Recommendation methods, devices, equipment, and storage media for service outlets

Abstract

This disclosure provides a method, apparatus, device, and storage medium for recommending service points, relating to the field of computer technology, particularly to the fields of cloud computing, big data, deep learning, and intelligent recommendation technology. The specific implementation scheme includes: determining the attribute information of each service point in the target service point dataset and the point of interest information of each service point within a first preset range; determining a first score for each service point based on the attribute information and point of interest information; and determining a preset number of recommended service points from the target service point dataset based on the first score of each service point. This disclosure can determine the score of each service point based on its attribute information and point of interest information, providing a basis for evaluating the commercial potential of each service point and making entry decisions.

Description

Technical Field

[0001] This disclosure relates to the field of computer technology, and in particular to the fields of cloud computing, big data, deep learning, and intelligent search technology. Background Technology

[0002] To achieve greater revenue targets, brands need to establish a presence in more retail outlets to sell their products. However, determining which outlets offer the highest commercial potential is a current challenge for brands. Summary of the Invention

[0003] This disclosure provides a method, apparatus, device, and storage medium for recommending service locations.

[0004] According to one aspect of this disclosure, a method for recommending service outlets is provided, comprising:

[0005] Determine the attribute information of each point in the target point dataset and the point of interest information of each point within a first preset range.

[0006] Based on attribute information and point of interest information, determine the first score for each site.

[0007] Based on the first rating of each location, a predetermined number of recommended locations are determined from the target location dataset.

[0008] According to another aspect of this disclosure, a site recommendation device is provided, comprising:

[0009] The first determining module is used to determine the attribute information of each point in the target point dataset and the point of interest information of each point within a first preset range.

[0010] The first rating module is used to determine the first rating for each site based on attribute information and point of interest information.

[0011] The recommendation module is used to determine a preset number of recommended outlets from the target outlet dataset based on the first rating of each outlet.

[0012] According to another aspect of this disclosure, an electronic device is provided, comprising:

[0013] At least one processor; and

[0014] The memory is communicatively connected to the at least one processor; wherein,

[0015] The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the recommended method for the aforementioned network points.

[0016] According to another aspect of this disclosure, a non-transitory computer-readable storage medium is provided storing computer instructions for causing the computer to perform the recommended method for the aforementioned network.

[0017] According to another aspect of this disclosure, a computer program product is provided, including a computer program that, when executed by a processor, implements the recommended method for the aforementioned network points.

[0018] This disclosure can determine the score of each site based on its attribute information and point of interest information, providing a basis for evaluating the commercial potential of each site and making entry decisions.

[0019] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of this disclosure, nor is it intended to limit the scope of this disclosure. Other features of this disclosure will become readily apparent from the following description. Attached Figure Description

[0020] The accompanying drawings are provided to better understand this solution and do not constitute a limitation of this disclosure. Wherein:

[0021] Figure 1 This is a flowchart illustrating a method for recommending service outlets according to an embodiment of this disclosure;

[0022] Figure 2 This is a schematic diagram illustrating the implementation flow of the site recommendation method according to an embodiment of the present disclosure.

[0023] Figure 3A This is a schematic diagram of a branch office dataset page according to an embodiment of the present disclosure;

[0024] Figure 3B This is a schematic diagram of a cleaning task page according to an embodiment of the present disclosure;

[0025] Figure 4A This is a schematic diagram illustrating the implementation process of creating an extended task according to an embodiment of this disclosure;

[0026] Figure 4B This is a schematic diagram of the first configuration page for creating an extended task according to an embodiment of this disclosure;

[0027] Figure 4C This is a schematic diagram of the second configuration page for creating an extended task according to an embodiment of this disclosure;

[0028] Figure 5 This is a schematic diagram of the process for calculating the potential score of a business outlet according to an embodiment of this disclosure;

[0029] Figure 6A ~E is a schematic diagram of the display page for recommended outlets according to an embodiment of this disclosure;

[0030] Figure 7 This is a schematic diagram of the structure of a site recommendation device according to an embodiment of the present disclosure;

[0031] Figure 8 This is a block diagram of an electronic device that can be used to implement the site recommendation method of the embodiments of this disclosure. Detailed Implementation

[0032] The exemplary embodiments of this disclosure are described below with reference to the accompanying drawings, including various details of the embodiments to aid understanding, and should be considered merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of this disclosure. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description.

[0033] The fast-moving consumer goods (FMCG) industry is highly competitive, with a wide variety of products within the same category. Brands typically use one of two methods to introduce their products to new retail outlets:

[0034] The first approach involves brands hiring employees to build their own store database and then manually deciding which stores to add based on experience. The drawback of this approach is that store selection relies on subjective personal experience and lacks objective decision-making criteria.

[0035] The second approach involves brands purchasing original local outlet data and hiring staff to visit and verify the authenticity and potential of these outlets. Based on feedback from these on-site visits, they then decide which outlets to establish. The drawback of this approach is that because the outlet data is purchased in a one-time transaction, information such as the closure or addition of outlets cannot be updated in the data, resulting in low data validity. Furthermore, the manual visits to outlets lead to infrequent updates and the possibility of different staff entering the same information multiple times due to varying data entry habits, resulting in poor data quality.

[0036] This disclosure provides a method for recommending service outlets, which predicts the commercial potential of each outlet by using its attribute information and points of interest information in the surrounding area. This method can effectively avoid the shortcomings of the above methods, such as slow outlet data updates, poor outlet data quality, and lack of basis for entry decisions.

[0037] The present application will now be described in detail with reference to the accompanying drawings and embodiments.

[0038] Figure 1 This is a flowchart illustrating a site recommendation method according to an embodiment of the present disclosure. The method includes at least the following steps:

[0039] S101: Determine the attribute information of each point in the target point dataset and the point of interest information of each point within a first preset range.

[0040] In this embodiment of the disclosure, the outlet can be a commercial store selling various goods. The outlet's attribute information may include the outlet's location information (e.g., the outlet belongs to Chaoyang District, Beijing), outlet type information, outlet chain attribute information, outlet address, and latitude and longitude coordinates. The outlet type information can indicate whether the outlet is a shopping mall, supermarket, convenience store, restaurant, or specialty store, etc. The outlet chain attribute can indicate whether the outlet belongs to a chain. The outlet chain attribute can also indicate the chain scale of the outlet, such as a large chain shopping mall / supermarket, a medium-sized chain shopping mall / supermarket, or a chain convenience store, etc.

[0041] The target dataset can be obtained from the needs of users (brand owners) who want to add new outlets. For example, the target dataset can be constructed based on one or more of the user-specified requirements, such as the outlet's region information, outlet type information, and outlet chain attribute information. The specific user requirements used can be selected and adjusted; this embodiment is only for illustrative purposes. Each outlet in the target dataset includes at least the outlet's attribute information and the outlet's point of interest information within a first preset range.

[0042] The point of interest information of each network point within the first preset range can be understood as the point of interest information covered within the coverage area formed after spreading outwards by a preset distance from the network point as the center.

[0043] Points of Interest (POI) information can include various geographic objects and facilities, such as bus stops, parking lots, residential areas, shops, and plazas. To better differentiate the commercial potential of the area where a point is located, in one example, geographic objects related to shopping, leisure and entertainment, transportation facilities, education and training, tourist attractions, real estate, and hotels can be selected as POI information. POI information can also include the number of each specific POI and the total number of various types of POIs, such as the number of supermarkets and the total number of shopping-related POIs obtained by summing the number of supermarkets, department stores, convenience stores, etc.

[0044] S102: Determine the first score for each site based on attribute information and point of interest information.

[0045] The first score can be understood as a predicted score of the business potential of a location based on its attribute information and POI information. The higher the score, the higher the business potential of the location.

[0046] S103: Based on the first score of each site, determine a preset number of recommended sites from the target site dataset.

[0047] The preset quantity can be determined based on user needs and budget. For example, if the user needs to add 50 sales outlets, the preset quantity is 50. Or, if the user's budget is 100,000 yuan and the cost of a single recommended outlet is 1,000 yuan, the preset quantity is 100.

[0048] According to the scheme of this disclosure embodiment, the score of each site is determined by the attribute information of the site and the POI information of the surrounding site, thereby assessing the business potential of each site and providing a basis for decision-making for site entry.

[0049] In one possible implementation, the site recommendation method of this disclosure embodiment includes steps S101 to S103, wherein step S102: determining a first score for each site based on attribute information and POI information, may further include the step:

[0050] S201: Determine the commercial reach score for each site based on attribute information and POI information.

[0051] In this embodiment of the disclosure, the commercial radiation score can be understood as a quantitative representation of the extent to which the outlet can benefit from the foot traffic of the nearest commercial center, based on the distance of the outlet to the nearest commercial center. Its calculation can be performed through the following steps:

[0052] Based on the POI information of the outlet, determine the location of each business center corresponding to that outlet.

[0053] Calculate the distance from the outlet to each business center based on the outlet's attribute information.

[0054] The commercial radiation score is determined based on the minimum distance from the outlet to each business center. The smaller the minimum distance, the higher the commercial radiation score.

[0055] The commercial radiation score can also be understood as a quantitative representation of the extent to which a location can benefit from the foot traffic of commercial centers, based on the number of commercial centers within a third preset range. Its calculation can be performed through the following steps:

[0056] Based on the POI information of the outlet, determine the number of business centers corresponding to that outlet.

[0057] The business radiation score is determined based on the number of business centers corresponding to the outlet; the more business centers, the higher the business radiation score.

[0058] S202: Perform feature encoding on attribute information and POI information to obtain the first feature of each site.

[0059] Feature encoding is performed on attribute information and POI information to obtain the first feature of each location. This includes: one-hot encoding of the qualitative descriptive features (e.g., store type, whether it is a chain store) contained in both attribute information and POI information. The continuous value features (e.g., the number of surrounding stores in the POI information) are discretized through feature binning before feature encoding. In one example, the number of surrounding stores can be divided into three intervals: 0 to 9, 10 to 29, and 30 or more, thus discretizing the continuous value.

[0060] S203: Determine the target model based on the target product category.

[0061] S204: Input the commercial radiation score and the first feature into the target model to determine the first score for each site.

[0062] The target product category can be understood as the category of products that need to be sold at the outlet. For example, the target product category could be dairy products, personal care products, cosmetics, alcoholic beverages, condiments, etc. For different target product categories, a target model can be pre-trained using the corresponding product sales data. If there is no model corresponding to the target product category, a general model can be used as the target model. The general model can be trained using the overall sales data of the outlet. The type, structure, and computational logic of the target model used are not specifically limited here, as long as it can score each outlet and obtain a first-place rating.

[0063] According to the solution of this disclosure embodiment, corresponding models are provided for multiple target product categories, thereby making the network recommendation for the target product more accurate.

[0064] In one possible implementation, the site recommendation method of this embodiment includes steps S101 to S103 and steps S201 to S204, wherein step S202: feature encoding is performed on the attribute information and POI information to obtain the first feature of each site, which may further include the following steps:

[0065] S301: Encode the attribute information to obtain the second feature of each network point.

[0066] In this embodiment of the disclosure, feature encoding of attribute information can be understood as performing feature encoding on one or more of the information included in the attribute information, such as the area information to which the outlet belongs, the outlet type information, the outlet chain attribute information, the outlet address and latitude and longitude coordinates, so as to obtain the second features of the outlet in multiple different dimensions.

[0067] S302: Perform feature crossing on the second feature to obtain the third feature for each point.

[0068] Feature crossing of the second feature can be understood as crossing at least some second features from different dimensions to obtain the third feature. In one example, feature crossing could be crossing the second feature obtained from the network chain attribute information with the second feature obtained from the network type information, such as crossing the second feature of the "chain" dimension with the second feature of the "supermarket" dimension, or crossing the second feature of the "non-chain" dimension with the second feature of the "convenience store" dimension.

[0069] S303: Perform feature encoding on the point of interest information to obtain the fourth feature of each point.

[0070] S304: Combine the third and fourth features to obtain the first feature of each network point.

[0071] In this embodiment of the disclosure, by performing feature cross-referencing on the various features contained in the attribute information and POI information of the network points, more features of different dimensions and higher-dimensional features can be obtained. Utilizing these features can help improve the accuracy of the target model prediction, so as to obtain a more valuable and objective first score for each network point.

[0072] In one possible implementation, the site recommendation method of this disclosure embodiment includes steps S101 to S103, and may further include the step:

[0073] S401: Based on the target product category, determine the demographic information of each outlet within the second preset range.

[0074] Based on step S401, step S103: determining a preset number of recommended locations from the target location dataset according to the first rating of each location, may further include:

[0075] S1031: Determine the second score for each outlet based on the demographic information.

[0076] In this embodiment, the population profile information may include category parameters such as population size, age, gender, life stage, education level, monthly income level, consumption level, and dining level within a second preset range. The second preset range can be understood as the Area of ​​Interest (AOI) formed around the location. In one example, the population size can be divided into working population and residential population. Age can be divided into multiple ranges: 0-18, 18-25, 25-35, 45-55, 55-65, and over 65. Life stage may include several sub-values ​​such as pregnancy, having a child aged 0-1 years, primary school student, junior high school student, university student, and working.

[0077] The second rating can be understood as a predictive rating of the business potential of a location based on demographic data.

[0078] S1032: Based on the first and second scores of each site, determine a preset number of recommended sites from the target site dataset.

[0079] The first and second scores can be weighted to obtain the overall potential score of the site. Based on the overall potential score, a predetermined number of recommended sites are determined from the target site dataset.

[0080] According to the scheme of this disclosure embodiment, the commercial potential of the outlet is comprehensively predicted by combining the first score with the second score obtained from the crowd profile information, thereby improving the accuracy of the prediction.

[0081] In one possible implementation, step S1031: determining the second score for each location based on the crowd profile information, may further include the following steps:

[0082] S501: Determine the parameter information for each category parameter in the population profile information.

[0083] In this embodiment of the disclosure, the parameter information for each category parameter includes the percentage of people in each numerical range of that category parameter obtained from the population profile information. Category parameters may include age, gender, and monthly income level. Gender parameters may include: 55% male and 45% female. Age parameters may include: 10% aged 0-18, 15% aged 18-25, 30% aged 25-45, 18% aged 45-55, 12% aged 55-65, and 15% aged 65 and above. Monthly income level parameters may include: 20% earning less than 3000 yuan, 40% earning 3000-5000 yuan, 30% earning 50000-8000 yuan, 7% earning 8000-20000 yuan, and 3% earning more than 20000 yuan.

[0084] S502: Determine the weight of each category parameter.

[0085] For different target products, the weight of each category parameter can be set based on the product's appeal to different groups and their purchasing power. For example, age could be weighted at 30%, monthly income at 30%, and gender at 40%. Furthermore, the weight of each numerical range within each category parameter can be set. For instance, for beauty products, the weight of females could be set at 90%, and males at 10%. Similarly, the weight of monthly income ranges could be set as follows: below 3000 yuan at 10%, 3000-5000 yuan at 20%, 50000-8000 yuan at 30%, 8000-20000 yuan at 30%, and above 20000 yuan at 10%. The weights for the 0-18 age range are set at 5%, the 18-25 age range at 35%, the 25-45 age range at 35%, the 45-55 age range at 15%, the 55-65 age range at 5%, and the age range over 65 at 5%.

[0086] S503: Determine the second score for each site based on the weight of each category parameter and the parameter information of each category parameter.

[0087] The score for each category parameter is calculated by multiplying the percentage of people in each interval by the weight of that interval and then summing the results. The second score is obtained by multiplying the score of each category parameter by its corresponding weight. In the example above, the second score = age category parameter score * 30% + monthly income level category parameter score * 40% + gender category parameter score * 40%. The age category parameter score is calculated by multiplying the percentage of people in each interval by the corresponding interval weight:

[0088] The percentage of people aged 0-18 multiplied by the corresponding interval weight: 10% * 5% = 0.005.

[0089] The percentage of people aged 18-25 multiplied by the corresponding interval weight: 15% * 35% = 0.0525.

[0090] The percentage of people aged 25-45 multiplied by the corresponding interval weight: 30% * 35% = 0.105.

[0091] The percentage of people aged 45-55 multiplied by the corresponding interval weight: 18% * 15% = 0.027.

[0092] The percentage of people aged 55-65 multiplied by the corresponding interval weight: 12% * 5% = 0.006.

[0093] The percentage of people aged 65 and above multiplied by the corresponding interval weight: 15% * 5% = 0.0075.

[0094] Adding the above scores together and multiplying by 100 yields a score of 20.3 for the age category parameter. Similarly, the monthly income level score is 21.4, and the gender score is 46. Based on the weights of these category parameters, the second score is 20.3*30% + 21.4*30% + 46*40% = 30.91.

[0095] It should be noted that, for different target products, the selection of which category parameters to use from the user profile data can be set based on the brand's user profile data or on relevant experience.

[0096] According to the scheme of this disclosure embodiment, the second score of the outlet is obtained based on the outlet's customer profile information. When determining the second score, the weight of each category parameter in the customer profile information can be set according to the target product, so that the second score is more relevant to the user group of the target product and can better reflect the commercial potential of each outlet for the target product category.

[0097] In one possible implementation, the site recommendation method of this disclosure embodiment includes steps S101 to S103, and may further include the step:

[0098] S601: Obtain information on registered outlets.

[0099] In this embodiment of the disclosure, the user can provide information about the outlets they have registered with, and the outlet information includes at least the store name.

[0100] S602: Based on the network point information and the point of interest information in the target map, remove erroneous network points from the network points already registered, and / or merge duplicate network points from the network points already registered to obtain the network point dataset.

[0101] The locations that need to be removed are those that do not exist on the target map (e.g., Baidu Maps) or those that have been entered repeatedly. Examples are as follows:

[0102] The user has entered the following registered outlets: "Xihongmen Bianlifeng", "Xihongmen Convenience Store", and "Convenience Store Xihongmen Branch".

[0103] The system compared the three locations with the "Bianlifeng (Xihongmen)" location on Baidu Maps and obtained the following results:

[0104] i. The match between "Xihongmen Convenience Store" and "Bianlifeng (Xihongmen Point)" is low, so it is assumed that the point does not exist in the Baidu Maps database and is therefore removed.

[0105] ii. “Xihongmen Bianlifeng”, “Xihongmen Branch of Bianlifeng” and “Bianlifeng (Xihongmen Point)” have a high degree of matching. Rewrite the names of these two outlets as “Bianlifeng (Xihongmen Point)”. At this time, there are two “Bianlifeng (Xihongmen Point)”. The system will randomly remove one of the “Bianlifeng (Xihongmen Point)” outlets. In the end, only one outlet will be retained out of the three outlets.

[0106] According to the solution of this disclosure embodiment, duplicate and erroneous outlets in the user-provided registered outlets can be cleaned to accurately obtain the registered outlets dataset, which facilitates management and deduplication in recommended outlets.

[0107] In one possible implementation, the site recommendation method of this disclosure embodiment includes steps S101 to S103, and may further include the following steps:

[0108] S701: Obtain the initial network point dataset based on location information and network point category information.

[0109] S702: When the network recommendation mode is to remove existing networks, based on the network dataset of already registered networks, remove the corresponding networks from the initial network dataset and the network dataset of already registered networks to obtain the target network dataset.

[0110] In this embodiment, the location information and outlet category information are determined by the user's needs. The user's outlet entry requirements typically include specific geographical and outlet category requirements. For example, a user might need to add outlets in Suzhou City, with the outlet category requirement being convenience stores and supermarkets. Based on these requirements, all convenience stores and supermarkets in Suzhou City can be obtained as the initial outlet dataset. Then, based on the user-provided dataset of already entered outlets, the user-provided outlets are removed to obtain the target outlet dataset for which ratings need to be calculated.

[0111] According to the solution of this disclosure embodiment, by removing user-registered locations from the target dataset before calculating recommended locations, user-registered locations are avoided from appearing in the recommended locations, thus reducing user spending.

[0112] In one possible implementation, the site recommendation method of this disclosure embodiment includes steps S101 to S103, and may further include the step:

[0113] S801: Obtain the initial network point dataset based on location information and network point category information.

[0114] S802: When the network recommendation mode is to retain existing networks, the target network dataset is obtained by matching the initial network dataset with the corresponding network tags in the existing network dataset based on the already registered network dataset.

[0115] In this embodiment, instead of removing user-registered service points, they are marked. Whether each service point is a registered service point can be displayed in the recommended service points list.

[0116] According to the scheme of this disclosure embodiment, the calculated recommended service points may include service points that the user has already registered with, thereby obtaining a rating for the service points the user has registered with. This can provide a basis for the user to decide whether to adjust or withdraw from their registered service points.

[0117] In one possible implementation, the site recommendation method of this disclosure embodiment includes steps S101 to S103, and may further include the step:

[0118] S901: Visually display a preset number of recommended service points in the form of an information table. And / or

[0119] S902: Visually display a preset number of recommended service points on the map.

[0120] In this embodiment of the disclosure, the information table may include the location information, name information, and rating information of the recommended service points. When the recommended service points are visually displayed on a map, at least one of the service point's name information, location information, rating information, and user profile information may be displayed in a preset area on the map.

[0121] According to the solution of this disclosure embodiment, by visualizing the recommended outlets, users can quickly and intuitively understand the detailed information of each outlet.

[0122] Figure 2 This is a schematic diagram illustrating the implementation process of a site recommendation method according to an embodiment of the present disclosure. This method can be applied to a data processing device, for example, the data processing device can be deployed on a terminal, server or other processing device to implement site recommendation.

[0123] In one example, the data processing device is a server, which can be a standalone server, a server cluster, a distributed system, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, and big data and artificial intelligence platforms. The server installs and runs an application or system that supports the recommended methods for running the site. Users can send requests to the server via a terminal to run the recommended methods for running the site. The terminal connects to the server via a wireless or wired network. Optionally, the terminal can be a smartphone, tablet, laptop, desktop computer, smartwatch, in-vehicle terminal, etc., but is not limited to these.

[0124] like Figure 2 As shown, the method for recommending service outlets includes the following steps:

[0125] Step 1: Upload existing branch data.

[0126] Users can access the web-based system on the server through their terminals, such as... Figure 3A The "Branch Dataset" page shown uploads the branch information it has collected and organized from the branches that have already joined the network.

[0127] Step 2: Clean the network data.

[0128] Users can, for example Figure 3B Create a cleaning task on the "Cleaning Task" page shown. During the cleaning process, the system will compare each point in the dataset with a Baidu Map POI, first keeping matching points and then removing duplicate points. After cleaning is complete, the user can save the cleaned points as a new version of the dataset. The cleaning method can refer to steps S601-S602 in the above method embodiment and the examples provided in those steps.

[0129] Step 3: Create an extended task.

[0130] The specific process for creating an extended task is as follows: Figure 4A As shown, it includes the following steps:

[0131] 1) The user selects the "product category" (i.e. the product category of the target product).

[0132] In one example, such as Figure 4B As shown, the user selected shower gel as the product category on the "Create Extended Task" page.

[0133] 2) The system automatically matches evaluation indicators and indicator weights, with the indicator weights adopting expert experience weights.

[0134] like Figure 4B As shown, the system matches "work-residence scale" and "age composition" based on the shower gel.

[0135] The initial assessment indicators are "Consumption Level". Users can manually customize and modify these indicators.

[0136] 3) Depending on whether the user has customized the evaluation metrics, proceed to step 4 if yes, otherwise proceed to step 5.

[0137] 4) Normalize the weights of the evaluation metrics selected by the user.

[0138] 5) Depending on whether the user has enabled advanced configuration, proceed to step 6 if yes, otherwise proceed to step 7.

[0139] When users have a clear consumer profile, "Advanced Metric Configuration" can be enabled, such as... Figure 4CAs shown, users can select evaluation indicators and enter indicator weights on the page, as well as select target intervals for each indicator and enter interval weights. The system will then calculate the population profile score based on the indicator weights and interval weights.

[0140] 6) The system calculates the population profile score based on the evaluation indicators, indicator weights, and interval weights selected by the user.

[0141] 7) Users can choose to upload their existing network datasets.

[0142] 8) Users select to expand to other cities (i.e., location information).

[0143] 9) Users select expansion channels (i.e., outlet category information).

[0144] 10) The system obtains the initial network point dataset based on the target location information and target network point category information, and calculates the upper limit of the existing network points.

[0145] 11) The user inputs the number of new outlets to be added (i.e., the preset number).

[0146] 12) Calculate the task cost.

[0147] 13) The user chooses whether to deduplicate the recommended outlets (i.e., whether to remove the outlets that have already been registered). If yes, proceed to step 14; otherwise, proceed to step 15.

[0148] 14) Before calculation, the system removes user-entered outlets (i.e., registered outlets) from the initial outlet dataset. The method for removing registered outlets can refer to steps S701 to S702, S801 to S802 in the above method embodiment and the examples provided for each step.

[0149] 15) Calculate the extended results.

[0150] After clicking "Start Expansion," the system begins calculating the expansion network. Based on the selected evaluation indicators, the system evaluates each selected channel outlet within the selected city, retaining the top-scoring outlets as recommended outlets based on the entered "Number of Outlets." The calculation method is as follows: Figure 5 As shown, it includes the following steps:

[0151] a. Retrieve Baidu Maps based on the user's selected location and service point type information.

[0152] POIs are used to obtain the initial point dataset.

[0153] b. Obtain the site's own attributes (i.e., attribute information), surrounding POI information (i.e., point of interest information), and surrounding AOI population profiles (i.e., population profile information) for each site.

[0154] c. Perform feature engineering on the attribute information and point-of-interest information to obtain the commercial radiation index (i.e., commercial radiation score) and first feature for each location. Feature engineering includes:

[0155] The first step is to calculate the commercial radiation index of the outlet using a commercial radiation model, representing the commercial potential of the outlet's geographical location. The second step is to perform hot-coded unique encoding on qualitative descriptive features (such as the outlet's distribution channels and whether it is a chain store). For continuous features, they are discretized using feature binning. The third step is to perform feature cross-referencing to obtain more features, which will serve as raw materials for the model input.

[0156] d. Input the commercial radiation score and the first feature into a pre-defined regression model based on the product category to obtain the POI comprehensive score (i.e., the first score). The pre-defined regression model can include personal care products, dairy products, alcoholic beverages, condiments, and general models, etc.

[0157] e. For demographic information, if advanced indicator configuration is not enabled, the system will use preset indicator weights and interval weights derived from expert experience and market reports to calculate the demographic profile score. If advanced indicator configuration is enabled, the system will calculate the demographic profile score (i.e., the second score) based on the input consumer profile.

[0158] f. Take a weighted average of the first and second scores to obtain the overall potential score of the outlet.

[0159] g. Sort all sites by their comprehensive potential score, and output the top-ranked sites as recommended sites based on the expansion quantity entered during configuration.

[0160] The method for calculating recommended locations can refer to steps S101 to S103, S201 to S204, S301 to S304, S401, S1031 to S1032, and S501 to S503 in the above method embodiment, as well as the examples provided for each step.

[0161] Step 4: View the expanded results.

[0162] After the expansion task is completed, users can view the recommended locations obtained from the expansion results, including the following steps:

[0163] 1) such as Figure 6A As shown, the recommended service points are visualized in the form of an information table. The table contains multiple attribute information and potential scores for each recommended service point.

[0164] 2) Users can click "Map View" to visualize the recommended service points on the map. For example... Figure 6B As shown, you can view the number of recommended outlets for each province in turn. Clicking on a province will allow you to view the number of recommended outlets for that city.

[0165] 3) Click on a city, such as Figure 6C As shown, you can view the potential distribution of different areas in the city.

[0166] 4) Click on a specific area, such as... Figure 6D As shown, you can view the specific distribution of recommended service points in this area. Click on a service point, as shown... Figure 6E As shown, detailed information about the branch can be viewed, along with a predicted potential score. The method for displaying recommended branches can be found in steps S901 to S902 of the above method embodiments, as well as the examples provided for each step.

[0167] The steps in the embodiments of this disclosure can be found in the relevant descriptions in the above method embodiments of this disclosure, and will not be repeated here.

[0168] The acquisition, storage, and application of user personal information involved in the technical solution disclosed herein comply with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0169] This disclosure provides an embodiment of a service point recommendation device. Figure 7 This is a schematic diagram of the structure of a site recommendation device according to an embodiment of the present disclosure, as shown below. Figure 7 As shown, the recommendation device 700 for this outlet includes at least:

[0170] The first determining module 710 is used to determine the attribute information of each point in the target point dataset and the point of interest information of each point within a first preset range.

[0171] The first scoring module 720 is used to determine the first score for each site based on attribute information and point of interest information.

[0172] The recommendation module 730 is used to determine a preset number of recommended outlets from the target outlet dataset based on the first rating of each outlet.

[0173] This disclosure also proposes another branch recommendation device, which includes one or more features of the branch recommendation device 700 of the above embodiments. In one possible implementation, the branch recommendation device includes:

[0174] The system comprises a first determination module 710, a first scoring module 720, a recommendation module 730, a first acquisition module, a cleaning module, a removal module, a labeling module, and a visualization module. The first determination module 710, the first scoring module 720, and the recommendation module 730 are identical to their corresponding modules described above, and will not be repeated here.

[0175] In one possible implementation, the first scoring module 720 further includes:

[0176] The first determination submodule is used to determine the commercial radiation score of each site based on attribute information and point of interest information.

[0177] The encoding submodule is used to perform feature encoding on attribute information and point of interest information to obtain the first feature of each point.

[0178] The second determination submodule is used to determine the target model based on the target product category.

[0179] The third determination submodule is used to input the commercial radiation score and the first feature into the target model to determine the first score of each network point.

[0180] In one possible implementation, the encoding submodule is used for:

[0181] The attribute information is feature-encoded to obtain the second feature of each network point.

[0182] The second feature is cross-referenced to obtain the third feature for each network point.

[0183] The interest point information is feature-encoded to obtain the fourth feature of each point.

[0184] By combining the third and fourth features, the first feature of each network point is obtained.

[0185] In one possible implementation, the second determining module is used to: determine the demographic information of each outlet within a second preset range based on the target product category.

[0186] Recommendation module 730 further includes:

[0187] The fourth sub-module is used to determine the second score for each location based on the demographic profile information.

[0188] The fifth determination submodule is used to determine a preset number of recommended outlets from the target outlet dataset based on the first and second scores of each outlet.

[0189] In one possible implementation, the fourth determining submodule is used for:

[0190] Determine the parameter information for each category in the population profile information.

[0191] Determine the weight of each category parameter.

[0192] The second score for each site is determined based on the weight of each category parameter and the parameter information for each category parameter.

[0193] In one possible implementation, the first acquisition module is used to: acquire network information of the network points that have been registered.

[0194] The cleaning module is used to: remove erroneous sites from the existing sites based on the site information and the point of interest information in the target map, and / or merge duplicate sites from the existing sites to obtain the existing site dataset.

[0195] In one possible implementation, the second acquisition module is used to: acquire an initial network point dataset based on location information and network point category information.

[0196] The elimination module is used to: when the network recommendation mode is to remove existing networks, based on the network dataset of already registered networks, eliminate the corresponding networks in the initial network dataset and the network dataset of already registered networks to obtain the target network dataset.

[0197] In one possible implementation, the second acquisition module is used to: acquire an initial network point dataset based on location information and network point category information.

[0198] The tagging module is used to: when the network recommendation mode is to retain existing network points, and based on the network point dataset already in the network, tag the corresponding network points in the initial network point dataset and the network point dataset already in the network to obtain the target network point dataset.

[0199] In one possible implementation, the visualization module is used to: visualize the recommended service points in the form of an information table; and / or visually display the recommended service points on a map.

[0200] The specific functions and examples of each module and submodule of the data processing apparatus in this disclosure can be found in the relevant descriptions of the corresponding steps in the above method embodiments, and will not be repeated here.

[0201] According to embodiments of this disclosure, this disclosure also provides an electronic device, a readable storage medium, and a computer program product.

[0202] Figure 8 A schematic block diagram of an example electronic device 800 that can be used to implement embodiments of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0203] like Figure 8As shown, device 800 includes a computing unit 801, which can perform various appropriate actions and processes based on a computer program stored in read-only memory (ROM) 802 or a computer program loaded from storage unit 808 into random access memory (RAM) 803. RAM 803 may also store various programs and data required for the operation of device 800. The computing unit 801, ROM 802, and RAM 803 are interconnected via bus 804. Input / output (I / O) interface 805 is also connected to bus 804.

[0204] Multiple components in device 800 are connected to I / O interface 805, including: input unit 806, such as a keyboard, mouse, etc.; output unit 807, such as various types of displays, speakers, etc.; storage unit 808, such as a disk, optical disk, etc.; and communication unit 809, such as a network card, modem, wireless transceiver, etc. Communication unit 809 allows device 800 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0205] The computing unit 801 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 801 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 801 performs the various methods and processes described above, such as the point recommendation method. For example, in some embodiments, the point recommendation method may be implemented as a computer software program tangibly contained in a machine-readable medium, such as storage unit 808. In some embodiments, part or all of the computer program may be loaded and / or installed on device 800 via ROM 802 and / or communication unit 809. When the computer program is loaded into RAM 803 and executed by the computing unit 801, one or more steps of the point recommendation method described above may be performed. Alternatively, in other embodiments, the computing unit 801 may be configured to perform the point recommendation method by any other suitable means (e.g., by means of firmware).

[0206] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0207] The program code used to implement the methods of this disclosure may be written in any combination of one or more programming languages. This program code may be provided to a processor or controller of a general-purpose computer, special-purpose computer, or other programmable data processing apparatus, such that when executed by the processor or controller, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program code may be executed entirely on a machine, partially on a machine, as a standalone software package partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0208] In the context of this disclosure, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0209] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the computer. Other types of devices can also be used to provide interaction with the user. For example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback). Input from the user can be received in any form (including sound input, voice input, or tactile input).

[0210] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as a data server), or computing systems that include middleware components (e.g., an application server), or computing systems that include frontend components (e.g., a user computer with a graphical user interface or web browser through which a user can interact with embodiments of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., a communication network). Examples of communication networks include local area networks (LANs), wide area networks (WANs), and the Internet.

[0211] Computer systems can include clients and servers. Clients and servers are generally located far apart and typically interact via communication networks. Client-server relationships are created by computer programs running on the respective computers and having a client-server relationship with each other. Servers can be cloud servers, servers in distributed systems, or servers incorporating blockchain technology.

[0212] It should be understood that the various forms of processes shown above can be used to rearrange, add, or delete steps. For example, the steps described in this disclosure can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution disclosed in this disclosure can be achieved, and this is not limited herein.

[0213] The specific embodiments described above do not constitute a limitation on the scope of protection of this disclosure. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this disclosure should be included within the scope of protection of this disclosure.

Claims

1. A method for recommending service outlets, comprising: Determine the attribute information of each point in the target point dataset and the point of interest information of each point within a first preset range; Based on the attribute information and the point of interest information, determine the commercial radiation score of each network point; The attribute information and the point of interest information are feature-encoded to obtain the first feature of each network point; Determine the target model based on the target product category; The commercial radiation score and the first feature are input into the target model to determine the first score for each outlet. Based on the target product category, determine the customer profile information for each outlet within a second preset range; Determine the parameter information for each category parameter in the aforementioned population profile information; Determine the weight of each category parameter; Based on the weight of each category parameter and the parameter information of each category parameter, a second score is determined for each outlet; Based on the first and second ratings of each location, a preset number of recommended locations are determined from the target location dataset.

2. The method according to claim 1, wherein, The step of feature encoding the attribute information and the point of interest information to obtain the first feature of each network point includes: The attribute information is feature-encoded to obtain the second feature of each network point; The second feature is subjected to feature cross-interaction to obtain the third feature of each network point; The interest point information is feature-encoded to obtain the fourth feature of each point; The third feature and the fourth feature are combined to obtain the first feature of each network point.

3. The method according to claim 1 or 2, further comprising: Obtain information on branches that have already joined the network; Based on the network point information and the point of interest information in the target map, erroneous network points are removed from the network points already registered, and / or duplicate network points are merged to obtain a dataset of registered network points.

4. The method according to claim 3, further comprising: Based on location information and branch category information, obtain the initial branch dataset; When the network recommendation mode is to remove existing networks, the corresponding networks in the initial network dataset and the existing network dataset are removed based on the already registered network dataset to obtain the target network dataset.

5. The method according to claim 3, further comprising: Based on location information and branch category information, obtain the initial branch dataset; When the network recommendation mode is to retain existing network points, the target network point dataset is obtained by matching the initial network point dataset with the corresponding network point tags in the existing network point dataset based on the already registered network point dataset.

6. The method according to claim 1 or 2, further comprising: The recommended outlets are displayed visually in the form of an information table; and / or The recommended locations are visualized on a map.

7. A network recommendation device, comprising: The first determining module is used to determine the attribute information of each point in the target point dataset and the point of interest information of each point within a first preset range; The first scoring module includes: The first determining submodule is used to determine the commercial radiation score of each network point based on the attribute information and the point of interest information; The encoding submodule is used to perform feature encoding on the attribute information and the point of interest information to obtain the first feature of each network point; The second determination submodule is used to determine the target model based on the target product category; The third determining submodule is used to input the commercial radiation score and the first feature into the target model to determine the first score of each outlet; and The second determining module is used to determine the demographic information of each outlet within a second preset range based on the target product category; Recommendation modules include: The fourth determining submodule is used to determine the parameter information of each category parameter in the population profile information; Determine the weight of each category parameter; Based on the weight of each category parameter and the parameter information of each category parameter, a second score is determined for each outlet; The fifth determination submodule is used to determine a preset number of recommended outlets from the target outlet dataset based on the first and second scores of each outlet.

8. The apparatus according to claim 7, wherein, The encoding submodule is used for: The attribute information is feature-encoded to obtain the second feature of each network point; The second feature is subjected to feature cross-interaction to obtain the third feature of each network point; The interest point information is feature-encoded to obtain the fourth feature of each point; The third feature and the fourth feature are combined to obtain the first feature of each network point.

9. The apparatus according to claim 7 or 8, further comprising: The first acquisition module is used to acquire information about the registered outlets; The cleaning module is used to remove erroneous points from the already registered points based on the network point information and the point of interest information in the target map, and / or to merge duplicate points from the already registered points to obtain a dataset of registered points.

10. The apparatus according to claim 9, further comprising: The second acquisition module is used to acquire the initial network point dataset based on location information and network point category information; The elimination module is used to eliminate the corresponding network points in the initial network point dataset and the existing network point dataset according to the network point recommendation mode when the network point recommendation mode is to remove existing network points, so as to obtain the target network point dataset.

11. The apparatus according to claim 9, further comprising: The third acquisition module is used to acquire the initial network point dataset based on location information and network point category information; The tagging module is used to tag the corresponding outlets in the initial outlet dataset and the existing outlet dataset according to the already registered outlet dataset when the outlet recommendation mode is to retain the existing outlets, so as to obtain the target outlet dataset.

12. The apparatus according to claim 7 or 8, further comprising: The visualization module is used to visually display the recommended outlets in the form of an information table; And / or, the recommended locations can be visualized on a map.

13. An electronic device, comprising: At least one processor; as well as A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-6.

14. A non-transitory computer-readable storage medium storing computer instructions, wherein, The computer instructions are used to cause the computer to perform the method according to any one of claims 1-6.

15. A computer program product comprising a computer program that, when executed by a processor, implements the method according to any one of claims 1-6.

Images