Target enterprise data and target personnel data matching method and system

Abstract

The embodiment of the invention provides a target enterprise data and target personnel data matching method and system, and relates to the technical field of data processing. The method comprises the steps of screening out a target enterprise data group based on historical data; to-be-distributed enterprise data and candidate person list data are screened out based on real-time data, and the real-time data comprise a target enterprise data group; calculating a first matching degree value of the to-be-allocated enterprise data and the candidate person list data; determining the to-be-allocated enterprise data and the candidate person list data of which the first matching degree values are greater than or equal to a preset threshold as target enterprise data and target person data; the specified target personnel data are in butt joint with the target enterprise data. Through one embodiment of the invention, the problem that the success rate of matching the customer manager data with the to-be-served enterprise data is low in the prior art is at least solved, and the effect of improving the success rate of matching the customer manager data with the to-be-served enterprise data is achieved.

Description

Technical Field

[0001] This application relates to the field of data processing technology, and more specifically, to a method and system for matching target enterprise data with target personnel data. Background Technology

[0002] In the current economic environment, small and medium-sized enterprises (SMEs) have become an important part of the national economy, and their financial service needs are becoming increasingly diverse and personalized. As the core providers of financial services to SMEs, commercial banks have widely adopted intelligent business opportunity management systems to improve referral efficiency and expand business coverage. These systems are used to identify potential customers, push business opportunities, and connect relevant service personnel.

[0003] The core operational logic of relevant intelligent business opportunity management systems largely revolves around the construction of business opportunity pools and opportunity delivery. These systems typically build business opportunity pools based on static data such as enterprise registration information and basic financial indicators, combined with the bank's internal product system. Then, through a standardized rule engine, they push business opportunity information to recommendation service personnel (such as account managers) in a broadcast-style, all-encompassing manner. In this model, the system focuses more on the screening and listing of business opportunity information, lacking a comprehensive consideration of multi-dimensional data such as the enterprise's deep needs, operational characteristics, and risk attributes. Furthermore, it fails to accurately characterize and quantify the individual characteristics of account managers, such as their professional capabilities, business experience, client resources, and historical service performance.

[0004] Because the relevant systems lack a deep correlation analysis mechanism between enterprise data and account manager data, lead to a lack of targeted and adaptive business opportunity delivery. Account managers often need to manually sift through a large amount of homogeneous business opportunity information to identify enterprise clients that match their service capabilities. This not only increases their workload and reduces service response efficiency but also easily leads to a "capability mismatch." This means experienced account managers are assigned to enterprises with low-complexity needs, while novice account managers are assigned to enterprises with high-difficulty needs, or the account manager's industry service experience does not match the enterprise's field. This extensive matching model prevents account managers from efficiently focusing on truly suitable enterprise clients, ultimately resulting in a low success rate of matching account manager data with the data of enterprises to be served. This severely impacts the conversion rate of financial services and the interaction experience for both parties. Summary of the Invention

[0005] This application provides a method and system for matching target enterprise data with target personnel data, which at least solves the problem of low success rate in matching customer manager data with enterprise data to be served in related technologies.

[0006] According to one embodiment of this application, a method for matching target enterprise data with target personnel data is provided, including:

[0007] Target enterprise data groups were selected based on historical data;

[0008] The data of enterprises to be assigned and the list of candidates are filtered based on real-time data, wherein the real-time data includes the target enterprise data group;

[0009] Calculate the first matching degree value between the enterprise data to be assigned and the candidate list data;

[0010] The data of enterprises to be assigned and the list of candidates whose first matching degree value is greater than or equal to the preset threshold are identified as the target enterprise data and the target personnel data.

[0011] Specify that the target personnel data is linked to the target enterprise data.

[0012] In one implementation, the target enterprise data group is filtered based on historical data, including:

[0013] Based on the company's historical operating data, new operating data for a future preset time period is determined, and based on the new operating data and the company's credit score in the system, the company's operating score is determined.

[0014] The enterprise intention score is determined based on historical enterprise-product pairing data.

[0015] Based on the enterprise's historical risk data, new risk data for a future preset time period is determined, and the enterprise's risk score is determined based on the new risk data.

[0016] The target enterprise data group is selected based on the enterprise operation score, the enterprise intention score, and the enterprise risk score. The historical data includes: the enterprise's historical operation data, the enterprise's historical pairing data with products, and the enterprise's historical risk data.

[0017] In one implementation, determining the enterprise intention score based on historical enterprise-product pairing data includes:

[0018] The company's intent score is determined based on industry data and the company's needs data.

[0019] The second matching degree value between the enterprise and the product is determined based on enterprise data and product data;

[0020] The enterprise intention score is determined based on the enterprise intention score and the second matching score.

[0021] In one implementation, it further includes:

[0022] The revenue value is determined based on the value data of the target enterprise data, the business capability data of the target personnel data, the historical feedback data of the target personnel data, and the business record data of the target personnel data.

[0023] The revenue value is used as an incentive value for the target personnel data.

[0024] In one implementation, it further includes:

[0025] After filtering out the data of enterprises to be assigned and the list of candidates, a window period with a preset time period is allocated to the data of enterprises to be assigned, so that the data of the target personnel can be connected to the data of the target enterprises within the window period;

[0026] If the target personnel data fails to be successfully linked to the target enterprise data within the specified window period, the target enterprise data will be reallocated in the next window period.

[0027] In one implementation, determining the revenue value based on the value data of the target enterprise data, the business capability data of the target personnel data, the historical feedback data of the target personnel data, and the business record data of the target personnel data includes: determining the revenue value based on a first algorithm, wherein the first algorithm includes:

[0028] ;

[0029] in, The fundamental value data for the target enterprise; Value scoring data for target company data; The highest historical task value corresponding to the target company's data; The matching score between target company data and target personnel data; The average difficulty of completing tasks based on the target personnel's historical data; This refers to the number of tasks completed by the target personnel within a preset time period prior to the current moment; This refers to the historical average number of transactions completed by the target personnel within a preset time period.

[0030] In one implementation, determining a company's intention score based on historical company-product pairing data includes: determining the company's intention score based on a second algorithm, wherein the second algorithm includes:

[0031] ;

[0032] in, Weighting of the enterprise intent score; The weight of the second matching degree value; A function for scoring enterprise intent; The coefficient weights of the enterprise intent scoring function; This is the function for the second matching degree value; The coefficient weights are the values ​​of the second matching degree function.

[0033] According to another embodiment of this application, a target enterprise data and target personnel data matching system is provided, comprising:

[0034] The evaluation module is used to filter out target enterprise data groups based on historical data;

[0035] A matching module is used to filter out enterprise data to be assigned and candidate personnel list data based on real-time data, wherein the real-time data includes the target enterprise data group; and to calculate a first matching degree value between the enterprise data to be assigned and the candidate personnel list data; and to determine the enterprise data to be assigned and candidate personnel list data with the first matching degree value greater than or equal to a preset threshold as target enterprise data and target personnel data; and to assign the target personnel data to the target enterprise data.

[0036] According to yet another embodiment of this application, a computer-readable storage medium is also provided, wherein a computer program is stored therein, and the computer program is configured to perform the steps in any of the above method embodiments when it is run.

[0037] According to yet another embodiment of this application, an electronic device is also provided, including a memory and a processor, wherein the memory stores a computer program and the processor is configured to run the computer program to perform the steps in any of the above method embodiments.

[0038] According to yet another embodiment of this application, a computer program product is also provided, including computer instructions that, when executed by a processor, implement the steps in any of the above method embodiments.

[0039] In one embodiment of this application, firstly, a target enterprise data group is filtered using historical data, overcoming the limitations of relying solely on static data and uncovering more comprehensive characteristics of enterprises from a historical perspective to initially identify a range of valuable enterprises. Then, based on real-time data containing this target enterprise data group, the data of enterprises to be assigned and the list of candidates are filtered, balancing the cumulative nature of historical data with the dynamic nature of real-time data to ensure the timeliness and relevance of the matching targets. Subsequently, a first matching degree value is calculated between the data of enterprises to be assigned and the list of candidates, establishing a quantitative correlation analysis mechanism between enterprise data and personnel data, replacing the crude, indiscriminate push. Finally, those meeting the first matching degree value are identified as target enterprise data and target personnel data and assigned for connection, achieving precise matching to avoid "capability mismatch." Therefore, this effectively solves the problem of low matching success rate between account manager data and enterprise data to be served in related technologies, thereby improving the matching success rate between account manager data and enterprise data to be served. Attached Figure Description

[0040] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:

[0041] Figure 1 This is a hardware structure block diagram of the target enterprise data and target personnel data matching method according to an embodiment of this application;

[0042] Figure 2 This is a flowchart of a method for matching target enterprise data and target personnel data according to an embodiment of this application;

[0043] Figure 3 This is a flowchart of a method for filtering target enterprise data groups based on historical data according to an embodiment of this application;

[0044] Figure 4 This is a flowchart of a method for determining a company's intention score based on historical pairing data of companies and products, according to an embodiment of this application.

[0045] Figure 5 This is a flowchart of a method for configuring incentive value data for target personnel data according to an embodiment of this application;

[0046] Figure 6 This is a flowchart of a method for allocating a preset time window period for enterprise data to be allocated according to an embodiment of this application;

[0047] Figure 7 This is a structural block diagram of a target enterprise data and target personnel data matching system according to an embodiment of this application;

[0048] Figure 8This is a structural block diagram of another target enterprise data and target personnel data matching system according to an embodiment of this application;

[0049] Figure 9 This is a structural block diagram of another target enterprise data and target personnel data matching system according to an embodiment of this application. Detailed Implementation

[0050] The present application will be described in detail below with reference to the accompanying drawings and embodiments. It should be noted that, unless otherwise specified, the embodiments and features described in the embodiments of the present application can be combined with each other.

[0051] It should be noted that the terms "first," "second," etc., in the specification, claims, and drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence.

[0052] The collection, storage, use, processing, transmission, provision, and disclosure of financial data or user data involved in the technical solution of this application all comply with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0053] It should be noted that in the embodiments of this application, certain existing solutions in the industry, such as software, components, and models, may be mentioned. These should be considered as exemplary, and their purpose is merely to illustrate the feasibility of implementing the technical solution of this application. However, it does not mean that the applicant has already used or necessarily used such a solution. Of course, in the embodiments of this application, bank business data is often used as an example for explanation and illustration, but it is not limited to bank business data.

[0054] The methods and embodiments provided in this application can be executed on a mobile terminal, a computer terminal, or a similar computing device. Taking running on a computer terminal as an example, Figure 1 This is a hardware structure block diagram of the target enterprise data and target personnel data matching method according to an embodiment of this application, such as... Figure 1 As shown, a hardware board may include one or more ( Figure 1 Only one is shown in the diagram. A processor 12 (which may include, but is not limited to, a microprocessor MCU or programmable logic device, etc.) and a memory 14 for storing data are also shown. The computer terminal may further include a transmission device 16 for communication functions and an input / output device 18. Those skilled in the art will understand that... Figure 1 The structure shown is for illustrative purposes only and does not limit the structure of the computer terminal described above. For example, the computer terminal may also include components that are more complex than those described above. Figure 1 The more or fewer components shown, or having the same Figure 1 The different configurations shown.

[0055] The memory 14 can be used to store computer programs, such as application software programs and modules, like the computer program corresponding to the target enterprise data and target personnel data matching method in this embodiment. The processor 12 executes various functional applications and implements the above-described methods by running the computer programs stored in the memory 14. The memory 14 may include high-speed random access memory and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 14 may further include memory remotely located relative to the processor 12, and these remote memories can be connected to a computer terminal via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0056] The transmission device 16 is used to receive or send data via a network. Specific examples of the network described above may include a wireless network provided by a telecommunications provider. In one example, the transmission device 16 includes a Network Interface Controller (NIC), which can connect to other network devices via a gateway to communicate with the Internet. In another example, the transmission device 16 may be a Radio Frequency (RF) module used for wireless communication with the Internet.

[0057] This application provides a method for matching target enterprise data with target personnel data. Figure 2 This is a flowchart of a method for matching target enterprise data and target personnel data according to an embodiment of this application, such as... Figure 2 As shown, the process includes:

[0058] Step S201: Filter out the target enterprise data group based on historical data;

[0059] In one exemplary implementation, for example, a commercial bank collects historical operating data (such as revenue fluctuations and deposit flows) of SMEs over the past three years, historical pairing records with bank products (such as whether they have applied for credit loans or corporate financial management), and historical risk data (such as whether there are overdue repayment records). Based on this data, a group of enterprises with potential cooperation value (high-quality customers) is initially screened. This step breaks through the limitations of relying solely on static data such as the registered capital of enterprises, and mines more comprehensive operating, demand, and risk characteristics of enterprises from a historical perspective, successfully identifying a range of enterprises with cooperation potential and laying the foundation for subsequent precise matching.

[0060] Step S202: Based on real-time data, filter out the data of enterprises to be assigned and the list of candidates, wherein the real-time data includes the target enterprise data group;

[0061] In one exemplary implementation, for example, the bank system captures the latest developments of target companies in real time (such as recent large capital inflows and newly released financing demand announcements), while simultaneously integrating information on currently employed account managers with reasonable account management workloads, to filter out company data to be assigned and candidate manager lists. This step balances the accumulation of historical data with the dynamism of real-time data, ensuring that the data on companies and account managers to be matched is timely and relevant, avoiding invalid matching due to information lag.

[0062] Step S203: Calculate the first matching degree value between the enterprise data to be assigned and the candidate list data;

[0063] In one exemplary implementation, a quantitative evaluation model is constructed, which combines the industry attributes and demand type of the enterprise to be allocated (such as whether it needs a science and technology innovation loan) with the candidate's industry service experience, product expertise, and historical recommendation success rate, among other dimensions, and assigns weights accordingly. The evaluation model then calculates a comprehensive matching score between the two. For example, a small and medium-sized manufacturing enterprise has a need for equipment upgrade loans, and candidate account manager C is skilled in manufacturing services and has a high success rate in recommending equipment loans. The evaluation model calculates a matching score between the two. This step establishes a quantitative correlation analysis mechanism between enterprise and personnel data, replacing the previous indiscriminate broadcasting and extensive push model, and providing clear data support for the matching results.

[0064] Step S204: The data of enterprises to be assigned and the list of candidates whose first matching degree value is greater than or equal to the preset threshold are determined as the data of target enterprises and the data of target personnel.

[0065] In one exemplary implementation, for example, a preset threshold of 75 points is used to filter the calculated first matching scores, identifying combinations with scores greater than or equal to 75 points as target company data and target personnel data. For instance, an IT company and account manager D have a matching score of 82 points, reaching the preset threshold; therefore, the IT company is identified as target company data, and account manager D is identified as target personnel data. This step accurately filters out highly compatible company and account manager combinations, effectively avoiding capability mismatches such as "novices connecting with companies with highly complex needs," and improving the effectiveness of subsequent connections.

[0066] Step S205: Specify the target personnel data and connect it to the target enterprise data.

[0067] In one exemplary implementation, for example, detailed information about the target company (an IT company) is sent to the target individual (such as account manager D) through the bank's internal system. This information includes the company's needs, operating status, and historical cooperation records. Simultaneously, dedicated access permissions are assigned, and the company is informed of the account manager's contact information and service scope. This step clarifies the responsible party for the connection, ensuring the matching results are effectively implemented and avoiding issues such as no follow-up after the opportunity is pushed out or multiple people making duplicate follow-ups, significantly improving connection efficiency.

[0068] Through steps S201 to S205, the process begins by filtering the target enterprise data group using historical data. This overcomes the limitations of relying solely on static data, allowing for the extraction of more comprehensive enterprise characteristics from a historical perspective to initially identify valuable enterprises. Next, real-time data containing this target enterprise data group is used to filter the data of enterprises to be assigned and the list of candidates, balancing the cumulative nature of historical data with the dynamism of real-time data to ensure the timeliness and relevance of the matching targets. Subsequently, the first matching degree value between the data of enterprises to be assigned and the list of candidates is calculated, establishing a quantitative correlation analysis mechanism between enterprise data and personnel data, replacing the crude, indiscriminate push approach. Finally, those meeting the first matching degree value are identified as target enterprise data and target personnel data and assigned for precise matching to avoid "capability mismatch." Therefore, this effectively solves the problem of low matching success rate between account manager data and enterprise data to be served in related technologies, thereby improving the matching success rate between account manager data and enterprise data to be served.

[0069] Figure 3 This is a flowchart of a method for filtering target enterprise data groups based on historical data according to an embodiment of this application, such as... Figure 3 As shown, in one implementation, filtering out a target enterprise data group based on historical data includes:

[0070] Step S301: Determine new operating data for a future preset time period based on the company's historical operating data, and determine the company's operating score based on the new operating data and the company's credit score in the system;

[0071] In one exemplary implementation, for example, a science and technology innovation SME is selected, and its historical operating data, such as revenue growth rate (average annual growth of 15%), R&D investment ratio (average annual growth of 8%), and average monthly deposit balance over the past two years, are collected. Linear trend analysis is used to predict revenue (expected growth of 12%) and R&D investment ratio (maintained at 7.8%) for the next six months. Then, the company's credit score (e.g., "Longxin Business Score" of 85) is retrieved from the system. The predicted operating data (reflecting revenue growth potential) and the credit score are weighted at a 4:6 ratio to calculate the company's operating score of 83. Therefore, this approach overcomes the limitations of relying solely on current registered capital and static revenue data. By predicting future operating trends through historical data, it can more accurately reflect the company's operating potential, providing a "growth potential" dimension reference for subsequent screening and avoiding overlooking promising but currently small-scale SMEs.

[0072] Step S302: Determine the enterprise intention score based on historical enterprise and product pairing data;

[0073] In one exemplary implementation, for example, for a small or medium-sized manufacturing enterprise (SME), historical pairing data of its interactions with bank products over the past year is extracted. This includes instances where the enterprise has taken out two equipment-secured loans, regularly purchases short-term wealth management products monthly, and has consulted about supply chain finance services. By statistically analyzing the frequency of pairings with "financing" and "wealth management" products (60% for financing and 40% for wealth management), the enterprise's demand for production-related financing products is determined, and a corporate intention score of 78 is calculated (the clearer the demand and the higher the fit with bank products, the higher the score). Therefore, mining enterprise needs based on real historical product cooperation records can accurately capture enterprise product preferences, ensuring that subsequently selected enterprises have clear service needs, reducing the probability of "enterprises without actual cooperation intentions" entering the data set, and enhancing the effective value of the data set.

[0074] Step S303: Determine new risk data for a future preset time period based on the company's historical risk data, and determine the company's risk score based on the new risk data;

[0075] In one exemplary implementation, for example, historical risk data for a small or medium-sized trading company (SME) is collected over the past three years. If there are no overdue payment records, accounts receivable turnover is stable at 6 times per year, and the fluctuation range of cash inflows and outflows in the past 12 months is less than 10%, stability analysis of the historical risk data is used to predict the probability of overdue payments in the next three months (estimated at 0.5%) and the liquidity risk level (low risk). The risk prediction result is then converted into a risk score of 92 points (the lower the risk and the lower the probability of future default, the higher the score). Therefore, by predicting the future risk status of enterprises through historical risk data, high-risk enterprises (such as those with frequent historical overdue payments and large cash flow fluctuations) can be excluded in advance. This avoids the failure of subsequent business opportunities due to enterprise risk issues, ensuring the overall quality of the target enterprise data group from a risk perspective, and reducing bank service costs and risk losses.

[0076] Step S304: Target enterprise data group is selected based on enterprise operation score, enterprise intention score, and enterprise risk score. Historical data includes: historical enterprise operation data, historical enterprise-product pairing data, and historical enterprise risk data.

[0077] In one exemplary implementation, for example, the weights of the enterprise's operational score, intention score, and risk score are each set at 1 / 3. A comprehensive score is calculated (e.g., an enterprise with an operational score of 83, an intention score of 78, and a risk score of 92 has a comprehensive score of 84.3). A comprehensive score threshold of 75 is preset. Enterprises with a comprehensive score greater than or equal to 75 are included in the target enterprise data group, while enterprises with a score below the threshold (e.g., high-risk enterprises with a comprehensive score of 68 and a risk score of only 60) are excluded. Therefore, by integrating the three core dimensions of operational potential, demand intention, and risk control, the one-sidedness of screening based on a single dimension (e.g., only looking at enterprises with good operational data but high risk, or only looking at enterprises with low risk but no demand) is avoided. Ultimately, a high-quality group of enterprises with "potential, demand, and low risk" is selected, providing a high-quality enterprise data foundation for subsequent accurate matching with account managers.

[0078] Through steps S301 to S304, in-depth mining of historical data and prediction of future trends replace the traditional, crude screening method that relies solely on business registration information and static financial indicators. The entire process considers both the growth potential of enterprises (future operating data) and the possibility of cooperation (demand intentions), while also ensuring the security of cooperation (risk control). The resulting target enterprise data set is of higher quality and more compatible with banking services. This effectively solves the problems of single-dimensional target enterprise screening and lack of foresight and comprehensiveness in related technologies, laying a solid data foundation for accurate matching in subsequent business opportunity allocation stages and improving the overall efficiency and success rate of business opportunity management.

[0079] Figure 4This is a flowchart illustrating a method for determining a company's intention score based on historical pairing data of companies and products, according to an embodiment of this application. Figure 4 As shown, in one implementation, determining the enterprise intention score based on historical enterprise-product pairing data includes:

[0080] Step S401: Determine the enterprise intent score based on the industry data and the enterprise's demand data;

[0081] In one exemplary implementation, taking a small and medium-sized enterprise (SME) in the new energy sector as an example, industry data is first collected. The new energy industry is supported by policies, and enterprises generally have a need to expand production capacity and purchase production equipment. Their core financial needs are concentrated in products such as fixed asset loans and green credit. Next, data on the enterprise's own needs is extracted. This enterprise submitted a green credit consultation form once in the past two months and has a historical record of a small equipment loan (used for upgrading a photovoltaic module production line). Based on this, industry demand matching degree (40%, the enterprise's needs overlap with the core needs of the new energy industry by 95%) and enterprise demand clarity (60%, the consultation record and historical loan prove the continuity of needs, scoring 92 points) are set. A weighted calculation yields the enterprise's intention score: 40% × 95 + 60% × 92 = 93.2 points. Therefore, by combining common industry needs with individual enterprise needs, the true cooperation intention of enterprises is accurately captured, avoiding the inclusion of enterprises with no clear needs (such as never having consulted about related products) or whose needs are unrelated to banking services (such as only requiring ordinary settlement services) in subsequent evaluations, providing a core basis for determining intention at the "intention level."

[0082] Step S402: Determine the second matching degree value between the enterprise and the product based on enterprise data and product data;

[0083] In one exemplary implementation, for example, core data of the aforementioned new energy enterprise is extracted: annual photovoltaic module production capacity of 100MW, green energy project investment of 30 million yuan in the past year, and enterprise credit rating of AA. Simultaneously, data on the bank's "Green Capacity Expansion Loan" product is retrieved, with eligibility criteria of: new energy sector enterprise, annual production capacity ≥ 50MW, green project investment ≥ 10 million yuan, and credit rating ≥ A. Matching dimensions and weights are set as follows: industry affiliation matching (20%, conforming to the new energy sector, 100 points), production capacity compliance (30%, 100MW ≥ 50MW, 100 points), green investment matching (30%, 30 million yuan ≥ 10 million yuan, 100 points), and credit rating matching (20%, AA ≥ A, 100 points). The weighted average is used to calculate the second matching score: 20%×100 + 30%×100 + 30%×100 + 20%×100 = 100 points. Therefore, by converting "whether a company is suitable for a product" into a quantifiable score, and replacing subjective judgment with objective data comparison, we can avoid mismatches where "a company has a need but does not meet the product's access requirements" (such as failing to meet credit rating standards), and provide objective "fit-match" basis for determining intent.

[0084] Step S403: Determine the enterprise intention score based on the enterprise intention score and the second matching score.

[0085] In one exemplary implementation, for example, the enterprise intent score (93.2 points) obtained in step S401 has a weight of 30%, and the second matching degree value (100 points) obtained in step S402 has a weight of 70%. The enterprise intent score can be calculated using the formula "Enterprise Intent Score = Intent Score × 30% + Second Matching Degree Value × 70%", resulting in: 93.2 × 30% + 100 × 70% = 97.96 points. Therefore, by combining the two key dimensions of "enterprise wants to cooperate" (intent score) and "enterprise can cooperate" (second matching degree), the bias of a single dimension is avoided. This excludes enterprises that are "intent but not suitable for the product" (e.g., wanting a green loan but lacking creditworthiness) and enterprises that are "suitable for the product but lack intent" (e.g., meeting the green loan conditions but only requiring settlement services), ensuring that the final score truly reflects the cooperation potential between the enterprise and the bank's products, providing a precise quantitative standard for subsequent screening.

[0086] Through steps S401 to S403, the genuine willingness to cooperate is first identified from both industry and enterprise perspectives, addressing the question of "whether the enterprise wants to cooperate." Then, through quantitative comparison of enterprise data and product conditions, the question of "whether the enterprise can cooperate" is addressed. Finally, a comprehensive score is generated through weighted integration, achieving dual verification of "willingness and suitability." This entire process is data-driven, replacing the traditional, crude judgment method that relies solely on historical matching records. It effectively avoids the problems of a single dimension of intention judgment and strong subjectivity, providing accurate and reliable scoring criteria for subsequent screening of target enterprise data groups. This further ensures that enterprises entering the opportunity allocation stage are "willing and highly suitable," laying a crucial foundation for improving the success rate of matching account managers with enterprises.

[0087] In one implementation, determining a company's intention score based on historical company-product pairing data includes: determining the company's intention score based on a second algorithm, wherein the second algorithm includes:

[0088] ;

[0089] in, Weighting of the enterprise intent score; The weight of the second matching degree value; A function for scoring enterprise intent; The coefficient weights of the enterprise intent scoring function; This is the function for the second matching degree value; The coefficient weights are the values ​​of the second matching degree function.

[0090] In one exemplary implementation, an evaluation framework is constructed with "corporate intent (reflecting willingness to cooperate)" and "product matching (reflecting feasibility of cooperation)" as its dual cores. Through a design of "layered weight control + functional quantification," it addresses the pain points of traditional corporate intent evaluations, which are often one-sided (focusing only on needs or fit), subjective and ambiguous (relying on human experience), and rigid (applying uniform standards to all situations). Overall weighting... / The internal coefficient weights of the function can be dynamically adjusted based on business priorities (e.g., prioritizing willingness when promoting new businesses, and prioritizing compatibility when controlling risks). / It can adapt to the evaluation needs of different industries and products, and at the same time, it can transform unstructured data such as industry attributes, demand frequency, and product access conditions into standardized scores, eliminating subjective errors and ensuring evaluation consistency. It also provides accurate and traceable quantitative basis for subsequent screening of target enterprise data groups based on historical data (such as setting intention score thresholds to lock in "high intention + high fit" enterprises) and business opportunity allocation. It fills the gap in traditional evaluation where "intention and fit are separated and cannot support accurate screening", and becomes a key technical support for ensuring the quality of enterprise intention evaluation and promoting the accurate allocation of overall business opportunities.

[0091] Figure 5 This is a flowchart of a method for configuring incentive value data for target personnel data according to an embodiment of this application, such as... Figure 5 As shown, in one embodiment, the method further includes:

[0092] Step S501: Determine the revenue value based on the value data of the target enterprise data, the business capability data of the target personnel data, the historical feedback data of the target personnel data, and the business record data of the target personnel data;

[0093] In one exemplary implementation, for example, taking a client manager surnamed Zhang (the target person) from a commercial bank connecting with a small and medium-sized biopharmaceutical enterprise (the target company) as an example, four types of core data are extracted first: target company value data (company assessment score of 92 points, corresponding to a potential business revenue of 100,000 yuan), Zhang's business capability data (5 years of service in the biopharmaceutical industry, 85% conversion rate of business opportunities in this field, corresponding to a capability coefficient of 1.1), Zhang's historical feedback data (customer satisfaction rate of 94% in the past year, corresponding to a feedback coefficient of 1.05), and Zhang's business record data (6 biopharmaceutical business opportunities completed in the past 3 months, higher than the team average of 4, corresponding to a record coefficient of 1.08). The calculation logic is set as follows: revenue value = potential revenue of the enterprise × capability coefficient × feedback coefficient × record coefficient. Substituting the data, the revenue value is 100,000 × 1.1 × 1.05 × 1.08 = 124,740 yuan. Therefore, by integrating enterprise value with personnel capabilities, feedback, and records of multi-dimensional data, it replaces the traditional single model that only determines revenue based on the enterprise's potential revenue. This allows the revenue value to reflect both the value of the business opportunity itself and the professionalism and past performance of the personnel, avoiding the unfair situation of "highly capable personnel receiving low-value business opportunities with low revenue" and ensuring that the revenue value calculation is more comprehensive and reasonable.

[0094] Step S502: Use the benefit value as incentive value data for the target personnel data.

[0095] In one exemplary implementation, for example, the commercial bank system directly uses the revenue value of 124,740 yuan calculated in step S501 as the incentive value for account manager Zhang. On one hand, this amount is entered into Zhang's performance system, labeled as "Special Incentive for Business Opportunities in Biopharmaceutical Companies," clearly indicating that the incentive is directly linked to this business opportunity. On the other hand, Zhang is notified through system notification of the incentive amount, calculation basis (such as industry service years, customer satisfaction, and other specific dimensions), and it is agreed that after the business opportunity is successfully secured, the incentive will be paid in two installments (60% upon signing the contract, and 40% after three months of problem-free service). Therefore, by transforming the abstract revenue value into a concrete and redeemable incentive, account managers clearly understand "the returns they can obtain from securing this business opportunity," and the incentive rules are transparent and traceable, effectively eliminating concerns that "the returns are roughly the same regardless of how much work is done," directly enhancing their initiative and sense of responsibility in engaging with target companies, and avoiding the problem of slackness in business opportunity follow-up due to unclear incentives.

[0096] Through steps S501 and S502, step S501 integrates data from both the enterprise and personnel sides to ensure that the calculation of revenue value takes into account both fairness and objectivity, thus solving the one-sidedness of traditional incentives that "emphasize business opportunities while neglecting personnel." Step S502 translates revenue value into perceptible and redeemable incentives, directly linking incentives to personnel contributions through transparent rules, thus solving the problems of "vague incentives and insufficient motivation." The synergistic effect of the two steps ensures both the rationality of the incentives (not ignoring personnel capabilities and past performance) and the effectiveness of the incentives (truly stimulating personnel action), effectively avoiding the problems of mismatched referral resources and low personnel enthusiasm. This provides stable driving force for the efficient connection between target enterprises and target personnel, further improving the success rate of business opportunity conversion.

[0097] In one implementation, determining the revenue value based on the value data of the target enterprise data, the business capability data of the target personnel data, the historical feedback data of the target personnel data, and the business record data of the target personnel data includes: determining the revenue value based on a first algorithm, wherein the first algorithm includes:

[0098] ;

[0099] in, The fundamental value data for the target enterprise; Value scoring data for target company data; The highest historical task value corresponding to the target company's data; The matching score between target company data and target personnel data; The average difficulty of completing tasks based on the target personnel's historical data; This refers to the number of tasks completed by the target personnel within a preset time period prior to the current moment; This refers to the historical average number of transactions completed by the target personnel within a preset time period.

[0100] In one exemplary implementation, the first algorithm overcomes the limitation of traditional business opportunity revenue using only the company's expected revenue as a single benchmark, by integrating the target company's value (based on...). Based on, / (calibrating the actual value level of the enterprise) and the capabilities and performance of the target personnel (based on) / Assess the suitability of personnel for business opportunities. / This framework, reflecting both recent and historical employee activity levels, constructs a three-dimensional revenue calculation framework linking "enterprise value, employee fit, and employee performance." This design ensures both the rationality of revenue calculation and fairness. It avoids excessive resource consumption for high-value opportunities that are extremely easy for employees to complete. Even if the enterprise's basic value is low, if recommending employees to complete the opportunity is difficult (…), it will still generate revenue. Higher than ) or outstanding recent business performance ( Higher than Even with additional compensation, the system effectively incentivizes recommenders to proactively connect with suitable business opportunities and improve business efficiency by quantifying the impact of their abilities and performance on earnings. This provides quantitative support for addressing the issues of resource imbalance and insufficient motivation among recommenders in traditional earnings distribution mechanisms, and lays a fair and motivating foundation for a virtuous cycle of dynamic allocation of overall business opportunities.

[0101] Figure 6 This is a flowchart of a method for allocating a preset time window period for enterprise data to be allocated according to an embodiment of this application, such as... Figure 6 As shown, in one embodiment, it further includes:

[0102] Step S601: After filtering out the data of enterprises to be assigned and the list of candidates, a window period with a preset time period is allocated to the data of enterprises to be assigned, so that the data of the target personnel can be connected to the data of the target enterprises within the window period.

[0103] In one exemplary implementation, for example, a commercial bank, through its internal system, filters out data on a company to be allocated loans as "a small and medium-sized enterprise engaged in the research and development of new energy batteries" (this enterprise has a technological transformation loan demand of 3 million yuan, which, according to preliminary assessment, is a highly suitable business opportunity). Simultaneously, it filters out four account managers with experience in the new energy industry and a loan business conversion rate exceeding 75% in the past six months as candidate personnel data, ultimately determining account manager Chen as the target personnel data. Combining the standard process of technological transformation loan business (requiring the completion of three core stages: enterprise demand verification, collateral evaluation, and loan plan formulation, typically requiring 12 to 18 working days), the system allocates a 15-working-day window period (November 1st to November 20th, 2024) to the data of the company to be allocated loans. The system also sends a connection instruction to manager Chen via system message, specifying the key actions to be completed within the window period (completing the on-site visit to the enterprise before November 5th and submitting the collateral evaluation report before November 15th); at the same time, the system sends a connection notification letter to the new energy enterprise, indicating manager Chen's contact information and the service progress milestones within the window period. Therefore, by setting clear window periods and key nodes, time boundaries are defined for the connection between target personnel and enterprises, avoiding the problems of "no time limit for connection and slow follow-up" in the traditional model. This not only allows Manager Chen to have a clear work rhythm plan, but also allows enterprises to have clear expectations for service progress, effectively improving the initiative and efficiency of the connection between the two parties and preventing business opportunities from being lost due to indefinite postponement.

[0104] Step S602: If the target personnel data is not successfully linked to the target enterprise data within the window period, the target enterprise data will be reallocated in the next window period.

[0105] In one exemplary implementation, for example, during the 15-working-day window period, although Account Manager Chen completed the on-site visit to the company on November 4th, the new energy company failed to provide the property ownership certificate for the collateral on time (requiring supplementary historical ownership change materials). As a result, by the end of the window period on November 20th, the collateral appraisal report had not been submitted, and the system automatically determined that the connection was unsuccessful through "key node not completed". At this time, the system puts the new energy company's pending allocation data back into the revenue opportunity pool, and when the next 15-working-day window period (November 22nd to December 9th, 2024) starts, combined with the latest account load of the candidate personnel list data (excluding the two high-priority opportunities that Manager Chen has currently added), the company's data is reassigned to another new energy field Account Manager, Manager Zhao. At the same time, the connection records of the previous window period are synchronized (such as the list of materials already provided by the company, and supplementary matters regarding the ownership change of the collateral), helping Manager Zhao to quickly grasp the previous progress. Therefore, the problem of wasted business opportunities due to single-time rejection is avoided. Through the mechanism of "returning and redistributing if unsuccessful", the data of enterprises to be assigned is always in an effective circulation state. This not only provides enterprises with the opportunity to obtain services again, but also reduces the information gap of new contact persons through historical record synchronization, reduces the cost of repeated communication, and maximizes the preservation of the potential value of business opportunities.

[0106] Through steps S601 and S602, step S601 addresses the pain points of "loose processes and uncontrolled progress" in traditional opportunity management by pre-setting reasonable window periods and key nodes to drive efficiency through time constraints. Step S602 addresses the pain points of "abandonment after a single failure and waste of resources" by reassigning opportunities after failed connections and dynamically reactivating their value through backflow. These two steps form a closed loop, ensuring both the timeliness and standardization of opportunity matching while maximizing the value of data from companies awaiting allocation. This prevents opportunities from being lost due to delays or single failures, providing a process-level safety net for successful subsequent connections between target personnel and companies. It further optimizes the allocation efficiency of opportunity resources and helps improve the overall success rate of opportunity conversion.

[0107] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods according to the above embodiments can be implemented by adding necessary general-purpose hardware platforms with the aid of software. Of course, they can also be implemented using hardware, but in many cases, the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0108] This application also provides a target enterprise data and target personnel data matching system, which is used to implement the above embodiments and preferred embodiments, and will not be repeated hereafter. As used below, a "module" can be a combination of software and / or hardware that implements a predetermined function. Although the apparatus described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0109] Figure 7 This is a structural block diagram of a target enterprise data and target personnel data matching system according to an embodiment of this application, such as... Figure 7 As shown, the system includes:

[0110] Evaluation module 71 is used to filter out target enterprise data groups based on historical data;

[0111] In one exemplary implementation, for example, the evaluation module 71 may employ an intelligent model to filter out a target enterprise data set based on historical data using a trained and converged intelligent model.

[0112] The matching module 72 is used to filter out enterprise data to be assigned and candidate personnel list data based on real-time data, wherein the real-time data includes a target enterprise data group; and to calculate a first matching degree value of the enterprise data to be assigned and the candidate personnel list data; and to determine the enterprise data to be assigned and the candidate personnel list data with the first matching degree value greater than or equal to a preset threshold as target enterprise data and target personnel data; and to specify the target personnel data to be matched with the target enterprise data.

[0113] In one exemplary implementation, for example, the matching module 72 can be an XGBoost classification model. Specifically, the XGBoost classification model can be used to comprehensively evaluate the data of companies to be assigned (business opportunity data) and the list of candidate personnel (recommended personnel list data), generate a first matching degree value between the data of companies to be assigned and the list of candidate personnel, and finally determine the appropriate target personnel data (the list of target recommended personnel). The model data of the XGBoost classification model mainly consists of two parts: corporate customer characteristics and recommended personnel characteristics.

[0114] Among them, the characteristics of corporate clients mainly include corporate business information (registered capital, years of establishment, industry type, organizational nature, enterprise size, enterprise qualifications, etc.), financial assets (average daily financial volume, deposit amount, investment and wealth management amount, etc.), financial products (number of products covered in the past 1 / 3 / 6 / 12 months, historical product holdings, product preferences, etc.), fund flows (debit funds and number of transactions in the past 1 / 3 / 6 / 12 months, loan amount and number of transactions in the past 1 / 3 / 6 / 12 months, etc.), corporate credit (debt balance in the past 1 / 3 / 6 / 12 months, credit method, number of overdue transactions in the past 1 / 3 / 6 / 12 months), enterprise value (current EVA value, average EVA (Economic Value Added) in the past 1 / 3 / 6 / 12 months, increase in EVA compared to the previous quarter, etc.), corporate legal person (internal customer level, number of mobile banking logins in the past 1 / 3 / 6 / 12 months, customer's average daily AUM (Assets Under Management) in the past year, etc.). The data encompasses 16 data dimensions and 218 data features, including (Assets under management), number of personal products covered, investment preferences, number of mobile banking visits, average mobile banking browsing time, and average monthly in-store visits).

[0115] The characteristics of the recommenders mainly include basic user information (age, gender, years of service, etc.), customer information within the industry (distribution of managed companies by size, industry, establishment time, and company affiliation, etc.), and recommendation performance (number of business opportunities recommended in the last 1 / 3 / 6 / 12 months, average recommendation success rate for each type of company, distribution of recommended products, etc.), covering a total of 8 data dimensions and 46 data features.

[0116] Finally, the two sets of data are integrated to form the final input data table.

[0117] Regarding the data modeling part, through comparison of various models in terms of accuracy and stability, the final model adopted is the XGBoost model to model multi-classification problems and calculate the matching degree between recommended personnel and business opportunities.

[0118] 1. Data preprocessing: Data processing is performed on the wide data table generated after data collection. The main processing steps are as follows: (1) Outlier handling: Outlier handling is performed according to different types of features and business definitions, mainly including missing value supplementation and extreme value handling. (2) Data cleaning: Standardization or binning is adopted for continuous variables to eliminate the influence of units and increase business interpretability. For example, AUM of corporate entities is classified and quantified. (3) Feature engineering: New business feature indicators are constructed based on single-column derivation, multi-column grouping and aggregation, such as the average amount and number of transactions of enterprises in the past 12 months. The variance selection method, correlation coefficient method, IV threshold, model selection and other methods are used to comprehensively evaluate the feature relevance and eliminate low relevance features.

[0119] 2. Model Construction: Input data features are used as independent variables, and the output is the matching degree between business opportunity companies and recommenders. The data training model can use the XGBoost classification model, which iteratively builds multiple decision trees to gradually correct the residuals of previous models, while introducing regularization and engineering optimization to improve performance.

[0120] To address the lack of historical business opportunity recommendation data during the initial model training phase, the model utilizes nearly one year's worth of internal corporate product recommendation records from the bank during initial training. In subsequent model iterations, the recommended results assigned by the model are periodically used as training data for iterative optimization of model parameters. Recall, AUC (Average Unit Cost), and F1 score (F1-Score, a comprehensive performance evaluation metric for classification models) are used as core performance evaluation metrics. A hyperparameter space is constructed, including the number of learners, maximum tree depth, minimum sample size per leaf node, maximum number of leaf nodes, and gradient boosting iterations. A grid search algorithm is used to traverse parameter combinations, combined with a k-fold cross-validation strategy for model validation, ultimately yielding the optimal model.

[0121] 3. Model Result Validation: (1) Accuracy: The model performed stably on the training set, test set, and validation set, with accuracy rates of 91%, 88%, and 85%, respectively. Using a multi-dimensional performance evaluation framework based on the confusion matrix, its recall and precision balance significantly outperformed the industry benchmark model, meeting business requirements. (2) Probability Calibration Validation: Through binning KS test (e.g., setting...), the model achieved a stable performance on the training set, test set, and validation set. (3) Key features of the model: The key features of the model are the number of years the enterprise has been established, the number of business opportunities recommended, and the degree of enterprise relevance. After verification with relevant business personnel, it was found that the model basically conforms to the recommendation logic.

[0122] By adopting the above technical solution, the evaluation module 71 filters the target enterprise data group based on historical data, breaking through the limitation of existing technologies that rely solely on static data. It can mine more comprehensive characteristics of enterprises' operations and needs from a historical perspective, initially identifying a range of enterprises with matching value, laying the foundation for subsequent precise matching. The matching module 72 further filters the data of enterprises to be assigned and the list of candidates based on real-time data containing the target enterprise data group, taking into account both the accumulation of historical data and the dynamism of real-time data. This ensures that the data of enterprises to be matched and account managers are timely and targeted, avoiding the problem of the extensive business opportunity pool data in existing technologies. Meanwhile, this module also establishes a quantitative correlation analysis mechanism between enterprise data and account manager data by calculating the first matching degree value between the data of enterprises to be assigned and the list of candidates, replacing the indiscriminate broadcast push. Then, objects with a first matching degree value greater than or equal to a preset threshold are identified as target enterprise data and target personnel data and designated for connection, effectively avoiding the phenomenon of "capability mismatch". Account managers do not need to identify suitable business opportunities themselves. Finally, through the functional collaboration of the evaluation module 71 and the matching module 72, the problem of low matching success rate between account manager data and data of enterprises to be served in related technologies is solved, and the effect of improving the matching success rate of the two is achieved.

[0123] In one implementation, the evaluation module 71 includes: a data evaluation model and a data scoring model.

[0124] The data evaluation models include:

[0125] The business evaluation unit is used to determine new business data for a future preset time period based on the company's historical business data, and to determine the company's business score based on the new business data and the company's credit score in the system.

[0126] In one exemplary implementation, for example, the operational assessment unit uses the IGBM model to predict the company's future revenue growth based on data from various dimensions, including basic enterprise information (registered capital, company age, deposit amount, industry type, etc.), scientific and technological innovation capabilities (number / growth rate of patents, number / growth rate of software copyrights, enterprise qualifications, etc.), market performance (number / amount of bidding, number / amount of financing, etc.), internal relationships within the industry (number of products covered, average daily balance of deposits / loans, average daily balance of loans / loans, average daily number / amount of transactions, customer five-category rating, etc.), and the actual controller's internal situation (customer rating within the industry, average daily balance of deposits / loans / loans, investment and wealth management AUM, product coverage, WeChat account, etc.). This data is then combined with the internal enterprise credit score to form the final enterprise operational score.

[0127] The business performance evaluation unit integrates a business performance scoring algorithm, which includes:

[0128] ;

[0129] ;

[0130] ;

[0131] in, Weights for each evaluation representative (e.g., pre-defined) , , ); To represent the company's current operating level, take the company's current Longxin Business Score; To represent the industry level, the average value of Longxin Business, a company in the same industry, is taken. To assess the company's future potential, the IGBM algorithm was used to predict the company's future revenue growth. For enterprises and industries.

[0132] The intent assessment unit is used to determine the enterprise's intent score based on historical pairing data between enterprises and products.

[0133] The risk assessment unit is used to determine new risk data within a preset future time period based on the company's historical risk data, and to determine the company's risk score based on the new risk data.

[0134] In one exemplary implementation, for example, the risk assessment unit is based on 272 data features, including business registration data (industry category, registered capital, etc.), industry business data (average daily balance of deposits / loans, number of products covered, etc.), and enterprise cash flow (number of inflows / outflows of funds in the past 3 months, etc.). Using the IGBM model algorithm, based on historical enterprise data, the probability of the enterprise developing overdue and non-performing loans in the next month is calculated, and then the probability value is converted into an enterprise risk score using a binning method.

[0135] The data scoring model is used to: filter target enterprise data groups based on the enterprise operation score, the enterprise intention score, and the enterprise risk score, wherein the historical data includes: the enterprise's historical operation data, the enterprise's historical pairing data with products, and the enterprise's historical risk data.

[0136] In one exemplary implementation, for example, ;in, Weights for operational evaluation units; The weights for the units to be evaluated; Weights for risk assessment units; Assign a business performance score to the enterprise; This is the score for the company's interest. Assign a risk score to the enterprise.

[0137] In one implementation, the intent assessment unit includes:

[0138] The hybrid expert subunit is used to determine the enterprise's intent score based on industry data and the enterprise's needs data.

[0139] In one exemplary implementation, for example, the hybrid expert subunit obtains the enterprise intent score by identifying the industry to which the enterprise belongs and by calling different industry sub-models and customer demand response models in a weighted manner.

[0140] The AI ​​large model subunit is used to determine a second matching degree value between the enterprise and the product based on enterprise data and product data; and to determine an enterprise intention score value based on the enterprise intention score value and the second matching degree value.

[0141] In one exemplary implementation, for example, an AI large model subunit is used to calculate a matching degree value (second matching degree value) between the target enterprise and various financial product information based on enterprise data and product data.

[0142] In one implementation, the intent evaluation unit integrates a second algorithm. The second algorithm includes:

[0143] ;

[0144] in, Weighting of the enterprise intent score; The weight of the second matching degree value; A function for scoring enterprise intent; The coefficient weights of the enterprise intent scoring function; This is the function for the second matching degree value; The coefficient weights are the values ​​of the second matching degree function.

[0145] Figure 8 This is a structural block diagram of another target enterprise data and target personnel data matching system according to an embodiment of this application, such as... Figure 8 As shown, in one embodiment, the system further includes: an incentive module 73, used to determine a revenue value based on the value data of the target enterprise data, the business capability data of the target personnel data, the historical feedback data of the target personnel data, and the business record data of the target personnel data; and to use the revenue value as incentive value data for the target personnel data.

[0146] In one implementation, the excitation module 73 integrates a first algorithm. The first algorithm includes:

[0147] ;

[0148] in, The fundamental value data for the target enterprise; Value scoring data for target company data; The highest historical task value corresponding to the target company's data; The matching score between target company data and target personnel data; The average difficulty of completing tasks based on the target personnel's historical data; This refers to the number of tasks completed by the target personnel within a preset time period prior to the current moment; This refers to the historical average number of transactions completed by the target personnel within a preset time period.

[0149] Figure 9 This is a structural block diagram of another target enterprise data and target personnel data matching system according to an embodiment of this application, such as... Figure 9 As shown, in one embodiment, the system further includes: a window period generation module 74, used to allocate a preset time period to the enterprise data to be allocated when filtering out enterprise data to be allocated and candidate personnel list data, so that the target personnel data can be connected to the target enterprise data within the window period; and, if the target personnel data fails to be successfully connected to the target enterprise data within the window period, to reallocate the target enterprise data in the next window period.

[0150] In one exemplary implementation, for example, to ensure that business opportunities can be continuously and cyclically allocated in the long term, a fixed time window mode is adopted for the release of business opportunities. The recommender needs to complete the business opportunity within the window period. If the window period is exceeded, the business opportunity will be put back into the business opportunity pool for reallocation.

[0151] In one implementation, the matching module 72 is further configured to: perform model iteration based on the data records of enterprises to be assigned in the previous time window period; if the data of enterprises to be assigned is completed within the window period, it is a positive sample; if the recommendation is unsuccessful, it is a negative sample.

[0152] In one exemplary implementation, for example, the matching module 72 iterates the model based on the recommended business opportunity records in the previous time window. If the business opportunity is completed within the window, it is a positive sample; if it is not successfully recommended, it is a negative sample.

[0153] In one implementation, the evaluation module 71 is also used to: perform monthly data iteration updates based on internal system data.

[0154] In one exemplary implementation, for example, the evaluation module 71 performs monthly model iteration updates based on internal system data.

[0155] In summary, the embodiments of this application can establish a dynamic allocation model for business opportunities for SMEs, which includes an evaluation module 71, a matching module 72, an incentive module 73, a window period generation module 74, and model iteration. This realizes dynamic matching of business opportunities for SMEs with recommending personnel, and further enables banks to accurately target business opportunities for SMEs.

[0156] It should be noted that the above modules can be implemented by software or hardware. For the latter, they can be implemented in the following ways, but are not limited to: all the above modules are located in the same processor; or, the above modules are located in different processors in any combination.

[0157] This application also provides a computer-readable storage medium storing a computer program configured to execute the steps in any of the above method embodiments when running.

[0158] In one exemplary embodiment, the aforementioned computer-readable storage medium may include, but is not limited to, various media capable of storing computer programs, such as a USB flash drive, read-only memory (ROM), random access memory (RAM), portable hard disk, magnetic disk, or optical disk.

[0159] This application also provides an electronic device including a memory and a processor, wherein the memory stores a computer program and the processor is configured to run the computer program to perform the steps in any of the above method embodiments.

[0160] In one exemplary embodiment, the electronic device may further include a transmission device and an input / output device, wherein the transmission device is connected to the processor and the input / output device is connected to the processor.

[0161] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps in any of the above method embodiments.

[0162] Specific examples in the embodiments of this application can be found in the examples described in the above embodiments and exemplary implementations, and will not be repeated here.

[0163] Obviously, those skilled in the art should understand that the modules or steps of this application described above can be implemented using general-purpose computing devices. They can be centralized on a single computing device or distributed across a network of multiple computing devices. They can be implemented using computer-executable program code, and thus can be stored in a storage device for execution by a computing device. In some cases, the steps shown or described can be performed in a different order than those presented here, or they can be fabricated as separate integrated circuit modules, or multiple modules or steps can be fabricated as a single integrated circuit module. Thus, this application is not limited to any particular combination of hardware and software.

[0164] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the principles of this application should be included within the protection scope of this application.

Claims

1. A method for matching target enterprise data with target personnel data, characterized in that, include: Target enterprise data groups were selected based on historical data; The data of enterprises to be assigned and the list of candidates are filtered based on real-time data, wherein the real-time data includes the target enterprise data group; Calculate the first matching degree value between the enterprise data to be assigned and the candidate list data; The data of enterprises to be assigned and the list of candidates whose first matching degree value is greater than or equal to the preset threshold are identified as the target enterprise data and the target personnel data. Specify that the target personnel data is linked to the target enterprise data.

2. The method according to claim 1, characterized in that, Based on historical data, a target enterprise data group was selected, including: Based on the company's historical operating data, new operating data for a future preset time period is determined, and based on the new operating data and the company's credit score in the system, the company's operating score is determined. The enterprise intention score is determined based on historical enterprise-product pairing data. Based on the enterprise's historical risk data, new risk data for a future preset time period is determined, and the enterprise's risk score is determined based on the new risk data. The target enterprise data group is selected based on the enterprise operation score, the enterprise intention score, and the enterprise risk score. The historical data includes: the enterprise's historical operation data, the enterprise's historical pairing data with products, and the enterprise's historical risk data.

3. The method according to claim 2, characterized in that, The enterprise intention score is determined based on historical enterprise-product pairing data, including: The company's intent score is determined based on industry data and the company's needs data. The second matching degree value between the enterprise and the product is determined based on enterprise data and product data; The enterprise intention score is determined based on the enterprise intention score and the second matching score.

4. The method according to claim 1, characterized in that, Also includes: The revenue value is determined based on the value data of the target enterprise data, the business capability data of the target personnel data, the historical feedback data of the target personnel data, and the business record data of the target personnel data. The revenue value is used as an incentive value for the target personnel data.

5. The method according to claim 1, characterized in that, Also includes: After filtering out the data of enterprises to be assigned and the list of candidates, a window period with a preset time period is allocated to the data of enterprises to be assigned, so that the data of the target personnel can be connected to the data of the target enterprises within the window period; If the target personnel data fails to be successfully linked to the target enterprise data within the specified window period, the target enterprise data will be reallocated in the next window period.

6. The method according to claim 4, characterized in that, Determining revenue based on the value data of the target enterprise data, the business capability data of the target personnel data, the historical feedback data of the target personnel data, and the business record data of the target personnel data includes: determining revenue based on a first algorithm, wherein the first algorithm includes: ； in, The fundamental value data for the target enterprise; Value scoring data for target company data; The highest historical task value corresponding to the target company's data; The matching score between target company data and target personnel data; The average difficulty of completing tasks based on the target personnel's historical data; This refers to the number of tasks completed by the target personnel within a preset time period prior to the current moment; This refers to the historical average number of transactions completed by the target personnel within a preset time period.

7. The method according to claim 3, characterized in that, Determining a company's intention score based on historical company-product pairing data includes: determining the company's intention score based on a second algorithm, wherein the second algorithm includes: ； in, Weighting of the enterprise intent score; The weight of the second matching degree value; A function for scoring enterprise intent; The coefficient weights of the enterprise intent scoring function; This is the function for the second matching degree value; The coefficient weights are the values ​​of the second matching degree function.

8. A system for matching target enterprise data with target personnel data, characterized in that, include: The evaluation module is used to filter out target enterprise data groups based on historical data; A matching module is used to filter out enterprise data to be assigned and candidate personnel list data based on real-time data, wherein the real-time data includes the target enterprise data group; and to calculate a first matching degree value between the enterprise data to be assigned and the candidate personnel list data; and to determine the enterprise data to be assigned and candidate personnel list data with the first matching degree value greater than or equal to a preset threshold as target enterprise data and target personnel data; and to assign the target personnel data to the target enterprise data.

9. A computer-readable storage medium, characterized in that, The storage medium stores a computer program, wherein the computer program is executed by a processor to perform the method described in any one of claims 1 to 7.

10. An electronic device comprising a memory and a processor, characterized in that, The memory stores a computer program, and the processor is configured to run the computer program to perform the method of any one of claims 1 to 7.

11. A computer program product comprising computer instructions, characterized in that, When the computer instructions are executed by the processor, they implement the method of any one of claims 1 to 7.

Images