Network site selection recommendation method, device and electronic equipment

Abstract

The method, device and electronic device for recommending network site selection provided by the embodiment of the present invention include: obtaining various evaluation data and expected business indicator data of network sites in multiple reference areas, and using these data to train multiple weight allocation models to evaluate the importance of various evaluation data in network site selection, and then using the importance of various evaluation data to score the importance of each reference area, and taking the area with a score higher than a preset threshold as the final recommended area. The embodiment of the present invention uses a large amount of actual data to train the weight allocation model, which can reduce subjective factors, quickly and accurately evaluate the impact of various evaluation data in network site selection, provide a scientific basis for network site selection, and improve the accuracy of network site selection.

Description

Technical Field

[0001] The embodiments of the present invention relate to the field of data processing technology, and in particular, to a network site selection recommendation method, device and electronic equipment. Background Art

[0002] Outlets generally refer to the grid system of various business organizations set up by enterprises or service organizations in the region to better serve customers, in order to support business and services. It is an important part of the radiation capacity and customer contact channels of enterprises or organizations. Reasonable outlet layout helps enterprises or service organizations quickly cover a wider user base and strengthen connections with users.

[0003] Traditional methods of selecting network sites are easily affected by subjective factors, are inefficient, and the results are not accurate enough, which cannot meet the increasing requirements of modern service agencies for efficiency and accuracy. Therefore, how to improve the accuracy and reliability of network site selection is a technical problem that needs to be solved. Summary of the invention

[0004] One of the purposes of the embodiments of the present invention is to provide a method, device and electronic device for recommending network site selection, which uses multi-source data to train a weight distribution model, and quickly completes the importance analysis of various evaluation data and their corresponding reference areas for site selection through the model, providing a scientific and operational basis for subsequent network site selection decisions, and improving the accuracy of site selection. The embodiments of the present invention can be implemented as follows:

[0005] In the first aspect, an embodiment of the present invention provides a method for network site selection recommendation, the method comprising: obtaining a multi-source evaluation data set and a network point expected business indicator data set; wherein the multi-source evaluation data set includes various evaluation data corresponding to each reference area; the network point expected business indicator data set includes network point expected business indicator data corresponding to each of the reference areas; using the multi-source evaluation data set and the network point expected business indicator data set to train multiple weight allocation models, and then using the weight allocation model to assign importance to the various evaluation data of the reference area; determining the importance score of each of the reference areas based on the multi-source evaluation data set after the importance assignment; and using the reference area whose importance score exceeds a preset threshold as the network site selection area.

[0006] In an optional embodiment, the weight allocation model is used to assign importance to various evaluation data of the reference area, including: determining the accuracy of each weight allocation model; using the accuracy to determine the assignment coefficient corresponding to each weight allocation model; obtaining the feature weights corresponding to the various evaluation data output by each weight allocation model after training; and weighting the feature weights of each evaluation data under each weight allocation model through the assignment coefficient of the weight allocation model to complete the importance assignment.

[0007] In an optional implementation, using the accuracy to determine the assignment coefficient corresponding to each of the weight distribution models includes: taking the ratio of the accuracy of each of the weight distribution models to the sum of all the accuracies as the assignment coefficient.

[0008] In an optional embodiment, determining the accuracy of each of the weight allocation models includes: testing each of the weight allocation models using various evaluation data corresponding to a portion of the reference area that has not participated in the model training and the expected business indicator data of the outlets, to obtain the predicted expected business indicator data of the outlets output by each of the weight allocation models; and determining the accuracy of each of the weight allocation models using the predicted expected business indicator data of the outlets and the expected business indicator data of the outlets.

[0009] In an optional implementation, the accuracy of each of the weight allocation models is determined using the predicted network point expected business indicator data and the network point expected business indicator data, including: determining the prediction error between the predicted network point expected business indicator data and the network point expected business indicator data and the overall true value corresponding to all the network point expected business indicator data; taking the ratio of the prediction error to the overall true value as the prediction error rate, and subtracting the prediction error rate from one to obtain the accuracy.

[0010] In an optional embodiment, multiple weight allocation models are trained using the multi-source evaluation data set and the expected business indicator data set for the outlets, including: preprocessing each of the multi-source evaluation data set; selecting the various evaluation data and the expected business indicator data for the outlets of a part of the reference area, training different machine learning models, and obtaining multiple weight allocation models.

[0011] In an optional embodiment, the importance score of each reference area is determined based on the multi-source evaluation data set after importance assignment, including: for each reference area, determining the sum of the various evaluation data after importance assignment as the importance score of the reference area.

[0012] In a second aspect, an embodiment of the present invention provides a network site selection recommendation device, comprising: an acquisition module, used to acquire a multi-source evaluation data set and a network point expected business indicator data set; wherein the multi-source evaluation data set includes various evaluation data corresponding to each reference area; the network point expected business indicator data set includes network point expected business indicator data corresponding to each reference area; a training module, used to train multiple weight allocation models using the multi-source evaluation data set and the network point expected business indicator data set; an assignment module, used to assign importance to the various evaluation data of the reference area using the weight allocation model; a determination module, used to determine the importance score of each reference area based on the multi-source evaluation data set after the importance assignment, and use the reference area with an importance score exceeding a preset threshold as the network site selection area.

[0013] In an optional embodiment, the assignment module is specifically used to: determine the accuracy of each of the weight allocation models; use the accuracy to determine the assignment coefficient corresponding to each of the weight allocation models; obtain the feature weights corresponding to the various evaluation data output by each of the weight allocation models after training; and weight the feature weights of each of the evaluation data under each of the weight allocation models through the assignment coefficient of the weight allocation model to complete the importance assignment.

[0014] In a third aspect, an embodiment of the present invention provides an electronic device, including a processor and a memory, wherein the memory stores a computer program executable by the processor, and the processor can execute the computer program to implement the network site recommendation method as described above.

[0015] The method, device and electronic device for recommending network site selection provided by the embodiment of the present invention first obtain a multi-source evaluation data set and a network point expected business indicator data set, use these data as training data, and train multiple weight allocation models. In the process of model training, the complex relationship between each evaluation data and the network point business performance can be mined. Then, each weight allocation model assigns an importance value to each evaluation data according to the training result. The higher the importance value, the greater the contribution of the corresponding evaluation data to the evaluation result of the reference area, that is, the greater the reference significance for the network point site selection in the area. Finally, according to the importance value obtained from each evaluation data, the importance score of each reference area for the network point site selection is determined. This provides a basis for subsequent site selection decisions. The embodiment of the present invention adopts a data-driven method to reduce the influence of subjective factors, and quickly analyzes the importance of each evaluation data and the corresponding area through a model, providing a scientific and operational basis for network point site selection decisions, which is conducive to improving the accuracy of network point site selection. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0017] Figure 1 A schematic flow chart of a network site selection recommendation method provided by an embodiment of the present invention;

[0018] Figure 2 An example diagram of an evaluation data matrix provided by an embodiment of the present invention;

[0019] Figure 3 A schematic flow chart of step S103 provided in an embodiment of the present invention;

[0020] Figure 4 A functional module diagram of a network site selection recommendation device 40 provided in an embodiment of the present invention;

[0021] Figure 5 This is a structural block diagram of an electronic device 50 provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0022] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of them. Generally, the components of the embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.

[0023] Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the embodiments of the present invention claimed for protection, but merely represents selected embodiments of the embodiments of the present invention. All other embodiments obtained by ordinary technicians in this field without creative work based on the embodiments in the embodiments of the present invention are within the scope of protection of the embodiments of the present invention.

[0024] It should be noted that similar reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined in one drawing, further definition and explanation thereof is not required in subsequent drawings.

[0025] In the description of the embodiments of the present invention, it should be noted that if the terms "upper", "lower", "inside", "outside", etc. appear, the orientation or position relationship indicated is based on the orientation or position relationship shown in the drawings, or is the orientation or position relationship in which the product of the invention is usually placed when used. It is only for the convenience of describing the embodiments of the present invention and simplifying the description, and does not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation. Therefore, it cannot be understood as a limitation on the embodiments of the present invention.

[0026] In addition, the terms “first”, “second”, etc., if used, are merely used to distinguish between the descriptions and should not be understood as indicating or implying relative importance.

[0027] It should be noted that, in the absence of conflict, the features in the embodiments of the present invention may be combined with each other.

[0028] See also Figure 1 , Figure 1 This is a schematic flow chart of a method for recommending network site selection provided by an embodiment of the present invention. The execution subject of the method may be an electronic device with data processing capability. The following steps may be performed by the electronic device:

[0029] S101, obtaining a multi-source evaluation data set and a network point expected business indicator data set; wherein the multi-source evaluation data set includes various evaluation data corresponding to each reference area; the network point expected business indicator data set includes network point expected business indicator data corresponding to each reference area;

[0030] S102: multiple weight allocation models are trained using multi-source evaluation data sets and network point expected business indicator data sets, and then the weight allocation models are used to assign importance to various evaluation data in the reference area.

[0031] S103: Determine the importance score of each reference region according to the multi-source evaluation data set after the importance is assigned.

[0032] S104: Using reference areas whose importance scores exceed a preset threshold as network site selection areas.

[0033] In the scheme of step S101 to step S104, firstly, a multi-source evaluation data set and a network point expected business indicator data set are obtained, and these data are used as training data to train multiple weight allocation models. In the process of model training, the complex relationship between each evaluation data and the network point business performance can be mined. Then, each weight allocation model assigns an importance value to each evaluation data according to the training result. The higher the importance value, the greater the contribution of the corresponding evaluation data to the evaluation result of the reference area, that is, the greater the reference significance for the network point site selection in the area. Finally, according to the importance value obtained from each evaluation data, the importance score of each reference area for the network point site selection is determined. This provides a basis for subsequent site selection decisions. The embodiment of the present invention adopts a data-driven method to reduce the influence of subjective factors, and quickly analyzes the importance of each evaluation data and the corresponding area through a model, providing a scientific and operational basis for network point site selection decisions, which is conducive to improving the accuracy of network point site selection.

[0034] Next, the embodiment of the present invention will Figure 1 Each step shown is described in detail.

[0035] In step S101, "outlets" can be business outlets or service outlets established by enterprises or service agencies in various places. Service agencies can be, but are not limited to: banks, supermarkets, shops, etc. The target research area can be an administrative division such as a county, town, etc., or a specific street or landmark surrounding area. It can be flexibly set according to actual needs and is not limited here. The determination of each reference area can refer to previous business experience and pilot survey results, and roughly divide some possible location categories of outlets from the target research area, such as business district areas, residential areas, etc. Within the final determined location category, several specific outlet addresses where outlets may be set can be further divided.

[0036] The multi-source evaluation data set contains various evaluation index data collected from different reference areas. These evaluation index data include not only the demographic characteristics of the region and the outside, but also comprehensive factors such as the business environment and surrounding competition. By measuring these evaluation index data, the ecological environment quality and operating environment conditions in each reference area can be evaluated, which is an important reference for selecting the location of the outlet. Using these multi-source evaluation data, a comprehensive evaluation of each reference area can be carried out from multiple aspects, providing a reference for selecting the best outlet.

[0037] In an optional implementation, the types of evaluation data include, but are not limited to, commercial, transportation, population, and peer categories. For example, commercial evaluation data may include, but are not limited to, the number of surrounding shopping malls, restaurants, office buildings, residences, hotels, subway stations, and bus stations; population evaluation data may include, but are not limited to, the number of non-working population, the number of working population, the number of people with affluent incomes among the non-working population and the working population, the number of app users of the institution to be studied, the number of users of other similar service agencies among the non-working population, etc.

[0038] It is understandable that users can obtain corresponding evaluation data according to the actual needs of different types of service agencies, which is not limited here.

[0039] It should be noted that the various evaluation data collected during the execution of the scheme involved in the above examples of the present invention, including data related to users, must be explicitly agreed to by the user before they can be applied to the scheme of the above embodiments of the present application.

[0040] Next, the embodiment of the present invention will be introduced by taking the site selection of a bank branch as an example. The site selection of branches of other types of service agencies is similar to that of banks, and will not be described in detail in the embodiment of the present invention.

[0041] In an optional implementation, the multi-source evaluation data set obtained in step S101 can be represented by a data matrix X, and the expected business indicator data of the outlets can be represented by a data matrix Y. Then the matrix forms of X and Y are as follows:

[0042]

[0043]

[0044] Among them, m is the total number of reference areas. One network site is selected in each reference area, so m can also be the total number of networks to be selected; n is the total number of evaluation data.

[0045] For ease of understanding, take Bank of China as an example, see Figure 2 , Figure 2 An example diagram of the evaluation data matrix provided for an embodiment of the present invention. Assuming that sites for five outlets of China Merchants Bank are to be selected, the multi-source evaluation data set may include: the number of Bank of Communications, China Construction Bank, Industrial and Commercial Bank of China, Agricultural Bank of China, bus stops, subway stations, hotels, the number of working people with affluent incomes, the number of Bank of China users, the number of Bank of China users among the household population, the number of surrounding shopping malls, the number of restaurants, the number of office buildings, the number of residences, and the number of public transportation.

[0046] The evaluation data and operating indicator values ​​of each branch of the bank are managed in a unified manner through the above-mentioned matrix form. This can not only monitor the changing trends of the performance indicators of each branch in real time, but also facilitate the subsequent rapid selection of training data and verification data.

[0047] In step S102, the purpose of the embodiment of the present invention is to train multiple weight distribution models through the data obtained in step S101. Therefore, the implementation of step S102 provided by the embodiment of the present invention includes the following steps:

[0048] Step a1: Preprocess the multi-source evaluation dataset.

[0049] Step a2: Select various evaluation data and expected business indicator data of some reference areas, train different machine learning models, and obtain multiple weight allocation models.

[0050] In the embodiment of the present invention, all evaluation data are preprocessed, which can be understood as performing a standardization operation on various evaluation data to ensure that the indicator values ​​are dimensionless.

[0051] Before training, the order of various evaluation data corresponding to each reference area can be shuffled, and then the multi-source data set can be divided according to the proportion of the reference area to obtain the evaluation data participating in the model training, that is, the training samples. The data that does not participate in the model training is used as the verification data for the subsequent verification of the accuracy of the weight allocation model obtained by the training.

[0052] In an optional implementation, the ratio of training samples to verification data can be customized according to actual needs. For example, the training samples account for 70% of the total data, and the verification data accounts for 30%, which is not limited here.

[0053] During the training process, the machine learning algorithms that can be selected include but are not limited to: linear regression, lasso, elastic network, random forest, xgboost, lightgbm, catboost, neural network, etc. Users can determine multiple machine learning algorithms based on the number of weight allocation models actually required, and then use training samples for training to obtain multiple weight allocation models, which is convenient for subsequent mining of the complex relationship between evaluation data and network performance.

[0054] In an optional implementation, in order to determine the importance of each evaluation data to the site selection of the network, the weights of various evaluation data can be determined. The current method of determining weights is easily affected by subjective factors, resulting in inaccurate results and affecting the accuracy of the site selection of the network. For example, the expert weighting method is easily affected by subjective factors, which may lead to inconsistent or biased weight distribution; the AHP technology requires the establishment of a relatively complex hierarchical model, and there are also problems of subjectivity and consistency; the principal component analysis method may ignore certain important influencing factors to a certain extent and cannot fully reflect the complex relationships between the various indicators; the fuzzy comprehensive evaluation method is affected by fuzzy mathematical methods when processing fuzzy relationships and determining weights, and is relatively complex to handle. In order to solve this problem, the embodiment of the present invention provides a solution in step S103, that is, the complex relationship between the evaluation data and the network performance is mined through the multiple weight distribution models obtained in step S102, so as to determine the importance of each evaluation data to the network performance.

[0055] Therefore, see Figure 3 , Figure 3 The schematic flow chart of step S103 provided in the embodiment of the present invention includes the following steps:

[0056] S103-1: Determine the accuracy of each weight allocation model.

[0057] The accuracy of the weight distribution model can reflect the ability of the model to mine the complex relationship between the evaluation data and the performance of the outlets. Generally, the higher the accuracy, the more it can mine the deep complex relationship. In the embodiment of the present invention, the importance of various evaluation data is assigned by analyzing the accuracy of the model, and the result is more accurate.

[0058] The embodiment of the present invention provides a method for determining the accuracy of a weight distribution model, including:

[0059] Step b1: Use various evaluation data corresponding to some reference areas that do not participate in model training and expected business indicator data of outlets to test each weight allocation model, and obtain the predicted expected business indicator data of outlets output by each weight allocation model.

[0060] From the foregoing content, it can be known that before model training, the embodiment of the present invention segments part of the data from the multi-source evaluation data set as verification data, so in the process of determining the accuracy of the model, the verification data can be used to verify the accuracy.

[0061] Step b2: Determine the accuracy of each weight allocation model using the predicted network point expected business indicator data and the network point expected business indicator data.

[0062] In the embodiment of the present invention, the prediction error between the predicted network point expected business indicator data and the network point expected business indicator data and the overall true value corresponding to all network point expected business indicator data can be determined first; the ratio of the prediction error to the overall true value is used as the prediction error rate, and the accuracy is obtained by subtracting the prediction error rate from one. Therefore, the accuracy calculation relationship of each weight allocation model is as follows:

[0063]

[0064] Among them, Pred represents the predicted expected business indicator data of the outlets, True represents the actual expected business indicator data of the outlets; ∑|Pred-True| is the prediction error between Pred and True; ∑True represents the overall true value, and finally the ratio of the prediction error to the overall true value is taken as the prediction error rate. Forecast error rate.

[0065] S103-2: Determine the assignment coefficient corresponding to each weight distribution model using the accuracy.

[0066] In the embodiment of the present invention, the calculation relationship of the assignment coefficient corresponding to each weight distribution model is as follows:

[0067]

[0068] Among them, coefficient i and Accuracy i are the assignment coefficient and accuracy of the i-th weight assignment model respectively; k is the number of weight assignment models.

[0069] S103-3: Obtain feature weights corresponding to various evaluation data output by each weight allocation model after training.

[0070] In the embodiment of the present invention, the weight distribution model implements multiple iterations of training by updating the weights during the training process, and the weights are fixed at the end of the training to obtain the final model. At this time, each model generates a column vector of feature weights, denoted as w=[w 11 , w 12 , w 13 ,…,w 1n ] T , n is the total number of evaluation data, each weight value corresponds to an evaluation data, and the feature weight column vectors output by the k weight distribution models can form an n×k feature weight matrix, denoted as W n×k , W n×k As shown below:

[0071]

[0072] In the above feature weight matrix, each column represents the feature weight corresponding to each of the n types of evaluation data output by a weight allocation model at the end of training. Through the assignment coefficient corresponding to each weight allocation model, the importance of the feature weights of the n evaluation data under each weight allocation model can be assigned, see S103-4.

[0073] S103-4: The feature weight of each evaluation data under each weight allocation model is weighted by the assignment coefficient of the weight allocation model to complete the importance assignment.

[0074] The feature weight matrix can be weighted averaged by the assignment coefficient of each weight distribution model, so as to complete the importance assignment of each evaluation data. The above step S103-4 can be expressed as follows:

[0075]

[0076] Among them, W1 to W n The importance values ​​of the 1st to nth evaluation data are assigned respectively.

[0077] The above implementation method quantifies the influence of various evaluation indicators on the performance of the dots, and then the importance score of each reference area can be determined in step S104.

[0078] In step S104, the sum of all evaluation data after importance assignment may be determined as the importance score of each reference region. The importance score calculation formula of each reference region is as follows:

[0079]

[0080] Among them, Score1 to Score m They correspond to the importance scores of the 1st to the mth reference areas respectively, which will provide a reference for the subsequent network site selection.

[0081] Finally, the reference area with importance score exceeding the preset threshold can be used as the network site selection area, wherein the preset threshold can be set according to actual needs. For example, if the preset threshold is 8, the reference area with importance score exceeding 8 is used as the best network site selection area.

[0082] Based on Figure 1 For the same inventive concept, see Figure 4 , Figure 4 The functional module diagram of the network site selection recommendation device 40 provided in the embodiment of the present invention includes: an acquisition module 401, a training module 402, an assignment module 403 and a determination module 404.

[0083] The acquisition module 401 is used to acquire a multi-source evaluation data set and a network point expected business indicator data set; wherein the multi-source evaluation data set includes various evaluation data corresponding to each reference area; the network point expected business indicator data set includes network point expected business indicator data corresponding to each reference area;

[0084] A training module 402 is used to train multiple weight allocation models using a multi-source evaluation data set and a network point expected business indicator data set;

[0085] The value assignment module 403 is used to assign importance to various evaluation data of the reference area by using the weight distribution model;

[0086] The determination module 404 is used to determine the importance score of each reference area according to the multi-source evaluation data set after the importance is assigned, and use the reference area whose importance score exceeds a preset threshold as the network point location area.

[0087] It is understandable that the acquisition module 401, the training module 402, the assignment module 403 and the determination module 404 can be executed in coordination. Figure 1 Each step in the process is performed to achieve the corresponding technical effect.

[0088] In an optional implementation, the assignment module 403 is specifically used to determine the accuracy of each weight allocation model; use the accuracy to determine the assignment coefficient corresponding to each weight allocation model; obtain the feature weights corresponding to various evaluation data output by each weight allocation model after training; and weight the feature weights of each evaluation data under each weight allocation model through the assignment coefficient of the weight allocation model to complete the importance assignment.

[0089] In an optional implementation, the assignment module 403 is further specifically configured to use the ratio of the accuracy of each weight distribution model to the sum of all accuracies as an assignment coefficient.

[0090] In an optional implementation, the assignment module 403 is further specifically used to test each weight allocation model using various evaluation data corresponding to some reference areas that have not participated in the model training and the expected business indicator data of the outlets, to obtain the predicted expected business indicator data of the outlets output by each weight allocation model; and to determine the accuracy of each weight allocation model using the predicted expected business indicator data of the outlets and the expected business indicator data of the outlets.

[0091] In an optional implementation, the assignment module 403 is also specifically used to determine the prediction error between the predicted network point expected business indicator data and the network point expected business indicator data and the overall true value corresponding to all network point expected business indicator data; the ratio of the prediction error to the overall true value is used as the prediction error rate, and the accuracy is obtained by subtracting the prediction error rate from one.

[0092] In an optional implementation, the training module 402 is specifically used to preprocess each multi-source evaluation data set; select various evaluation data of some reference areas and expected business indicator data of outlets, train different machine learning models, and obtain multiple weight allocation models.

[0093] In an optional implementation, the determination module 404 is specifically configured to determine, for each reference region, the sum of various evaluation data after the importance is assigned as the importance score of the reference region.

[0094] It should be noted that the division of modules in the above embodiments of the present application is schematic and is only a logical function division. There may be other division methods in actual implementation. In addition, each functional unit in each embodiment of the present application may be integrated into a processing unit, or may exist physically separately, or two or more units may be integrated into one unit. The above integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0095] 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 is essentially 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, and the computer software product is stored in a storage medium, including a number of instructions to enable an electronic device (which can be a personal computer, server, or network device, etc.) or a processor (processor) to perform all or part of the steps of each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, ROM), random access memory (Random Access Memory, RAM), disk or optical disk and other media that can store program code.

[0096] Based on the above embodiments, the present application also provides an electronic device, see Figure 5 , Figure 5 This is a structural block diagram of an electronic device 50 provided in an embodiment of the present invention, and the electronic device 50 is used to execute the network site selection recommendation method provided in an embodiment of the present invention, and the electronic device 50 includes: a memory 501, a processor 502, a communication interface 503, and a bus 504. The memory 501, the processor 502, and the communication interface 503 are directly or indirectly electrically connected to each other to realize data transmission or interaction. For example, these components can be electrically connected to each other through one or more communication buses or signal lines.

[0097] Optionally, bus 504 may be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 5 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0098] In the embodiment of the present application, the processor 502 may be a general-purpose processor, a digital signal processor, an application-specific integrated circuit, a field programmable gate array or other programmable logic device, a discrete gate or transistor logic device, or a discrete hardware component, and may implement or execute the methods, steps, and logic block diagrams disclosed in the embodiment of the present application. The general-purpose processor may be a microprocessor or any conventional processor, etc. The steps of the method disclosed in the embodiment of the present application may be directly embodied as being executed by a hardware processor, or may be executed by a combination of hardware and software modules in the processor. The software module may be located in the memory 501, and the processor 502 reads the program instructions in the memory 501, and completes the steps of the above method in combination with its hardware.

[0099] In the embodiment of the present application, the memory 501 may be a non-volatile memory, such as a hard disk (HDD) or a solid-state drive (SSD), etc., or a volatile memory (VolatileMemory), such as RAM. The memory may also be any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto. The memory in the embodiment of the present application may also be a circuit or any other device that can implement a storage function, for storing instructions and / or data.

[0100] The memory 501 can be used to store software programs and modules, such as the instructions / modules of the network site selection recommendation device 40 provided in the embodiment of the present invention, which can be stored in the memory 501 in the form of software or firmware (Firmware) or fixed in the operating system (OS) of the electronic device 500. The processor 502 executes various functional applications and data processing by executing the software programs and modules stored in the memory 501. The communication interface 503 can be used to communicate signaling or data with other node devices.

[0101] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described devices and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0102] Understandably, Figure 5 The structure shown is for illustration only. The electronic device 50 may also include Figure 5 More or fewer components as shown, or with Figure 5 Different configurations shown. Figure 5 The components shown may be implemented in hardware, software or a combination thereof.

[0103] The electronic device 50 can be any electronic product that can perform human-computer interaction with a user, such as a personal computer, a tablet computer, a smart phone, a personal digital assistant (PDA), a game console, an interactive network television (IPTV), a smart wearable device, etc.

[0104] The electronic device 50 may also include a network device and / or a user device. The network device includes, but is not limited to, a single network server, a server group consisting of multiple network servers, or a cloud consisting of a large number of hosts or network servers based on cloud computing.

[0105] The network where the electronic device 50 is located includes but is not limited to the Internet, a wide area network, a metropolitan area network, a local area network, a virtual private network (VPN), etc.

[0106] Based on the above embodiments, the present application further provides a storage medium, in which a computer program is stored. When the computer program is executed by a computer, the computer executes the network site selection recommendation method provided in the above embodiments.

[0107] Based on the above embodiments, the embodiments of the present application further provide a computer program, which, when executed on a computer, enables the computer to execute the network site selection recommendation method provided in the above embodiments.

[0108] Based on the above embodiments, the embodiments of the present application further provide a chip, which is used to read a computer program stored in a memory and to execute the network site selection recommendation method provided in the above embodiments.

[0109] The present application also provides a computer program product in an embodiment, including instructions, which, when executed on a computer, enable the computer to execute the network site selection recommendation method provided in the above embodiment.

[0110] The embodiments of the present application are described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by instructions. These instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0111] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0112] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0113] The above are only specific implementations of the embodiments of the present invention, but the protection scope of the embodiments of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in the embodiments of the present invention should be included in the protection scope of the embodiments of the present invention. Therefore, the protection scope of the embodiments of the present invention shall be subject to the protection scope of the claims.

Claims

1. A network site selection recommendation method, characterized in that: The method comprises: Acquire a multi-source evaluation data set and a network point expected business indicator data set; wherein the multi-source evaluation data set includes various evaluation data corresponding to each reference area; the types of the evaluation data include: commercial type, transportation type, population type and peer type; the network point expected business indicator data set includes network point expected business indicator data corresponding to each reference area; Training a plurality of weight allocation models using the multi-source evaluation data set and the network point expected business indicator data set, and then using the weight allocation model to assign importance to various evaluation data of the reference area; The method further comprises: preprocessing the multi-source evaluation data set; selecting the various evaluation data and the expected business indicator data of the network point in a part of the reference area, training different machine learning models, and obtaining the plurality of weight allocation models; Then, the weight distribution model is used to assign importance to various evaluation data of the reference area, including: determining the accuracy of each weight distribution model; determining the assignment coefficient corresponding to each weight distribution model using the accuracy; obtaining the feature weights corresponding to various evaluation data output by each weight distribution model after training; and weighting the feature weights of each evaluation data under each weight distribution model by the assignment coefficient of the weight distribution model to complete the importance assignment; Determining an importance score for each of the reference regions according to the multi-source evaluation data set after importance assignment; The reference area whose importance score exceeds a preset threshold is used as a network site selection area.

2. The network site selection recommendation method according to claim 1, characterized in that: Determining the assignment coefficient corresponding to each of the weight distribution models using the accuracy includes: The ratio of the accuracy of each weight distribution model to the sum of all accuracies is used as the assignment coefficient.

3. The network site selection recommendation method according to claim 1, characterized in that: Determining the accuracy of each of the weight distribution models, comprising: Using various evaluation data corresponding to some reference areas that have not participated in model training and real expected business indicator data of outlets to test each of the weight allocation models, and obtaining predicted expected business indicator data of outlets output by each of the weight allocation models; The accuracy of each of the weight allocation models is determined using the predicted network point expected business indicator data and the actual network point expected business indicator data.

4. The network site selection recommendation method according to claim 3, characterized in that: Determining the accuracy of each of the weight allocation models using the predicted network point expected business indicator data and the actual network point expected business indicator data includes: Determine the prediction error between the predicted network point expected business indicator data and the actual network point expected business indicator data and the overall actual value corresponding to all the actual network point expected business indicator data; The ratio of the prediction error to the overall true value is taken as the prediction error rate, and the accuracy is obtained by subtracting the prediction error rate from one.

5. The network site selection recommendation method according to claim 1, characterized in that: Determining the importance score of each reference region according to the multi-source evaluation data set after the importance assignment includes: For each of the reference regions, the sum of the various evaluation data after the importance is assigned is determined as the importance score of the reference region.

6. A network site selection recommendation device, characterized in that: include: An acquisition module is used to acquire a multi-source evaluation data set and a network point expected business indicator data set; wherein the multi-source evaluation data set includes various evaluation data corresponding to each reference area; the types of the evaluation data include: commercial type, transportation type, population type and peer type; the network point expected business indicator data set includes network point expected business indicator data corresponding to each reference area; A training module, used to train multiple weight allocation models using the multi-source evaluation data set and the network point expected business indicator data set; an assignment module, used to assign importance to various evaluation data of the reference area using the weight allocation model; The training module is specifically used to: pre-process the multi-source evaluation data set; select the various evaluation data of some reference areas and the expected business indicator data of the network points, train different machine learning models, and obtain multiple weight allocation models; The assignment module is specifically used to determine the accuracy of each of the weight allocation models; determine the assignment coefficient corresponding to each of the weight allocation models using the accuracy; obtain the feature weights corresponding to the various evaluation data output by each of the weight allocation models after training; and weight the feature weights of each of the evaluation data under each of the weight allocation models by the assignment coefficient of the weight allocation model to complete the importance assignment; The determination module is used to determine the importance score of each reference area according to the multi-source evaluation data set after the importance is assigned, and use the reference area whose importance score exceeds a preset threshold as the network point location area.

7. An electronic device, characterized in that: The invention comprises a processor and a memory, wherein the memory stores a computer program executable by the processor, and the processor can execute the computer program to implement the network site selection recommendation method according to any one of claims 1 to 5.

Images