicator models are stored, wherein the storage index of the capability indicator model is the corresponding innovation stage identifier.

[0064] As mentioned above, there are five innovation stages in total. The capability indicator models can be pre-trained and stored for each of the five innovation stages.

[0065] Table 2 shows the correspondence between the capability indicator model and the innovation stage identification.

[0066]

[0067] Table 2

[0068] Further, the above S102 may specifically be: obtaining the target capability indicator model according to the innovation stage identifier of the target innovation stage. For example, when the target innovation stage of the target enterprise is determined to be innovation stage 2, i.e., concept creation stage, capability indicator model 2 is obtained locally using the identifier of innovation stage 2 as a storage index.

[0069] As described above, the capability index itself is relatively abstract. It is a kind of data that needs to be abstracted from the data of the matching object itself and can evaluate the matching object's ability to assist the target enterprise in innovating at a certain innovation stage. It is difficult to define manually. Therefore, when training the capability index model for any innovation stage, it is impossible to directly obtain the corresponding capability index as a label for training.

[0070] Furthermore, the present disclosure proposes that the capability indicator model can be obtained by training for different innovation stages in the following manner.

[0071] For each innovation stage, perform the following steps to obtain the capability indicator model corresponding to the innovation stage:

[0072] Obtaining auxiliary innovation capability data corresponding to the innovation stage of multiple enterprises, as well as auxiliary innovation capability scores at the innovation stage;

[0073] The acquired data is used to construct a third training sample set, wherein in each training sample, the auxiliary innovation capability data of the enterprise at the innovation stage is used as the feature value, and the auxiliary innovation capability score of the enterprise at the innovation stage is used as the label value;

[0074] Training the deep learning model using the constructed third training sample set;

[0075] The output layer of the trained deep learning model is deleted, and the input layer and hidden layer of the deep learning model are retained to obtain a capability indicator model corresponding to the innovation stage.

[0076] Capability indicators are relatively abstract and difficult to define manually, but it is relatively easy to define the auxiliary innovation capability score of each matching object at a certain innovation stage. For example, a percentage score can be used to score a company based on its auxiliary innovation results. In addition, on some professional evaluation websites, the company's auxiliary innovation capability score can also be directly obtained.

[0077] The auxiliary innovation capability data is usually different data at different innovation stages. For example, in the opportunity identification stage, it can be data used to evaluate scientific research capabilities, while in the development plan stage, it can be data used to evaluate production and manufacturing capabilities. After the auxiliary innovation capability data and auxiliary innovation capability scores are obtained for any innovation stage, the obtained data can be used to construct a third training sample set.

[0078] After constructing the third training sample set, the third training sample set is used to train the deep learning model. The input and output of the trained deep learning model can reflect the relationship between the auxiliary innovation capability data and the auxiliary innovation capability score. The deep learning model usually includes an input layer, at least one hidden layer and an output layer. The hidden layer of deep learning can abstract the feature vector of the input data to the same dimensional space to show its more abstract characteristics, that is, the input auxiliary innovation capability data originally belonging to different dimensions can be converted into more abstract feature data belonging to the same dimension, and the abstracted feature data contains the relationship between various input data, that is, various auxiliary innovation capability data, and the abstracted feature data can also be used to predict the auxiliary innovation capability score. Therefore, the present application proposes that the output of the last hidden layer of the deep learning model can be used as the capability index data of the matching object corresponding to the input auxiliary innovation capability data, that is, the feature abstraction ability of the hidden layer of the deep learning model itself can be used to extract the innovation capability index of each matching object.

[0079] Therefore, after obtaining the trained deep learning model, the output layer of the trained deep learning model is deleted, and the input layer and hidden layer of the deep learning model are retained, so that the capability indicator model corresponding to the innovation stage can be obtained.

[0080] like Figure 3 As shown, it is a schematic diagram of the capability index model obtained above, that is, after deleting the output layer of the deep learning model trained with the third training sample set, the capability index model is obtained.

[0081] As mentioned above, when training the multi-task model, each training sample in the second training sample set uses the enterprise's innovation operation data as the feature value and the capability index of the best matching object of the enterprise in its innovation stage as the label value. The best matching object can be determined manually, that is, the matching object can be judged as the best matching object based on the innovation results, but the capability index of the best matching object cannot be defined manually.

[0082] Therefore, the present application proposes that the second training sample set may be obtained for the second task in the following manner:

[0083] After obtaining the innovation operation data of any enterprise, obtain the auxiliary innovation capability data of all dimensions corresponding to the best matching object of the enterprise;

[0084] Screening the auxiliary innovation capability data of all dimensions of the best matching object to obtain the auxiliary innovation capability data of the best matching object at the innovation stage to which the target enterprise belongs;

[0085] Input the screened auxiliary innovation capability data into the capability indicator model corresponding to the innovation stage to obtain the capability indicator of the best matching object;

[0086] After executing the above steps for the best matching objects corresponding to all enterprises in the second training sample set, all label values ​​in the second training sample set are obtained.

[0087] That is, the full-dimensional auxiliary innovation capability data of the best matching object includes auxiliary innovation capability data of each innovation stage. In order to obtain the label value, it is necessary to screen out the auxiliary innovation capability data corresponding to the innovation stage to which the target enterprise belongs. For example, if the innovation stage to which the target enterprise belongs is innovation stage 1, then the scientific research capability-related data that is more concerned with innovation stage 1, i.e., the opportunity identification stage, is screened out, and the screened data is input into the capability indicator model corresponding to innovation stage 1 to obtain the capability indicator of the best matching object, and the obtained capability indicator is used as the label value in the second training sample set. The above method can be used to obtain the capability indicator that is difficult to define manually as the label value in the second training sample set, which is subsequently used for training the multi-task model.

[0088] In the above S103, since the target capability indicator model is pre-trained and can output capability indicators, after the auxiliary innovation capability data of several candidate matching objects are input into the target capability indicator model, the capability indicator of each candidate matching object can be obtained.

[0089] S104-105 may be specifically performed in the following manner.

[0090] In S104, specifically, the similarity between several capability indicators and the capability indicators of the best matching object may be determined, that is, the first matching result is the similarity between several capability indicators and the capability indicators of the best matching object. In S105, specifically, the candidate matching object having the highest similarity with the capability indicators of the best matching object may be determined as the target matching object.

[0091] That is, since the capability index is feature information in the same feature space output by the hidden layer of the deep learning model, a preset similarity algorithm can be used to determine the similarity between several capability indicators of several candidate matching objects and the capability indicators of the best matching object in the feature space. The higher the similarity with the capability indicators of the best matching object, the closer the auxiliary innovation capability of the candidate matching object is to the auxiliary innovation capability of the best matching object, and the auxiliary innovation capability data of various dimensions are also closer. Therefore, the candidate matching object with the highest similarity with the capability indicators of the best matching object can be determined as the target matching object. The preset similarity algorithm can be to calculate the Euclidean distance, calculate the Manhattan distance, etc., and the present disclosure does not limit the similarity algorithm.

[0092] Considering that when the target enterprise is looking for partners during the innovation process, in addition to looking for partners based on the needs of the current innovation stage, it usually also needs to look for partners based on the needs of the next innovation stage. For example, the target enterprise is currently in the opportunity identification stage, and the next innovation stage will enter the concept creation stage. If it can find a partner that can provide itself with the ability to assist innovation in both innovation stages, it does not need to look for a partner again in the next innovation stage. In addition, the target enterprise's own innovation operation data will change as the innovation progresses. Therefore, if the target enterprise's current innovation operation data is used to predict partners for more stages in the future, the predicted partners will not meet the actual needs of the target enterprise at that time.

[0093] Therefore, the present application proposes that the candidate matching objects that the target enterprise needs most in the current innovation stage and the next innovation stage can be determined as the target matching objects based on the target enterprise's innovation operation data.

[0094] Specifically, in one implementation, different matching object prediction models can be pre-trained according to different innovation stages. That is, there are different matching object prediction models for different innovation stages, the input of the matching object prediction model is the innovation operation data of the target enterprise, and the output is the capability index of the best matching object corresponding to the target enterprise in the innovation stage corresponding to the matching object prediction model.

[0095] The difference between the matching object prediction model and the capability indicator model mentioned above is that the matching object prediction model inputs the innovation operation data of the target enterprise and outputs the capability indicator data of the best matching object corresponding to the target enterprise at this stage, while the capability indicator model inputs the auxiliary innovation capability data of the matching object and outputs the capability indicator data of the matching object at this stage.

[0096] In each training sample in the training sample set used to train the matching object prediction model, the innovation operation data of the enterprise at this innovation stage is used as the feature value, and the capability index of the best matching object of the enterprise at this innovation stage is used as the label value. The acquisition of capability index can be referred to above, and will not be repeated here.

[0097] When predicting the target enterprise's current innovation stage and the target matching object of the next innovation stage based on the target enterprise's innovation operation data, the enterprise's innovation operation data can be specifically input into a pre-trained matching object prediction model corresponding to the next innovation stage of the target innovation stage to obtain the capability index of the best matching object corresponding to the target enterprise in the next innovation stage; according to the next innovation stage of the target enterprise, obtain the capability index model corresponding to the next innovation stage, for example, obtain the capability index model corresponding to the next innovation stage according to the identifier of the next innovation stage. Input the auxiliary innovation capability data of several candidate matching objects corresponding to the next innovation stage into the capability index model of the next innovation stage to obtain several capability indicators corresponding to several candidate matching objects respectively;

[0098] Using the plurality of capability indicators to match the capability indicators of the best matching object corresponding to the next innovation stage, respectively, to obtain a second matching result;

[0099] Based on the first matching result and the second matching result above, a target matching object of the target enterprise is screened out from a number of candidate matching objects.

[0100] That is, the matching object prediction model and capability index model of the next innovation stage of the corresponding target innovation stage that have been trained in advance are used to obtain the second matching results of several candidate matching objects and the best matching object of the next innovation stage of the target enterprise, and then the target matching object of the target enterprise is comprehensively determined by using the two matching results, namely the first matching result and the second matching result. The target matching object is screened out according to the matching results of the candidate matching object and the best matching object of the target enterprise in two consecutive innovation stages, so that the screened target matching object can meet the innovation needs of the target enterprise in two consecutive innovation stages as much as possible.

[0101] In one implementation, the first matching result and the second matching result are combined to select a target matching object of the target enterprise from a plurality of candidate matching objects, including:

[0102] Determine, based on the first matching result, a first similarity of capability indicators of any candidate matching object and the best matching object of the target enterprise at the target innovation stage for the target innovation stage;

[0103] Determine, based on the second matching result, a second similarity of capability indicators of any candidate matching object and the best matching object of the target enterprise in the next innovation stage for the next innovation stage;

[0104] The first similarity and the second similarity are combined to select a target matching object of the target enterprise from a plurality of candidate matching objects.

[0105] Specifically, the first similarity and the second similarity of any candidate matching object can be calculated in the manner described above, which will not be repeated here. After obtaining the first similarity and the second similarity, the average value of the first similarity and the second similarity can be calculated for each candidate matching object, and the candidate matching object with the largest average value is determined as the target matching object.

[0106] In addition, different weights may be assigned to the first similarity and the second similarity, respectively, and the target matching object may be determined based on the similarities to which the weights are assigned.

[0107] Considering that the target enterprise is currently in the innovation stage where its partners are more important, and that the target enterprise's innovation operation data may change slightly in the next innovation stage, the weight of the first similarity can be set larger, and the weight of the second similarity can be set smaller, and the target matching object can be determined according to the set weights. For example, the weight of the first similarity is set to 0.6, and the weight of the second similarity is set to 0.4, and the comprehensive similarity data corresponding to each candidate matching object is calculated by the formula A=0.6B1+0.4B2. Among them, B1 is the first similarity and B2 is the second similarity. If the comprehensive similarity A is larger, it means that in two consecutive innovation stages, the candidate matching object is closer to the ability index of the best matching object of the target enterprise, so the candidate matching object with the largest comprehensive similarity A is selected as the target matching object of the target enterprise.

[0108] like Figure 4 As shown, it is a flowchart of determining the candidate matching objects that the target enterprise needs most in the current innovation stage and the next innovation stage according to the innovation operation data of the target enterprise as the target matching object. Since the multi-task model can output the target innovation stage to which the target enterprise currently belongs, the identifier of the target innovation stage and the identifier of the next innovation stage of the target innovation stage can be obtained according to the target innovation stage to which the target enterprise currently belongs, and the target capability indicator model, the next stage capability indicator model, and the next stage matching object prediction model can be obtained respectively using the two identifiers.

[0109] like Figure 5 As shown, based on the same inventive concept, the embodiment of the present disclosure further provides a matching object determination device, the device comprising:

[0110] The multi-task model calling module 510 is used to input the innovation operation data of the target enterprise into the pre-trained multi-task model to obtain the target innovation stage to which the target enterprise currently belongs and the capability index of the best matching object corresponding to the target innovation stage of the target enterprise;

[0111] An indicator model acquisition module 520 is used to acquire a target capability indicator model corresponding to the target innovation stage according to the target innovation stage;

[0112] A capability indicator determination module 530 is used to input auxiliary innovation capability data of a plurality of candidate matching objects corresponding to the target innovation stage into the target capability indicator model to obtain a plurality of capability indicators corresponding to the plurality of candidate matching objects respectively;

[0113] A matching result acquisition module 540 is used to respectively match the plurality of capability indicators with the capability indicators of the best matching object to obtain a first matching result;

[0114] The matching object determination module 550 is used to select a target matching object of the target enterprise from the plurality of candidate matching objects according to the first matching result.

[0115] In one embodiment, the apparatus further includes a model training module 560;

[0116] The model training module 560 is used to obtain a first training sample set for a first task, wherein the first task is to predict the innovation stage of the enterprise based on the innovation operation data of the enterprise; each training sample in the first training sample set uses the innovation operation data of the enterprise as a feature value and the innovation stage of the enterprise as a label value;

[0117] A second training sample set is obtained for a second task, wherein the second task is to predict the capability index of the best matching object of the enterprise at its innovation stage according to the innovation operation data of the enterprise; each training sample in the second training sample set uses the innovation operation data of the enterprise as a feature value and uses the capability index of the best matching object of the enterprise at its innovation stage as a label value;

[0118] The initial multi-task model is supervisedly trained using the first training sample set and the second training sample set to obtain a trained multi-task model.

[0119] In one implementation, the model training module 560 is specifically configured to divide the first training sample set and the second training sample set into a plurality of groups, each group comprising a plurality of training samples;

[0120] Select any group of training samples in the first training sample set and any group of training samples in the second training sample set to combine, to obtain a plurality of groups of combined training samples;

[0121] The multi-task model is supervisedly trained using each group of combined training samples as training samples for each iteration, and the loss function of the multi-task model includes the losses of the first task and the second task.

[0122] In one embodiment, the model training module 560 is further used to pre-train different capability indicator models according to different innovation stages, and store the trained capability indicator models, wherein the storage index of the capability indicator model is the innovation stage identifier corresponding to the capability indicator model;

[0123] The indicator model acquisition module 520 is specifically used to acquire the target capability indicator model according to the innovation stage identifier of the target innovation stage.

[0124] In one embodiment, the model training module 560 is specifically configured to perform the following steps for each innovation stage:

[0125] Obtaining auxiliary innovation capability data corresponding to the innovation stage of multiple enterprises, as well as auxiliary innovation capability scores at the innovation stage;

[0126] The acquired data is used to construct a third training sample set, wherein in each training sample, the auxiliary innovation capability data of the enterprise at the innovation stage is used as a feature value, and the auxiliary innovation capability score of the enterprise at the innovation stage is used as a label value;

[0127] Training the deep learning model using the constructed third training sample set;

[0128] The output layer of the trained deep learning model is deleted, and the input layer and hidden layer of the deep learning model are retained to obtain a capability indicator model corresponding to the innovation stage.

[0129] In one embodiment, the model training module 560 is specifically used to obtain the full-dimensional auxiliary innovation capability data of the best matching object corresponding to any enterprise after obtaining the innovation operation data of the enterprise;

[0130] Screening the auxiliary innovation capability data of all dimensions of the best matching object to obtain the auxiliary innovation capability data of the best matching object at the innovation stage of the enterprise;

[0131] Inputting the screened auxiliary innovation capability data into the capability index model corresponding to the innovation stage to obtain the capability index of the best matching object;

[0132] After executing the above steps for the best matching objects corresponding to all enterprises in the second training sample set, all label values ​​in the second training sample set are obtained.

[0133] In one implementation, the matching result acquisition module 540 is specifically used to determine the similarity between the plurality of capability indicators and the capability indicator of the best matching object;

[0134] The matching object determination module 550 is specifically configured to determine the candidate matching object having the highest similarity with the capability indicator of the best matching object as the target matching object.

[0135] In one embodiment, the model training module 560 is further used to pre-train different matching object prediction models according to different innovation stages;

[0136] The capability indicator determination module 530 is further used to input the innovation operation data of the target enterprise into a pre-trained matching object prediction model corresponding to the next innovation stage of the target innovation stage, and obtain the capability indicator of the best matching object corresponding to the target enterprise in the next innovation stage;

[0137] The indicator model acquisition module 520 is further used to acquire a capability indicator model corresponding to the next innovation stage according to the next innovation stage;

[0138] The capability indicator determination module 530 is further used to input the auxiliary innovation capability data of the plurality of candidate matching objects corresponding to the next innovation stage into the capability indicator model of the next innovation stage, so as to obtain the plurality of capability indicators corresponding to the plurality of candidate matching objects respectively;

[0139] The matching result acquisition module 540 is further used to use the plurality of capability indicators to match the capability indicators of the best matching object corresponding to the next innovation stage to obtain a second matching result;

[0140] The matching object determination module 550 is specifically configured to integrate the first matching result and the second matching result to select a target matching object of the target enterprise from the plurality of candidate matching objects.

[0141] In one embodiment, the matching object determination module 550 is specifically used to determine, for the target innovation stage, a first similarity of capability indicators between any candidate matching object and the best matching object of the target enterprise at the target innovation stage according to the first matching result;

[0142] Determine, based on the second matching result, a second similarity of the capability indicators of any candidate matching object and the best matching object of the target enterprise in the next innovation stage for the next innovation stage;

[0143] The first similarity and the second similarity are combined to select a target matching object of the target enterprise from the plurality of candidate matching objects.

[0144] The solutions in the embodiments of the present application can be implemented using various computer languages, for example, the object-oriented programming language Java and the interpreted scripting language JavaScript.

[0145] In addition, an embodiment of the present invention further provides an electronic device, including a bus, a transceiver, a memory, a processor, and a computer program stored in the memory and executable on the processor. The transceiver, the memory, and the processor are respectively connected via a bus. When the computer program is executed by the processor, each process of each embodiment of the above-mentioned matching object determination method is implemented, and the same technical effect can be achieved. To avoid repetition, it will not be described here.

[0146] For details, see Figure 6 As shown, the electronic device includes a bus 1110 , a processor 1120 , a transceiver 1130 , a bus interface 1140 , a memory 1150 , and a user interface 1160 .

[0147] In the embodiment of the present invention, the electronic device further includes: a computer program stored in the memory 1150 and executable on the processor 1120 , and the computer program implements the above-mentioned matching object determination method when executed by the processor 1120 .

[0148] The transceiver 1130 is configured to receive and send data under the control of the processor 1120 .

[0149] In an embodiment of the present invention, the bus architecture (represented by bus 1110), bus 1110 may include any number of interconnected buses and bridges, and bus 1110 connects various circuits including one or more processors represented by processor 1120 and a memory represented by memory 1150.

[0150] Bus 1110 represents one or more of any of several types of bus structures, including a memory bus and memory controller, a peripheral bus, an Accelerate Graphical Port (AGP), a processor, or a local bus using any of a variety of bus architectures. By way of example and not limitation, such architectures include: Industry Standard Architecture (ISA) bus, Micro Channel Architecture (MCA) bus, Enhanced ISA (EISA) bus, Video Electronics Standards Association (VESA), Peripheral Component Interconnect (PCI) bus.

[0151] The processor 1120 may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method embodiment may be completed by an integrated logic circuit of hardware in the processor or an instruction in the form of software. The above processors include: a general-purpose processor, a central processing unit (CPU), a network processor (NP), a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA), a complex programmable logic device (CPLD), a programmable logic array (PLA), a microcontroller unit (MCU) or other programmable logic devices, discrete gates, transistor logic devices, discrete hardware components. The methods, steps and logic block diagrams disclosed in the embodiments of the present invention may be implemented or executed. For example, the processor may be a single-core processor or a multi-core processor, and the processor may be integrated into a single chip or located in multiple different chips.

[0152] Processor 1120 can be a microprocessor or any conventional processor. The method steps disclosed in conjunction with the embodiments of the present invention can be directly executed by a hardware decoding processor, or by a combination of hardware and software modules in the decoding processor. The software module can be located in a readable storage medium known in the art, such as a random access memory (RAM), a flash memory (FlashMemory), a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), a register, etc. The readable storage medium is located in a memory, and the processor reads the information in the memory and completes the steps of the above method in conjunction with its hardware.

[0153] The bus 1110 may also connect various other circuits such as peripheral devices, voltage regulators or power management circuits, and the bus interface 1140 provides an interface between the bus 1110 and the transceiver 1130, which are well known in the art. Therefore, the embodiments of the present invention will not be further described.

[0154] The transceiver 1130 may be one element or multiple elements, such as multiple receivers and transmitters, and provides a unit for communicating with various other devices on a transmission medium. For example, the transceiver 1130 receives external data from other devices, and the transceiver 1130 is used to send data processed by the processor 1120 to other devices. Depending on the nature of the computer system, a user interface 1160 may also be provided, such as a touch screen, a physical keyboard, a display, a mouse, a speaker, a microphone, a trackball, a joystick, and a stylus.

[0155] It should be understood that in an embodiment of the present invention, the memory 1150 may further include a memory remotely arranged relative to the processor 1120, and these remotely arranged memories may be connected to the server through a network. One or more parts of the above-mentioned network may be an ad hoc network, an intranet, an extranet, a virtual private network (VPN), a local area network (LAN), a wireless local area network (WLAN), a wide area network (WAN), a wireless wide area network (WWAN), a metropolitan area network (MAN), the Internet, a public switched telephone network (PSTN), a plain old telephone service network (POTS), a cellular telephone network, a wireless network, a wireless fidelity (Wi-Fi) network, and a combination of two or more of the above-mentioned networks. For example, the cellular telephone network and the wireless network can be a Global System for Mobile Communications (GSM) system, a Code Division Multiple Access (CDMA) system, a Worldwide Interoperability for Microwave Access (WiMAX) system, a General Packet Radio Service (GPRS) system, a Wideband Code Division Multiple Access (WCDMA) system, a Long Term Evolution (LTE) system, a LTE Frequency Division Duplex (FDD) system, a LTE Time Division Duplex (TDD) system, an Advanced Long Term Evolution (LTE-A) system, a Universal Mobile Telecommunications (UMTS) system, an Enhanced Mobile Broadband (Enhance Mobile Broadband, eMBB) system, a massive Machine Type of Communication (mMTC) system, an Ultra-Reliable Low Latency Communications (UltraReliable Low Latency Communications, uRLLC) system, etc.

[0156] It should be understood that the memory 1150 in the embodiment of the present invention may be a volatile memory or a non-volatile memory, or may include both a volatile memory and a non-volatile memory. Among them, the non-volatile memory includes: a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory.

[0157] Volatile memory includes: Random Access Memory (RAM), which is used as an external cache. By way of example but not limitation, many forms of RAM are available, such as: Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), Double Data Rate Synchronous Dynamic Random Access Memory (DDRSDRAM), Enhanced Synchronous Dynamic Random Access Memory (ESDRAM), Synchronous Link Dynamic Random Access Memory (SLDRAM) and Direct Rambus RAM (DRRAM). The memory 1150 of the electronic device described in the embodiment of the present invention includes but is not limited to the above and any other suitable types of memory.

[0158] In the embodiment of the present invention, the memory 1150 stores the following elements of the operating system 1151 and the application program 1152: executable modules, data structures, or subsets thereof, or extended sets thereof.

[0159] Specifically, the operating system 1151 includes various system programs, such as a framework layer, a core library layer, a driver layer, etc., which are used to implement various basic services and process hardware-based tasks. The application 1152 includes various applications, such as a media player (Media Player) and a browser (Browser), which are used to implement various application services. The program for implementing the method of the embodiment of the present invention may be included in the application 1152. The application 1152 includes applets, objects, components, logic, data structures, and other computer system executable instructions that perform specific tasks or implement specific abstract data types.

[0160] In addition, an embodiment of the present invention also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the various processes of the various embodiments in the above-mentioned matching object determination method are implemented, and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.

[0161] Computer readable storage media include: permanent and non-permanent, removable and non-removable media, which are tangible devices that can retain and store instructions for use by instruction execution devices. Computer readable storage media include: electronic storage devices, magnetic storage devices, optical storage devices, electromagnetic storage devices, semiconductor storage devices, and any suitable combination of the above. Computer readable storage media include: phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), non-volatile random access memory (NVRAM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassette storage, magnetic tape disk storage or other magnetic storage devices, memory sticks, mechanical encoding devices (such as punched cards or raised structures in grooves with instructions recorded thereon) or any other non-transmission medium that can be used to store information that can be accessed by a computing device. As defined in the embodiments of the present invention, computer-readable storage media do not include temporary signals themselves, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through waveguides or other transmission media (such as light pulses passing through fiber optic cables), or electrical signals transmitted through wires.

[0162] The computer program code contained in the above-mentioned computer-readable storage medium can be transmitted using any appropriate medium, including: wireless, wire, optical cable, radio frequency (RF) or any suitable combination thereof.

[0163] The computer program code for performing the operation of the embodiments of the present invention can be written in assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-related instructions, microcode, firmware instructions, state setting data, integrated circuit configuration data, or in one or more programming languages ​​or a combination thereof, wherein the programming language includes an object-oriented programming language, such as Java, Smalltalk, C++, and also includes a conventional procedural programming language, such as C language or a similar programming language. The computer program code can be executed completely on the user's computer, partially on the user's computer, as an independent software package, partially on the user's computer, partially on a remote computer, and completely on a remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer or to an external computer through any type of network, including a local area network (LAN) or a wide area network (WAN).

[0164] The embodiment of the present invention further provides a computer program, including one or more computer instructions. When the computer instructions are loaded and executed on a processor, the process or function described in the embodiment of the present application is generated in whole or in part.

[0165] Although the preferred embodiments of the present application have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the present application.

[0166] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is also intended to include these modifications and variations.

Claims

1. A matching object determination method, characterized in that: include: Inputting the target enterprise's innovation operation data into a pre-trained multi-task model to obtain the target innovation stage to which the target enterprise currently belongs and the capability index of the best matching object corresponding to the target innovation stage of the target enterprise; Obtaining a target capability indicator model corresponding to the target innovation stage; Inputting auxiliary innovation capability data of a plurality of candidate matching objects corresponding to the target innovation stage into the target capability indicator model to obtain a plurality of capability indicators corresponding to the plurality of candidate matching objects respectively; Using the plurality of capability indicators to match the capability indicators of the best matching object respectively to obtain a first matching result; A target matching object of the target enterprise is screened out from the plurality of candidate matching objects according to the first matching result.

2. The method according to claim 1, characterized in that: The training process of the multi-task model includes: A first training sample set is obtained for a first task, wherein the first task is to predict the innovation stage of an enterprise based on the innovation operation data of the enterprise; each training sample in the first training sample set uses the innovation operation data of the enterprise as a feature value and uses the innovation stage of the enterprise as a label value; A second training sample set is obtained for a second task, wherein the second task is to predict the capability index of the best matching object of the enterprise at its innovation stage according to the innovation operation data of the enterprise; each training sample in the second training sample set uses the innovation operation data of the enterprise as a feature value and uses the capability index of the best matching object of the enterprise at its innovation stage as a label value; The initial multi-task model is supervisedly trained using the first training sample set and the second training sample set to obtain a trained multi-task model.

3. The method according to claim 2, characterized in that The method of using the first training sample set and the second training sample set to perform supervised training on the initial multi-task model to obtain a trained multi-task model includes: Dividing the first training sample set and the second training sample set into a plurality of groups, each group comprising a plurality of training samples; Select any group of training samples in the first training sample set and any group of training samples in the second training sample set to combine, to obtain a plurality of groups of combined training samples; The multi-task model is supervisedly trained using each group of combined training samples as training samples for each iteration, and the loss function of the multi-task model includes the losses of the first task and the second task.

4. The method according to claim 1, characterized in that: Pre-train different capability indicator models according to different innovation stages, and store the trained capability indicator models, wherein the storage index of the capability indicator model is the innovation stage identifier corresponding to the model; The obtaining of the target capability indicator model corresponding to the target innovation stage includes: The target capability indicator model is obtained according to the innovation stage identifier of the target innovation stage.

5. The method according to claim 4, characterized in that The training of different capability indicator models according to different innovation stages includes: Perform the following steps for each of the innovation phases: Obtaining auxiliary innovation capability data corresponding to the innovation stage of multiple enterprises, as well as auxiliary innovation capability scores at the innovation stage; The acquired data is used to construct a third training sample set, wherein in each training sample, the auxiliary innovation capability data of the enterprise at the innovation stage is used as a feature value, and the auxiliary innovation capability score of the enterprise at the innovation stage is used as a label value; Training the deep learning model using the constructed third training sample set; The output layer of the trained deep learning model is deleted, and the input layer and hidden layer of the deep learning model are retained to obtain a capability indicator model corresponding to the innovation stage.

6. The method according to claim 5, characterized in that The step of obtaining a second training sample set for the second task includes: After obtaining the innovation operation data of any enterprise, obtain the auxiliary innovation capability data of all dimensions corresponding to the best matching object of the enterprise; Screening the auxiliary innovation capability data of all dimensions of the best matching object to obtain the auxiliary innovation capability data of the best matching object at the innovation stage of the enterprise; Inputting the screened auxiliary innovation capability data into the capability index model corresponding to the innovation stage to obtain the capability index of the best matching object; After executing the above steps for the best matching objects corresponding to all enterprises in the second training sample set, all label values ​​in the second training sample set are obtained.

7. The method according to claim 1, characterized in that The obtaining of a first matching result by matching the plurality of capability indicators with the capability indicators of the target matching object respectively includes: Determining the similarity between the plurality of capability indicators and the capability indicator of the best matching object; The step of selecting a target matching object of the target enterprise from the plurality of candidate matching objects according to the first matching result includes: The candidate matching object having the highest similarity with the capability index of the best matching object is determined as the target matching object.

8. The method according to claim 1, characterized in that Pre-training different matching object prediction models according to different innovation stages; the method also includes: Inputting the innovation operation data of the target enterprise into a pre-trained matching object prediction model corresponding to the next innovation stage of the target innovation stage, and obtaining the capability index of the best matching object corresponding to the target enterprise in the next innovation stage; Obtaining a capability indicator model corresponding to the next innovation stage; Inputting the auxiliary innovation capability data of the plurality of candidate matching objects corresponding to the next innovation stage into the capability indicator model of the next innovation stage, to obtain a plurality of capability indicators respectively corresponding to the plurality of candidate matching objects; Using the plurality of capability indicators to match the capability indicators of the best matching object corresponding to the next innovation stage, respectively, to obtain a second matching result; The step of selecting a target matching object of the target enterprise from the plurality of candidate matching objects according to the first matching result includes: The first matching result and the second matching result are combined to screen out a target matching object of the target enterprise from the plurality of candidate matching objects.

9. The method according to claim 8, characterized in that The step of combining the first matching result and the second matching result to select a target matching object of the target enterprise from the plurality of candidate matching objects includes: Determine, based on the first matching result, a first similarity of capability indicators of any candidate matching object and the best matching object of the target enterprise at the target innovation stage for the target innovation stage; Determine, based on the second matching result, a second similarity of the capability indicators of any candidate matching object and the best matching object of the target enterprise in the next innovation stage for the next innovation stage; The first similarity and the second similarity are combined to select a target matching object of the target enterprise from the plurality of candidate matching objects.

10. A matching object determination device, characterized in that: include: A multi-task model calling module is used to input the innovation operation data of the target enterprise into the pre-trained multi-task model to obtain the target innovation stage to which the target enterprise currently belongs and the capability index of the best matching object corresponding to the target innovation stage of the target enterprise; An indicator model acquisition module, used to acquire a target capability indicator model corresponding to the target innovation stage according to the target innovation stage; A capability indicator determination module, used to input auxiliary innovation capability data of a plurality of candidate matching objects corresponding to the target innovation stage into the target capability indicator model, and obtain a plurality of capability indicators corresponding to the plurality of candidate matching objects respectively; A matching result acquisition module, used to respectively match the several capability indicators with the capability indicators of the best matching object to obtain a first matching result; The matching object determination module is used to screen out a target matching object of the target enterprise from the plurality of candidate matching objects according to the first matching result.

11. An electronic device comprising a processor and a memory, wherein the memory stores a computer program, characterized in that: The processor executes the computer program stored in the memory to implement the steps in the method according to any one of claims 1 to 9.

12. A computer program comprising computer instructions, characterized in that When the computer instructions are executed by a processor, the steps in the method according to any one of claims 1 to 9 are implemented.

13. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps in the method according to any one of claims 1 to 9 are implemented.

Images