Service recommendation method, system and equipment based on supply two sides and medium

Abstract

The invention discloses a service recommendation method, system and device based on supply double sides and a medium. The method comprises the steps of obtaining service demand side data and service supply side data; according to the portrait association degree of the service demand side data and the service supply side data, performing service recommendation processing on a target object of the service demand side to obtain a candidate service set; performing combination processing on the candidate service set according to a population evolution algorithm to obtain a service combination; and performing optimization processing on the service combination according to a multi-criterion clustering model to obtain a recommendation result. The embodiment of the invention can improve the service recommendation efficiency, and can be widely applied to the technical field of computers.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a method, system, device, and medium for recommending services based on both supply and demand sides. Background Art

[0002] With economic and social development and improved living standards, the public's demand for health services is increasing. Currently, the variety of health services has increased dramatically, covering a wide range of areas, including medical care, nutrition and health care, fitness and leisure. Related technologies exist that recommend health services through health consultants and caregivers. However, in practical applications, these methods have found that it is difficult to accurately integrate health services based on the needs of health service users. This leads to deviations between recommendation results and actual resource distribution and service availability, making it difficult to achieve a global optimal solution and affecting the efficiency of service recommendations.

[0003] In summary, the technical problems existing in the relevant technologies need to be improved. Summary of the Invention

[0004] The main purpose of the embodiments of the present application is to propose a service recommendation method, system, device and medium based on both the supply and demand sides, which can improve the service recommendation effect.

[0005] To achieve the above objectives, one aspect of an embodiment of the present application proposes a service recommendation method based on both the supply and demand sides, the method comprising:

[0006] Obtain service demand-side data and service supply-side data;

[0007] Performing service recommendation processing on the target object on the service demand side according to the profile correlation degree between the service demand side data and the service supply side data to obtain a candidate service set;

[0008] Combining the candidate service sets according to a population evolution algorithm to obtain a service combination;

[0009] The service combination is optimized according to a multi-criteria clustering model to obtain a recommendation result.

[0010] In some embodiments, performing service recommendation processing on a target object on the service demand side according to the degree of association between the profiles of the service demand side data and the service supply side data to obtain a candidate service set includes the following steps:

[0011] An object portrait is constructed based on the service demand side data, and a service portrait is constructed based on the service supply side data;

[0012] Performing correlation analysis on the health attributes and economic attributes of the object portrait and the service portrait to obtain a correlation degree between the portraits;

[0013] The services provided by the service supply side are recommended to the target object according to the degree of association of the portraits to obtain the candidate service set.

[0014] In some embodiments, the process of recommending services provided by the service supply side to the target object based on the portrait association degree to obtain the candidate service set includes the following steps:

[0015] The services provided by the service supply side are traversed according to the random walk algorithm, and the selection probability of the walk path is adjusted according to the correlation degree of the portrait, and the service importance is calculated using the importance calculation formula;

[0016] Selecting services provided by a service supply side according to the service importance to obtain the candidate service set;

[0017] The candidate service set is recommended to the target object.

[0018] In some embodiments, combining the candidate service sets according to the population evolution algorithm to obtain the service combination includes the following steps:

[0019] Selecting the candidate service set through the target object to obtain a target service set;

[0020] Performing index analysis and topology structure combination processing on the target service set to obtain a service composition model;

[0021] The service composition model is optimized and solved according to the population evolution algorithm to obtain the service composition.

[0022] In some embodiments, performing optimization and solving processing on the service composition model according to the population evolution algorithm to obtain the service composition includes the following steps:

[0023] Performing population initialization processing on the service composition model according to the population evolution algorithm to obtain a population;

[0024] Perform algorithm selection processing on the population according to the literature analysis method to construct an optimization algorithm comprehensive strategy pool;

[0025] The population is updated through the optimization algorithm comprehensive strategy pool according to the evaluation criteria to obtain the service combination.

[0026] In some embodiments, optimizing the service combination according to the multi-criteria clustering model to obtain a recommendation result includes the following steps:

[0027] updating the service demand side data according to the service combination, and performing vector space mapping processing on the updated data and the service combination to obtain a data set;

[0028] Performing model building processing according to the intra-cluster data compactness index and the inter-cluster independence index to obtain the multi-criteria clustering model;

[0029] Performing clustering processing on the data set by using the multi-criteria clustering model to obtain clustering information;

[0030] The service combination is subjected to service optimization processing according to the clustering information to obtain the recommendation result.

[0031] In some embodiments, clustering the data set using the multi-criteria clustering model to obtain clustering information includes the following steps:

[0032] Performing a minimum spanning tree generation process on the data set, and performing edge removal and random insertion processes on the minimum spanning tree according to a density index to obtain an initial clustering scheme;

[0033] Performing index calculation processing on the initial clustering schemes, and merging the initial clustering schemes according to the calculated indexes to obtain a target clustering scheme;

[0034] A population update process is performed on the target clustering scheme to obtain the clustering information.

[0035] To achieve the above objectives, another aspect of the present application provides a service recommendation system based on both the supply and demand sides, the system comprising:

[0036] The first module is used to obtain service demand side data and service supply side data;

[0037] The second module is used to perform service recommendation processing on the target object on the service demand side according to the profile correlation degree between the service demand side data and the service supply side data, and obtain a candidate service set;

[0038] The third module is used to combine the candidate service sets according to the population evolution algorithm to obtain a service combination;

[0039] The fourth module is used to optimize the service combination according to a multi-criteria clustering model to obtain a recommendation result.

[0040] To achieve the above-mentioned purpose, another aspect of an embodiment of the present application provides an electronic device, which includes a memory and a processor, wherein the memory stores a computer program, and the processor implements the above-mentioned method when executing the computer program.

[0041] To achieve the above objectives, another aspect of an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method described above is implemented.

[0042] The embodiments of the present application include at least the following beneficial effects: the present application provides a service recommendation method, system, device and medium based on both the supply and supply sides. The scheme recommends services to the target objects on the service demand side according to the degree of correlation between the portraits of the service demand side data and the service supply side data to obtain a candidate service set. The portraits of both the supply and demand sides can be analyzed to make personalized recommendations, thereby improving the accuracy of service matching; and the scheme combines the candidate service sets according to the population evolution algorithm to obtain a service combination, and can select the optimal service combination scheme from the candidate service set, thereby reducing the solution complexity and improving the computational efficiency; in addition, the scheme optimizes the service combination according to the multi-criteria clustering model to obtain a recommendation result, which can optimize the health services on both the supply and demand sides and improve the user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 This is a flow chart of a service recommendation method based on both supply and demand sides provided in an embodiment of the present application;

[0044] Figure 2 yes Figure 1 Flowchart of step S102 in FIG.

[0045] Figure 3 This is a schematic diagram of a service recommendation process provided by an embodiment of the present application;

[0046] Figure 4 is a schematic diagram of a population evolution algorithm provided in an embodiment of the present application;

[0047] Figure 5 is a schematic diagram of a health service combination model provided in an embodiment of the present application;

[0048] Figure 6 is a schematic diagram of a multi-criteria clustering model provided in an embodiment of the present application;

[0049] Figure 7 is a schematic diagram of a clustering algorithm provided in an embodiment of the present application;

[0050] Figure 8 This is a schematic diagram of a system application scenario provided by an embodiment of the present application;

[0051] Figure 9 This is a schematic diagram of the structure of a service recommendation system based on both supply and demand sides provided in an embodiment of the present application;

[0052] Figure 10 This is a schematic diagram of the hardware structure of the electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0053] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application is further described in detail below with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are only used to explain the present application and are not intended to limit the present application. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the embodiments of the present application. They are merely examples of systems and methods consistent with some aspects of the embodiments of the present application as detailed in the appended claims.

[0054] It will be understood that the terms "first", "second", etc. used in this application may be used herein to describe various concepts, but unless otherwise specified, these concepts are not limited by these terms. These terms are only used to distinguish one concept from another. For example, without departing from the scope of the embodiments of the present application, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, the words "if" and "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining".

[0055] The terms "at least one", "plurality", "each", "any", etc. used in this application include "at least one", "two" or more, "plurality" or "each", "any" or "any one", "each" or "any one" as used herein.

[0056] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which this application pertains. The terms used herein are for the purpose of describing the embodiments of this application only and are not intended to limit this application.

[0057] Before explaining the embodiments of the present application in detail, some of the nouns and terms involved in the embodiments of the present application are first explained. The nouns and terms involved in the embodiments of the present application are subject to the following explanations.

[0058] 1) NP-hard (Nondeterministic Polynomial Hard) is a core concept in computational complexity theory, used to describe a class of problems with extremely high computational complexity. NP-hard problems are key research topics in theoretical computer science and algorithm design, and their difficulty is widely considered to exceed the limits of traditional computing capabilities (unless P = NP). NP-hard problems are not necessarily classified as NP, but their complexity has been shown to be no less complex than the most difficult problems in NP (i.e., NP-complete problems).

[0059] 2) PersonalRank, a core algorithm in the field of recommendation systems, is primarily used for personalized recommendations based on graph structures. PersonalRank simulates a user's random walk through items or social networks, calculating the importance weights of nodes (users or items) to achieve precise personalized recommendations. Its core concept is derived from the PageRank algorithm, but it introduces personalized preference parameters, significantly improving the relevance and accuracy of recommendations.

[0060] With economic and social development and rising living standards, people's demand for health services is increasing. Against the backdrop of rapid advancements in health service technologies, numerous health services are becoming increasingly sophisticated. The continuous development of a range of health services, including face-to-face consultations, elderly care, rehabilitation and health care, daily care, dietary management, medication supply and delivery, and medication decoction services, has significantly improved and enriched the current health service industry chain.

[0061] For the general public, the first thing they face is how to understand their own needs for health services. Currently, the types of health services have increased dramatically, covering many fields such as medical care, nutrition and health care, fitness and leisure. Due to the limitations of the health industry chain in which health consultants and caregivers themselves are located, it is difficult to accurately integrate health services based on the needs of health service users. On the other hand, even if health service users can understand their own health status very well, it is difficult for them to choose the best set of health services that meet their needs from the health services provided by many different service providers in the entire health service industry chain based on their own health status and their specific health needs, such as good efficacy, low cost, appropriate time, and close distance. Therefore, how to help health service users select and combine health services has become an important issue that must be solved in the health service field.

[0062] For example, for example, relevant health service portfolio optimization methods usually only consider the demand side or the supply side unilaterally, and lack a comprehensive analysis of both sides of supply and demand. This one-sided analysis leads to deviations between the recommendation results and the actual resource distribution and service availability, making it difficult to achieve a global optimal solution. Related methods often use a single indicator (such as cost minimization) as the core optimization goal, ignoring the comprehensive balance of multi-dimensional demands such as efficacy, risk, and user comfort, and are difficult to adapt to personalized scenarios. When health service portfolio optimization involves multi-constraint and multi-objective decision-making (such as cost, time, efficacy, etc.), its NP-hard characteristics cause traditional mathematical methods (such as linear programming) to have a surge in computational complexity in large-scale scenarios, which cannot meet real-time response requirements.

[0063] In view of this, an embodiment of the present application provides a service recommendation method, system, device and medium based on both the supply and supply sides. The scheme recommends services to the target objects on the service demand side according to the degree of correlation between the portraits of the service demand side data and the service supply side data to obtain a candidate service set. The portraits of both the supply and demand sides can be analyzed to make personalized recommendations, thereby improving the accuracy of service matching; and the scheme combines the candidate service sets according to the population evolution algorithm to obtain a service combination, and can select the optimal service combination scheme from the candidate service set, thereby reducing the solution complexity and improving the computational efficiency; in addition, the scheme optimizes the service combination according to the multi-criteria clustering model to obtain a recommendation result, which can optimize the health services on both the supply and demand sides and improve the user experience.

[0064] The embodiment of the present application provides a service recommendation method based on the supply side, which relates to the field of service recommendation technology. The embodiment of the present application provides a service recommendation method based on the supply side, which can be applied to a terminal, a server, or software running in a terminal or a server. In some embodiments, the terminal can be a smart phone, a tablet computer, a laptop computer, a desktop computer, a smart speaker, a smart watch, and a car terminal, etc., but is not limited to this; the server side can be configured as an independent physical server, or as a server cluster or distributed system composed of multiple physical servers, or as a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms. The server can also be a node server in a blockchain network; the software can be an application that implements a service recommendation method based on the supply side, etc., but is not limited to the above forms.

[0065] The present application can be used in many general or special computer system environments or configurations. For example: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, and the like. The present application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, and the like that perform specific tasks or implement specific abstract data types. The present application can also be practiced in distributed computing environments in which tasks are performed by remote processing devices connected via a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media, including storage devices.

[0066] It should be noted that in each specific embodiment of the present application, when it comes to the need to perform relevant processing based on data related to the user's identity or characteristics, such as user information, user behavior data, user historical data, and user location information, the user's permission or consent will be obtained first, and the collection, use, and processing of such data will comply with relevant laws, regulations, and standards. In addition, when the embodiment of the present application needs to obtain the user's sensitive personal information, the user's separate permission or consent will be obtained through a pop-up window or by jumping to a confirmation page. After clearly obtaining the user's separate permission or consent, the necessary user-related data for the normal operation of the embodiment of the present application will be obtained.

[0067] Figure 1 This is an optional flow chart of a service recommendation method based on both supply and demand sides provided in an embodiment of the present application. Figure 1 The method may include but is not limited to steps S101 to S104.

[0068] Step S101, obtaining service demand side data and service supply side data;

[0069] Step S102, performing service recommendation processing on the target object on the service demand side according to the profile correlation degree between the service demand side data and the service supply side data, to obtain a candidate service set;

[0070] Step S103, combining the candidate service sets according to a population evolution algorithm to obtain a service combination;

[0071] Step S104: Optimize the service combination according to a multi-criteria clustering model to obtain a recommendation result.

[0072] Steps S101 to S104 shown in the embodiment of the present application can be applied to the health service recommendation scenario, and service recommendations are made by obtaining service demand side data and service supply side data, wherein the service demand side data is relevant data of the service object, such as the health service demander's own portrait, the health service's own characteristics and other information. The service supply side data is relevant data of various health services provided by the health service provider. The embodiment of the present application provides an analysis of the degree of portrait correlation between the service demand side data and the service supply side data, so as to perform service recommendation processing on the health service of the service supply side to the target object of the service demand side, and obtain a candidate service set. The candidate service set is a plurality of recommended health services, which is used to provide a data basis for subsequent service combination. Then, the embodiment of the present application combines the candidate service set according to the population evolution algorithm to obtain a service combination, which can recommend a health service combination to the health service demander based on the population evolution algorithm considering the health service demander's own portrait, the health service's own characteristics and other information. In addition, the embodiment of the present application optimizes the service combination based on the multi-criteria clustering model, and optimizes the combination of health services by fully considering factors such as the user's health information and personalized needs and the level of cooperation between health service providers, thereby obtaining the final recommendation results.

[0073] One of the above technical solutions has the following advantages or beneficial effects: The embodiment of the present application analyzes the degree of correlation between health service users and health services, accurately screens out health services that match user needs, generates a set of candidate services, and more accurately identifies user potential needs, thereby improving the pertinence and accuracy of health service recommendations and providing users with a more personalized service experience. In addition, the embodiment of the present application optimizes the service portfolio through a multi-criteria clustering model, which can provide optimization directions for service providers and improve the efficient matching and collaborative optimization of both supply and demand sides.

[0074] In step S101 of some embodiments, service demand-side data and service supply-side data may be obtained through a service interface. The service demand-side data includes target user profile data, which may include basic user information such as age, gender, location, and occupation. The service supply-side data includes data such as the service name, type, description, price, service scope, category, and tags provided by the service provider. The data may also be obtained through other means, such as obtaining database permissions, but is not limited thereto.

[0075] See also Figure 2 In step S102 of some embodiments, performing service recommendation processing on a target object on the service demand side according to the degree of association between the profiles of the service demand side data and the service supply side data to obtain a candidate service set includes the following steps:

[0076] Step S201: constructing an object portrait based on the service demand-side data, and constructing a service portrait based on the service supply-side data;

[0077] Step S202: performing a correlation analysis of the health attributes and economic attributes of the object portrait and the service portrait to obtain a correlation degree of the portraits;

[0078] Step S203: recommend the services provided by the service supply side to the target object according to the portrait association degree to obtain the candidate service set.

[0079] In the embodiment of the present application, the object portrait can be constructed based on the service demand side data, the service portrait can be constructed based on the service supply side data, or the corresponding portrait data can be directly obtained from the service demand side data and the service supply side data. In a feasible embodiment, please refer to Figure 3 The service resources on the health service platform include face-to-face consultation, Chinese medicine decoction, drug delivery and other sub-services. For each sub-service, there will be many service providers, and the users in the health service combination are health service demanders. In the scenario of the health service combination, i represents a certain type of health service, and i will correspond to multiple health service providers. The embodiment of this application realizes personalized health service recommendations for users by constructing user health portraits and health service portraits. The health portrait of user j is recorded as PU j , then PU j It includes the values evaluated by user j on multiple health signs, and the profile of health service i is recorded as PS i , corresponding to,PS i It includes the correlation degree between the health service i and multiple health signs of the user.

[0080] Both health service users and health services possess health-related attributes and economic-related attributes, and these two attributes are correlated. For health service users, health-related attributes should include various health indicators of the health service user, such as blood pressure and blood sugar, including symptoms of some common chronic conditions and signs of sub-health. Economic attributes should also consider the health service user's economic characteristics, such as disposable income and income stability. Health services also possess these two attributes. From a health attribute perspective, many health services have varying degrees of effect on different health indicators of users. These effects on user health constitute the health attributes of the health service itself. At the same time, each health service has different economic attributes, such as long-term and short-term costs. This embodiment of the application constructs a user health profile and an economic profile based on the health and economic perspectives of the health service user, respectively. These two profiles together constitute the health service user profile. Similarly, a health service efficacy profile and a cost profile are constructed, which together constitute the health service profile. This embodiment of the application analyzes the correlation between the health and economic attributes of the object profile and the service profile to determine the degree of correlation between the profiles. Finally, the services provided by the service supply side are recommended to the target object based on the degree of portrait association to obtain a set of candidate services.

[0081] One of the above technical solutions has the following advantages or beneficial effects: The embodiment of the present application realizes personalized health service recommendations for users by constructing user health portraits and health service portraits, which can simultaneously consider the user's personalized needs and the service capabilities of the supply side, achieve accurate matching of both supply and demand sides, and improve the accuracy of recommendations.

[0082] In some embodiments, the process of recommending services provided by the service supply side to the target object based on the portrait association degree to obtain the candidate service set includes the following steps:

[0083] The services provided by the service supply side are traversed according to the random walk algorithm, and the selection probability of the walk path is adjusted according to the correlation degree of the portrait, and the service importance is calculated using the importance calculation formula;

[0084] Selecting services provided by a service supply side according to the service importance to obtain the candidate service set;

[0085] The candidate service set is recommended to the target object.

[0086] In the embodiment of the present application, the random walk mechanism of the traditional PersonalRank algorithm is improved, and the health service users and health service portraits are analyzed to obtain ω pqTo represent the degree of association between the node p corresponding to health service i and the node corresponding to health service user j. ij Applied to the random walk process. When the random walk reaches node q, the importance of node q is updated according to the importance calculation formula, where the expression of the importance calculation formula is shown as follows:

[0087]

[0088] Where RR(q) represents the importance of node q, α represents a parameter between 0 and 1, and u represents the user node to be recommended.

[0089] Finally, the services provided by the service supply side are selected according to the service importance, and multiple health services with greater service importance can be recommended to the target object, thereby obtaining a candidate service set.

[0090] One of the above technical solutions has the following advantages or beneficial effects: This embodiment dynamically adjusts the roaming path based on the correlation between the user profile and the service profile, accurately screening health services that match the user's needs and generating a set of candidate services. This allows for more accurate identification of potential user needs, improves the pertinence and accuracy of health service recommendations, and provides users with a more personalized service experience.

[0091] In some embodiments, combining the candidate service sets according to the population evolution algorithm to obtain the service combination includes the following steps:

[0092] Selecting the candidate service set through the target object to obtain a target service set;

[0093] Performing index analysis and topology structure combination processing on the target service set to obtain a service composition model;

[0094] The service composition model is optimized and solved according to the population evolution algorithm to obtain the service composition.

[0095] In the embodiment of the present application, the optimal service combination solution is selected from the candidate service set. Since the second-stage service combination model of the health service combination is a complex optimization problem with NP-hard characteristics, classical mathematical solutions cannot solve this type of problem well. In this embodiment of the present application, a population evolution algorithm is used. It can introduce a variety of adaptive mechanisms in the intelligent optimization algorithm framework to cooperate with the individual update strategy pool, adaptively update the individual generation method, and combine the population evolution direction evaluation mechanism to solve the optimal service combination.

[0096] After the recommendation results are pushed to the user, the embodiment of the present application will guide the user to explore his or her potential needs for health services and provide guidance for the user to select health services. All health services selected by the user are denoted as I, and I = {1, 2, ..., i, ... k}. The health service set I has different health service combinations X = {x1, x2, ..., x} according to different indicator analyses and certain topological structures (serial, parallel, selection and hybrid structures). i ,…x k},X∈Ω to complete, where x i represents the health service provided by the specific health service provider selected when completing subtask i, and Ω is the set of all feasible health service combinations to complete I. In general, service x i There are many different evaluation indicators, such as the time required for service, the cost of service, the reliability of service, etc. Therefore, for a service combination X, the embodiment of the present application also needs to use different evaluation indicators to measure and then select an optimal service combination. This problem is the health service combination optimization problem.

[0097] In one possible embodiment, the health service combination model uses T i 、C i ,η i Represents service x i The time, cost and efficacy required to complete, where η i The efficacy includes health services x i In terms of the treatment effect on all health signs, the user's health profile counts n health signs in total, so η i The efficacy of each dimension can be understood as the degree of positive impact on different health signs of users, and a new evaluation index, the health service cooperation level R, R ik Used to describe the candidate service x i 、x k The cooperative relationship between health services will indirectly affect many factors such as service cost and time. In order to improve the rationality of the model and make the results more feasible, the embodiment of this application establishes a health service combination model, as shown in the following formula:

[0098]

[0099] Where, α i , β i , γ i and θ i Represents T i 、C i ,η i With R ikRegarding the importance of sub-service i, for example, in the drug delivery service, time and cost are extremely important, but this health service does not have the indicator of efficacy, so a larger weight will be set for the cost and time of this service in the calculation of the total cost and total time of the service combination plan, and the weight of the efficacy indicator will be set to 0. Since the two health services of elderly entertainment and drug delivery are not closely related, and the cooperation between the two will not be too close, a lower weight will be set. For the two services with a greater correlation, drug delivery and Chinese medicine decoction, a larger weight will be set for the cooperation relationship between the two when calculating the cooperation relationship of the service combination plan. Among them, the degree of closeness will continue to change over time. In addition, using T max 、C max and η min They represent the longest time, maximum cost and minimum efficacy of the service plan that the user can accept, with R min Indicates the minimum level of cooperation between services. By solving the above model, we can get the user’s personalized service combination Y={y1,y2,L y i ,L y k},Y∈Ω,y i It indicates the health service provided by a specific service provider that is finally selected by the health service i.

[0100] One of the above technical solutions has the following advantages or beneficial effects: This embodiment of the application achieves personalized recommendations on the demand side by integrating user profiles and health service profiles, and optimizes based on factors such as the level of cooperation between the supply and demand sides. This framework overcomes the limitations of unilateral optimization, simultaneously considering the personalized needs of the demand side and the service capabilities and cooperation level of the supply side, achieving a precise match between supply and demand, and resolving the problem of mismatch between recommendation results and actual resource distribution in existing technologies, thereby achieving a global optimal solution.

[0101] In some embodiments, performing optimization and solving processing on the service composition model according to the population evolution algorithm to obtain the service composition includes the following steps:

[0102] Performing population initialization processing on the service composition model according to the population evolution algorithm to obtain a population;

[0103] Perform algorithm selection processing on the population according to the literature analysis method to construct an optimization algorithm comprehensive strategy pool;

[0104] The population is updated through the optimization algorithm comprehensive strategy pool according to the evaluation criteria to obtain the service combination.

[0105] In the examples of this application, please refer to Figure 4Initialize the population and related parameters, randomly generating an initial population where each individual represents a possible service combination solution. Relevant parameters such as population size, crossover probability, and mutation probability are also set. Fitness values are then calculated for each individual based on the service combination's evaluation metrics (such as time, cost, efficacy, and cooperation level). The fitness value reflects the quality of the service combination solution. A strategy is then selected to update the population, selecting appropriate operators (such as destroy and repair) from the strategy pool. The population is updated using an adaptive mechanism to control the operator selection probability. For example, operator selection weights are adjusted based on the evaluation criteria of the new solution, ensuring that the algorithm selects appropriate operators based on the optimization objective. The evolved population is then evaluated to determine whether it meets termination criteria, such as reaching a preset number of iterations or achieving convergence in fitness. If so, the optimal solution is obtained. If not, the algorithm proceeds to the fitness value calculation step.

[0106] Among them, the step of constructing a strategy pool fully considers the calculation methods, advantages and disadvantages, complementarity and diversity of different operators through literature analysis, and attempts to incorporate intelligent optimization algorithms such as the latest algorithms and classic algorithms into the strategy pool as operators to build a relatively complete comprehensive strategy pool of optimization algorithms. For example, in the traditional Chinese medicine pharmaceutical service combination optimization model, the strategy pool is equipped with a variety of destroy and repair operators for destroying and repairing the population, among which the destroy operator includes random destruction, destruction by priority, etc.; the repair operator includes random repair, repair by priority, etc. In addition, the embodiment of the present application establishes an adaptive mechanism to control the selection probability of the strategy pool operator, so that the algorithm selects relatively suitable operators to update the population around the optimization target. For example, in the traditional Chinese medicine pharmaceutical service combination optimization model, the algorithm is equipped with an evaluation criterion for the new solution as the weight for the selection of each operator for the next update, as shown in the following formula:

[0107]

[0108] Wherein ω1≥ω2≥ω3≥0 represent different weight values. The embodiment of the present application updates the population through the optimization algorithm integrated strategy pool, thereby obtaining the optimal service combination.

[0109] One of the above technical solutions has the following advantages or beneficial effects: This embodiment of the present application introduces a diverse adaptive mechanism and an individual update strategy pool, combined with a population evolution direction evaluation mechanism, to efficiently solve complex optimization problems and select the optimal service combination solution from a set of candidate services. This embodiment of the present application can effectively reduce the complexity of solving NP-hard problems, improve computational efficiency, meet real-time response requirements, and address the problem of low optimization efficiency caused by excessive computational complexity.

[0110] In some embodiments, optimizing the service combination according to the multi-criteria clustering model to obtain a recommendation result includes the following steps:

[0111] updating the service demand side data according to the service combination, and performing vector space mapping processing on the updated data and the service combination to obtain a data set;

[0112] Performing model building processing according to the intra-cluster data compactness index and the inter-cluster independence index to obtain the multi-criteria clustering model;

[0113] Performing clustering processing on the data set by using the multi-criteria clustering model to obtain clustering information;

[0114] The service combination is subjected to service optimization processing according to the clustering information to obtain the recommendation result.

[0115] In the embodiment of the present application, the service demand side data is updated according to the service combination, see Figure 5 In order to make full use of the information contained in the health service combination results, after completing the combination of health service set I, the user's health signs will inevitably be affected after receiving the combined health services, and a new health profile next_PU will be generated at this time. j , considering the impact of each health service provided by a specific health service provider in the combination on the user's health signs, that is, their efficacy, the user's new health profile is recalculated through the profile update formula. The profile update formula is shown as follows:

[0116]

[0117] In the formula, λ represents the weight coefficient, η i They represent the therapeutic effects required to complete the service.

[0118] In the embodiment of the present application, after updating the user health portrait, the updated user health portrait will replace the original health portrait. When the user health service is optimized next time, the updated portrait will affect the next health service optimization, which will make the next health service optimization more accurate. The entire process of health service optimization and combination will alternate and influence each other.

[0119] Then, the updated data and the service combination are processed into a vector space mapping process to obtain a data set. Based on the static and dynamic variables in the data set, a multi-criteria set is constructed, with different criteria for different types of variables. The criterion clustering model is used to independently select weight combinations, optimize multiple objectives at the same time, discover more complex data structures, and improve the accuracy of the clustering algorithm. Figure 6For data clustering, the data within the cluster should be compact, and there should be no overlapping areas between clusters, and each cluster should be independent. Based on this consideration, two indicators are used: the intra-cluster data compactness index (variance, VAR) and the inter-cluster independence index (connectivity, CNN). The calculation formula for the intra-cluster data compactness index is shown below:

[0120]

[0121] Where N is the number of data; C is the label of the current clustering scheme; c is the label of each cluster in the current scheme; υ c represents the centroid of cluster c. This formula measures the compactness of the data within the cluster by calculating the variance between each point in the cluster and the centroid, thereby observing the similarity of the subdivided population. The calculation formula for the inter-cluster independence index is as follows:

[0122]

[0123] Where L is a custom parameter, which means that the top L neighbors of a point participate in the calculation. When the sample size is too large, this parameter can be configured to improve the calculation efficiency and reduce computing resources; il is the point ranked l among the neighbors of point i. This formula measures the independence between clusters by determining whether each data point and its top-ranked neighbors are assigned to the same cluster. If not, a penalty value is used to measure the independence between clusters.

[0124] Based on the above indicators, a multi-objective clustering model is established. The multi-criteria clustering model is shown as follows:

[0125] minf(C)=(VAR(C),CNN(C))

[0126] The dataset is then clustered using a multi-criteria clustering model to generate cluster information. Based on this multi-criteria clustering model, an advanced adaptive intelligent optimization algorithm is employed to output the optimal cluster information. Using the information represented by the different dimensions of each cluster, service providers can determine the primary needs of different market segments and optimize their services accordingly. This clustering information allows for optimized service combinations and ultimately leads to recommendations.

[0127] One of the above technical solutions has the following advantages or beneficial effects: This embodiment dynamically updates the user profile based on the health service combination results, making subsequent health service recommendations more accurate. This mechanism not only improves the user experience but also enhances the accuracy and adaptability of system recommendations, ensuring that the system can continuously provide the optimal service combination solution based on changes in the user's health.

[0128] In some embodiments, clustering the data set using the multi-criteria clustering model to obtain clustering information includes the following steps:

[0129] Performing a minimum spanning tree generation process on the data set, and performing edge removal and random insertion processes on the minimum spanning tree according to a density index to obtain an initial clustering scheme;

[0130] Performing index calculation processing on the initial clustering schemes, and merging the initial clustering schemes according to the calculated indexes to obtain a target clustering scheme;

[0131] A population update process is performed on the target clustering scheme to obtain the clustering information.

[0132] In the examples of this application, please refer to Figure 7 , mapping health data and needs into a high-dimensional vector space to obtain a data set, and calculating the distance between data points (such as Euclidean distance, etc.), and performing normalization operations, generating a minimum spanning tree (MST) based on the prim algorithm, and then calculating the density index (DI value), and then selectively removing the top-ranked edges to form the initial cluster. The calculation formula of the density index is as follows:

[0133]

[0134] Among them, nn i (j) is the ranking of point j among the neighbors of point i, σ(i,j) represents the distance between point i and point j, σ max This is the maximum value among all distances.

[0135] Then, the VAR (intra-cluster data compactness index) and CNN (inter-cluster independence index) of the initial clustering scheme are calculated through the VAR and CNN definition formulas pre-set in the evaluator. At the same time, a ranking parameter L is customized to judge the top L neighbors of each data point. If the two do not belong to the same cluster, a cluster merge operation will be performed, and the values of VAR and CNN will be updated to obtain a final clustering scheme and its indicators. The edges selected and removed based on the DI value are randomly selected from all the data and inserted into the deleted parts, and the evaluator will judge the scheme each time to eventually form a clustering scheme set. The schemes in the set are used as the initial population, and based on the multi-criteria clustering model, an advanced adaptive intelligent optimization algorithm is used to output the best clustering information.

[0136] One of the above technical solutions has the following advantages or beneficial effects: The embodiment of the present application adopts a multi-criteria clustering model, optimizes the clustering effect through VAR (intra-cluster data compactness index) and CNN (inter-cluster independence index), and combines adaptive intelligent optimization technology to achieve health service optimization on both the supply and demand sides. This model can accurately identify the health service needs of different groups of people, provide optimization directions for service providers, solve the problem of missing multi-objective trade-offs caused by a single optimization goal, improve the comprehensive benefits of the health service combination, and provide service providers with accurate market insights, thereby improving resource allocation efficiency and market competitiveness.

[0137] The following is a detailed description of the embodiments of the present application with reference to specific application examples:

[0138] The present application embodiment can be applied to health service recommendation scenarios, please refer to Figure 8 , the embodiment of the present application constructs a PersonalRank algorithm that integrates health service users and health service profiles to achieve personalized health service recommendations for health service users, assist in completing the planning stage of health service combinations, preliminarily screen health services that match user needs, and generate a set of candidate services. Specifically, by constructing a health demander profile and a health service profile. Collect various health sign data (such as blood pressure, blood sugar, common chronic disease symptoms, etc.) and economic characteristics (such as disposable income, income stability, etc.) of the health demander to construct a health demander profile. Collect health service efficacy data (the treatment effect on different health signs of users) and cost data (long-term expenses, short-term expenses, etc.) to construct a health service profile. Then, initialize the degree of association, and calculate the initial degree of association between the health demander and the health service based on the health demander profile and the health service profile. For example, by calculating the similarity of the two portraits in the health sign dimension and the economic dimension, the initial degree of association is obtained. By integrating the degree of association with the random walk strategy, the initial degree of association is combined with the random walk strategy to form an improved random walk mechanism. During the random walk process, the probability of selecting the walk path is adjusted according to the degree of association, so that the walk is more inclined to the path with a high degree of association. Finally, a new random walk is executed to calculate the importance of health services. This improved random walk mechanism traverses the health services and calculates the importance of each one. The importance reflects the potential value of a health service to the health seeker. This generates and recommends a set of candidate services. Based on their importance, the top n health services are selected to form a candidate service set and recommended to the health seeker. This phase completes the planning of the health service portfolio and preliminarily selects health services that match user needs.

[0139] In the second-stage service combination model of the present application, the embodiment is responsible for further optimization and selecting the optimal service combination solution from the candidate service set. Since the second-stage service combination model of the health service combination is a complex optimization problem with NP-hard characteristics, classical mathematical solutions cannot solve this type of problem well. We introduce a variety of adaptive mechanisms in the intelligent optimization algorithm framework to cooperate with the individual update strategy pool, adaptively update the individual generation method, and combine the population evolution direction evaluation mechanism. Specifically, by initializing the population and related parameters. Randomly generate the initial population, each individual represents a possible service combination solution. At the same time, set relevant parameters such as population size, crossover probability, mutation probability, etc. Calculate the fitness value. According to the evaluation indicators of the service combination (such as time, cost, efficacy, cooperation level, etc.), calculate the fitness value of each individual. The fitness value reflects the pros and cons of the service combination solution. Select a strategy to update the population. Select appropriate operators (such as destroy and repair operators) from the strategy pool, control the selection probability of the operator according to the adaptive mechanism, and update the population. For example, adjust the operator selection weight according to the evaluation criteria of the new solution so that the algorithm selects the appropriate operator around the optimization goal. Whether the evolved population meets the termination conditions, such as reaching the preset number of iterations or fitness value convergence, if so, the optimal solution is obtained. Whether the evolved population meets the termination conditions, if not, go to the step of calculating the fitness value.

[0140] Finally, the embodiment of the present application constructs a clustering algorithm based on adaptive intelligent optimization technology to solve the health service optimization model for both supply and demand sides and obtain the optimal solution. By mapping health data and demand into a high-dimensional vector space, and calculating the distance between data points (such as Euclidean distance, etc.), and performing a normalization operation, a minimum spanning tree mst is generated based on the prim algorithm, and by calculating the DI value, the top-ranked edges are selected and removed accordingly to form an initial cluster. The VAR (intra-cluster data compactness index) and CNN (inter-cluster independence index) of the initial clustering scheme are calculated by pre-setting the VAR and CNN definition formulas in the evaluator. At the same time, a ranking parameter L is customized to judge the top L neighbors of each data point. If the two do not belong to the same cluster, a cluster merging operation will be performed, and the values of VAR and CNN will be updated to obtain a final clustering scheme and its indicators. The edges selected and removed based on the DI value are randomly selected from all data to be inserted into the deleted part, and each time the evaluator will judge the scheme, and finally form a clustering scheme set. Using the solutions in the collection as the initial population, we employ a multi-criteria clustering model and an advanced adaptive intelligent optimization algorithm to output optimal clustering information. By using the information represented by the different dimensions of each point in the cluster, service providers can determine the primary needs of different market segments and optimize their services accordingly.

[0141] See also Figure 9 The present application also provides a service recommendation system based on both supply and demand sides, which can implement the above-mentioned service recommendation method based on both supply and demand sides. The system includes:

[0142] The first module 901 is used to obtain service demand side data and service supply side data;

[0143] The second module 902 is configured to perform service recommendation processing on the target object on the service demand side according to the profile correlation degree between the service demand side data and the service supply side data, and obtain a candidate service set;

[0144] The third module 903 is configured to combine the candidate service sets according to a population evolution algorithm to obtain a service combination;

[0145] The fourth module 904 is configured to optimize the service combination according to a multi-criteria clustering model to obtain a recommendation result.

[0146] It can be understood that the contents of the above method embodiments are all applicable to the present system embodiments, the functions specifically implemented by the present system embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0147] The present application also provides an electronic device comprising a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the aforementioned method for recommending services based on both supply and demand sides. The electronic device can be any smart terminal, such as a tablet computer or an in-vehicle computer.

[0148] It can be understood that the contents of the above method embodiments are applicable to the present device embodiments, the functions specifically implemented by the present device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0149] See also Figure 10 , Figure 10 The hardware structure of an electronic device according to another embodiment is shown. The electronic device includes:

[0150] The processor 1001 can be implemented as a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of the present application.

[0151] The memory 1002 can be implemented in the form of a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 1002 can store an operating system and other application programs. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 1002 and is called by the processor 1001 to execute the service recommendation method based on both supply and demand sides of the embodiments of this application.

[0152] Input / output interface 1003, used to implement information input and output;

[0153] Communication interface 1004, used to implement communication interaction between this device and other devices, which can be achieved through wired means (such as USB, network cable, etc.) or wireless means (such as mobile network, WiFi, Bluetooth, etc.);

[0154] Bus 1005 , which transmits information between various components of the device (e.g., processor 1001 , memory 1002 , input / output interface 1003 , and communication interface 1004 );

[0155] The processor 1001 , the memory 1002 , the input / output interface 1003 and the communication interface 1004 are connected to each other in communication within the device via the bus 1005 .

[0156] An embodiment of the present application further provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the above-mentioned service recommendation method based on both sides of supply.

[0157] It can be understood that the contents of the above method embodiments are all applicable to the present storage medium embodiment, the functions specifically implemented by the present storage medium embodiment are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0158] The memory, as a non-transient computer-readable storage medium, can be used to store non-transient software programs and non-transient computer executable programs. In addition, the memory may include a high-speed random access memory and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some embodiments, the memory may optionally include a memory remotely arranged relative to the processor, and these remote memories may be connected to the processor via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0159] The embodiments of the present application provide a service recommendation method, system, device and medium based on both the supply and demand sides. The solution recommends services to target objects on the service demand side according to the degree of correlation between the portraits of the service demand side data and the service supply side data to obtain a candidate service set. The portraits of both the supply and demand sides can be analyzed to make personalized recommendations, thereby improving the accuracy of service matching. In addition, the solution combines the candidate service sets according to the population evolution algorithm to obtain a service combination, and can select the optimal service combination solution from the candidate service set, thereby reducing the solution complexity and improving the computational efficiency. In addition, the solution optimizes the service combination according to the multi-criteria clustering model to obtain a recommendation result, which can optimize the health services on both the supply and demand sides and improve the user experience.

[0160] The embodiments described in the embodiments of this application are intended to more clearly illustrate the technical solutions of the embodiments of this application and do not constitute a limitation on the technical solutions provided by the embodiments of this application. Those skilled in the art will appreciate that with the evolution of technology and the emergence of new application scenarios, the technical solutions provided in the embodiments of this application are also applicable to similar technical problems.

[0161] Those skilled in the art will understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of the present application, and may include more or fewer steps than shown in the figures, or a combination of certain steps, or different steps.

[0162] The system embodiment described above is merely illustrative. The units described as separate components may or may not be physically separate, i.e., they may be located in one place or distributed across multiple network units. Some or all of the modules may be selected based on actual needs to achieve the objectives of this embodiment.

[0163] Those skilled in the art will appreciate that all or some of the steps in the methods, systems, and functional modules / units in the devices disclosed above may be implemented as software, firmware, hardware, or appropriate combinations thereof.

[0164] The terms "first", "second", "third", "fourth", etc. (if any) in the specification of the present application and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequential order. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0165] It should be understood that in this application, "at least one (item)" means one or more, and "plurality" means two or more. "And / or" is used to describe the association relationship of associated objects, indicating that three relationships may exist. For example, "A and / or B" can mean: only A exists, only B exists, and A and B exist at the same time, where A and B can be singular or plural. The character " / " generally indicates that the previous and next associated objects are in an "or" relationship. "At least one of the following items" or similar expressions refers to any combination of these items, including any combination of single items or plural items. For example, at least one of a, b or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.

[0166] In the several embodiments provided in this application, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the system embodiments described above are merely illustrative. For example, the division of the above units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of systems or units, which can be electrical, mechanical or other forms.

[0167] The units described above as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0168] In addition, the functional units in the various embodiments of the present application may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0169] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium and includes multiple instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of various embodiments of the present application. The aforementioned storage medium includes: various media that can store programs, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0170] The preferred embodiments of the present invention are described above with reference to the accompanying drawings, but are not intended to limit the scope of the present invention. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and essence of the present invention should be within the scope of the present invention.

Claims

1. A service recommendation method based on both supply and demand sides, characterized in that: The method comprises the following steps: Obtain service demand-side data and service supply-side data; Performing service recommendation processing on the target object on the service demand side according to the profile correlation degree between the service demand side data and the service supply side data to obtain a candidate service set; Combining the candidate service sets according to a population evolution algorithm to obtain a service combination; The service combination is optimized according to a multi-criteria clustering model to obtain a recommendation result.

2. The method according to claim 1, characterized in that The process of performing service recommendation processing on the target object on the service demand side according to the profile correlation degree between the service demand side data and the service supply side data to obtain a candidate service set includes the following steps: An object portrait is constructed based on the service demand side data, and a service portrait is constructed based on the service supply side data; Performing correlation analysis on the health attributes and economic attributes of the object portrait and the service portrait to obtain a correlation degree between the portraits; The services provided by the service supply side are recommended to the target object according to the degree of association of the portraits to obtain the candidate service set.

3. The method according to claim 2, characterized in that The process of recommending the services provided by the service supply side to the target object based on the portrait association degree to obtain the candidate service set includes the following steps: The services provided by the service supply side are traversed according to the random walk algorithm, and the selection probability of the walk path is adjusted according to the correlation degree of the portrait, and the service importance is calculated using the importance calculation formula; Selecting services provided by the service supply side according to the service importance to obtain the candidate service set; The candidate service set is recommended to the target object.

4. The method according to claim 1, wherein The process of combining the candidate service sets according to the population evolution algorithm to obtain a service combination includes the following steps: Selecting the candidate service set through the target object to obtain a target service set; Performing index analysis and topology structure combination processing on the target service set to obtain a service composition model; The service composition model is optimized and solved according to the population evolution algorithm to obtain the service composition.

5. The method according to claim 4, characterized in that The optimizing and solving the service composition model according to the population evolution algorithm to obtain the service composition includes the following steps: Performing population initialization processing on the service composition model according to the population evolution algorithm to obtain a population; Perform algorithm selection processing on the population according to the literature analysis method to construct an optimization algorithm comprehensive strategy pool; The population is updated through the optimization algorithm comprehensive strategy pool according to the evaluation criteria to obtain the service combination.

6. The method according to claim 1, characterized in that Optimizing the service combination according to the multi-criteria clustering model to obtain a recommendation result includes the following steps: updating the service demand side data according to the service combination, and performing vector space mapping processing on the updated data and the service combination to obtain a data set; Performing model building processing according to the intra-cluster data compactness index and the inter-cluster independence index to obtain the multi-criteria clustering model; Performing clustering processing on the data set by using the multi-criteria clustering model to obtain clustering information; The service combination is subjected to service optimization processing according to the clustering information to obtain the recommendation result.

7. The method according to claim 6, characterized in that The clustering process of the data set by the multi-criteria clustering model to obtain clustering information includes the following steps: Performing a minimum spanning tree generation process on the data set, and performing edge removal and random insertion processes on the minimum spanning tree according to a density index to obtain an initial clustering scheme; Performing an index calculation process on the initial clustering schemes, and merging the initial clustering schemes according to the calculated indexes to obtain a target clustering scheme; A population update process is performed on the target clustering scheme to obtain the clustering information.

8. A service recommendation system based on both supply and demand sides, characterized in that: The system comprises: The first module is used to obtain service demand side data and service supply side data; The second module is used to perform service recommendation processing on the target object on the service demand side according to the profile correlation degree between the service demand side data and the service supply side data, and obtain a candidate service set; The third module is used to combine the candidate service sets according to the population evolution algorithm to obtain a service combination; The fourth module is used to optimize the service combination according to a multi-criteria clustering model to obtain a recommendation result.

9. An electronic device, characterized in that: The electronic device includes a memory and a processor, the memory stores a computer program, and the processor implements the method according to any one of claims 1 to 7 when executing the computer program.

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

Images