Intelligent selection method and system for regional hot products and medium

Abstract

The invention provides an intelligent selection method and system for regional hot products and a medium. The method comprises the steps of obtaining user feature data and generating a user consumption feature portrait, then extracting user consumption intention data, obtaining sales data and social media feature data of each commodity in a preset time period in a preset area, and calculating to obtain a commodity popularity index; performing threshold comparison on the commodity popularity index to obtain an effective commodity popularity index, and matching effective hot commodity category data with user consumption intention data to obtain a matched consumer commodity category; according to the sales data, the location distance weight data, the effective commodity popularity index and the user consumption intention data, processing to obtain a hot commodity matching recommendation index, and performing threshold comparison on the hot commodity matching recommendation index to obtain a hot commodity recommendation state; therefore, through calculation of the commodity popularity index and the hot product matching recommendation index and threshold comparison, regional hot products can be pushed and selected more quickly and accurately.

Description

Technical Field

[0001] The present application relates to the field of regional hot product recommendation, and more specifically, to an intelligent recommendation method, system and medium for regional hot products. Background Art

[0002] Due to different regional environments or cultural backgrounds, different regions often have different hot-selling products. During the flow of people, the push of regional hot-selling products can help people understand the current trends and regional characteristics. With the rapid development of e-commerce and the increasing diversification of consumer demand, how to accurately recommend regional hot products to users has become an important issue. Traditional hot product recommendation methods often rely on manual experience or simple data analysis, and there are problems such as inaccurate, untimely, and lack of personalization. Therefore, an intelligent recommendation method is needed that can combine technologies such as big data analysis and user profiling to provide users with more accurate and personalized regional hot product recommendation services.

[0003] In view of the above problems, effective technical solutions are urgently needed. Summary of the invention

[0004] The purpose of this application is to provide a method, system and medium for intelligently recommending regional hot products, which can achieve faster and more accurate recommendation of regional hot products through calculation and threshold comparison of product popularity index and hot product matching recommendation index.

[0005] The present application also provides a method for intelligently recommending regional hot products, comprising the following steps:

[0006] Obtain user feature data, generate user consumption feature portraits based on the user feature data, and extract user consumption intention data based on the user consumption feature portraits;

[0007] Obtain sales information and social media feature information of each product in a preset area within a preset time period, and extract sales data and social media feature data respectively;

[0008] Calculate the commodity popularity index based on the sales data and social media feature data;

[0009] Compare the commodity heat index with a preset commodity heat index threshold to obtain a valid commodity heat index;

[0010] Acquire effective hot product category data corresponding to the effective product popularity index, and match the user consumption intention data with the effective hot product category data to obtain a matching consumption category;

[0011] Obtain the user's territorial distance weight data, and obtain the hot product matching recommendation index based on the sales data of the hot products in the matching consumer category combined with the territorial distance weight data, the effective product popularity index and the user's consumption intention data;

[0012] The hot product matching recommendation index is compared with a preset matching recommendation index threshold to obtain a hot product recommendation status, and hot products that meet the preset requirements are pushed to the user's smart terminal and displayed.

[0013] Optionally, in the method for intelligently recommending regional hot products described in the present application, the acquiring of user characteristic data, generating a user consumption characteristic portrait according to the user characteristic data, and extracting user consumption intention data according to the user consumption characteristic portrait specifically includes:

[0014] Obtain user feature data, including basic user information, historical shopping records, and online shopping and browsing records;

[0015] Generate a user consumption feature profile based on the user's basic information data, historical shopping record data, and online shopping and browsing record data;

[0016] The user consumption intention data is extracted based on the user consumption characteristic portrait, including single product consumption budget data, consumption intention category data and consumption willingness index data.

[0017] Optionally, in the method for intelligently recommending regional hot products described in the present application, the obtaining of sales information and social media feature information of each product in a preset area within a preset time period, and extracting the sales data and social media feature data respectively, specifically includes:

[0018] Obtain sales data of each commodity in a preset area within a preset time period, including sales volume data and sales amount data;

[0019] Get social media feature data for each product in a preset area within a preset time period, including the total number of social media searches, topic tag popularity data, and product support data.

[0020] Optionally, in the intelligent recommendation method for regional hot products described in the present application, the calculating and obtaining the product popularity index according to the sales data and social media feature data specifically includes:

[0021] Input the sales volume data, the total number of social media searches, the topic tag popularity data, and the product support volume data into a preset product popularity evaluation model for processing to obtain a product popularity index;

[0022] The calculation formula of the commodity heat index in the commodity heat evaluation model is:

[0023]

[0024] Among them, S r is the commodity popularity index, X l , S z , B r , Z s They are sales volume data, total number of social media searches, topic tag popularity data and product support data, and α1 and α2 are preset feature coefficients.

[0025] Optionally, in the intelligent recommendation method for regional hot products described in the present application, the comparing the product heat index with a preset product heat index threshold to obtain a valid product heat index specifically includes:

[0026] Compare the commodity heat index with a preset commodity heat index threshold to obtain a commodity heat index category, including a valid commodity heat index or an invalid commodity heat index;

[0027] If the commodity heat index is greater than or equal to the preset commodity heat index threshold, the commodity heat index category is a valid commodity heat index and is marked as reserved;

[0028] If the commodity heat index is less than the preset commodity heat index threshold, the commodity heat index category is invalid commodity heat index and will not be counted.

[0029] Optionally, in the intelligent recommendation method for regional hot products described in the present application, obtaining effective hot product category data corresponding to the effective product popularity index, matching the user consumption intention data and the effective hot product category data to obtain matching consumption categories, specifically includes:

[0030] Obtaining effective hot product category data corresponding to the effective product popularity index, and calculating the effective hot product category data with the consumer intention category data to obtain category matching index data;

[0031] Comparing the category matching index data with a preset category matching index threshold to obtain a category matching status, including a strong match and a weak match;

[0032] If the category matching index data is less than or equal to the preset category matching index threshold, the category matching status is a strong match, and the corresponding hot product category is marked as a matching consumption category;

[0033] If the category matching index data is greater than a preset category matching index threshold, the category matching status is a weak match, and the corresponding hot product category is marked as a non-matching consumer category.

[0034] Optionally, in the intelligent recommendation method for regional hot products described in the present application, the acquisition of the user's territorial distance weight data is performed, and the hot product matching recommendation index is obtained according to the sales data of the hot products in the matching consumer category combined with the territorial distance weight data, the effective product heat index and the user consumption intention data, specifically including:

[0035] Obtain the user's geographical distance weight data, and calculate the average sales price data based on the sales volume data and sales revenue data of the hot products in the matching consumer category;

[0036] Input the average sales price data, territorial distance weight data, single product consumption budget data and consumption willingness index data into a preset hot product recommendation index evaluation model for processing to obtain a hot product matching recommendation index;

[0037] The calculation formula of the hot product matching recommendation index in the hot product recommendation index evaluation model is:

[0038]

[0039] Among them, T j is the hot product matching recommendation index, D y , X j , Q z , Y y They are single product consumption budget data, average sales price data, territorial distance weight data and consumption willingness index data, and β1, β2 and β3 are preset characteristic coefficients.

[0040] Optionally, in the method for intelligently recommending regional hot products described in the present application, comparing the hot product matching recommendation index with a preset matching recommendation index threshold to obtain a hot product recommendation status, and pushing the hot products that meet the preset requirements to the user's smart terminal and displaying them specifically includes:

[0041] Compare the hot product matching recommendation index with a preset matching recommendation index threshold to obtain a hot product recommendation status, including recommended hot products and non-recommended hot products;

[0042] If the hot product matching recommendation index is greater than or equal to the preset matching recommendation index threshold, the hot product recommendation status is recommended hot product;

[0043] If the hot product matching recommendation index is less than the preset matching recommendation index threshold, the hot product recommendation status is a non-recommended hot product;

[0044] The recommended hot products are sorted from large to small according to the hot product matching recommendation index and pushed to the user's smart terminal in order for display.

[0045] In a second aspect, the present application provides a system for intelligently recommending regional hot products, the system comprising: a memory and a processor, wherein the memory stores a program of an intelligent method for recommending regional hot products, and when the program of the intelligent method for recommending regional hot products is executed by the processor, the following steps are implemented:

[0046] Obtain user feature data, generate user consumption feature portraits based on the user feature data, and extract user consumption intention data based on the user consumption feature portraits;

[0047] Obtain sales information and social media feature information of each product in a preset area within a preset time period, and extract sales data and social media feature data respectively;

[0048] Calculate the commodity popularity index based on the sales data and social media feature data;

[0049] Compare the commodity heat index with a preset commodity heat index threshold to obtain a valid commodity heat index;

[0050] Acquire effective hot product category data corresponding to the effective product popularity index, and match the user consumption intention data with the effective hot product category data to obtain a matching consumption category;

[0051] Obtain the user's territorial distance weight data, and obtain the hot product matching recommendation index based on the sales data of the hot products in the matching consumer category combined with the territorial distance weight data, the effective product popularity index and the user's consumption intention data;

[0052] The hot product matching recommendation index is compared with a preset matching recommendation index threshold to obtain a hot product recommendation status, and hot products that meet the preset requirements are pushed to the user's smart terminal and displayed.

[0053] In a third aspect, the present application also provides a readable storage medium, which stores a program for a method for intelligently selecting regional hot products. When the program for intelligently selecting regional hot products is executed by a processor, the steps of a method for intelligently selecting regional hot products as described in any one of the above items are implemented.

[0054] As can be seen from the above, the intelligent recommendation method, system and medium for regional hot products provided by the present application. The method obtains user feature data and generates a user consumption feature portrait, then extracts user consumption intention data, obtains the sales data and social media feature data of each commodity in a preset area within a preset time period, calculates the commodity heat index, compares the commodity heat index with a threshold to obtain an effective commodity heat index, matches the effective hot product category data with the user consumption intention data to obtain a matching consumption category, and then obtains a hot product matching recommendation index based on sales data, territorial distance weight data, effective commodity heat index and user consumption intention data, and then obtains a hot product matching recommendation index based on a threshold comparison of the hot product matching recommendation index to obtain a hot product recommendation status; thereby, through the calculation and threshold comparison of the commodity heat index and the hot product matching recommendation index, a faster and more accurate recommendation of regional hot products can be achieved.

[0055] Other features and advantages of the present application will be described in the following description, and partly become apparent from the description, or understood by practicing the present application. The purpose and other advantages of the present application can be realized and obtained by the structures particularly pointed out in the written description and the drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments of the present application will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present application 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 paying creative work.

[0057] Figure 1 A flow chart of a method for intelligently recommending regional hot products provided in an embodiment of the present application;

[0058] Figure 2 A flowchart of obtaining user consumption intention data of a method for intelligently recommending regional hot products provided in an embodiment of the present application;

[0059] Figure 3 A flowchart of obtaining sales data and social media feature data for a method for intelligently recommending regional hot products provided in an embodiment of the present application;

[0060] Figure 4 A flowchart of obtaining an effective commodity popularity index in a method for intelligently recommending regional hot commodities provided in an embodiment of the present application. DETAILED DESCRIPTION

[0061] The technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all of the embodiments. The components of the embodiments of the present application described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the application claimed for protection, but merely represents the selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without making creative work belong to the scope of protection of the present application.

[0062] It should be noted that similar reference numerals and letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings. At the same time, in the description of this application, the terms "first", "second", etc. are only used to distinguish the description and cannot be understood as indicating or implying relative importance.

[0063] Please refer to Figure 1 , Figure 1 1 is a flow chart of a method for intelligently recommending regional hot products in some embodiments of the present application. The method for intelligently recommending regional hot products is used in terminal devices, such as computers, mobile phones, etc. The method for intelligently recommending regional hot products includes the following steps:

[0064] S11, obtaining user characteristic data, generating a user consumption characteristic portrait according to the user characteristic data, and extracting user consumption intention data according to the user consumption characteristic portrait;

[0065] S12, obtaining sales information and social media feature information of each commodity in a preset area within a preset time period, and extracting the sales data and social media feature data respectively;

[0066] S13, calculating and obtaining a commodity popularity index according to the sales data and social media feature data;

[0067] S14, comparing the commodity heat index with a preset commodity heat index threshold to obtain a valid commodity heat index;

[0068] S15, obtaining effective hot product category data corresponding to the effective product popularity index, matching the user consumption intention data with the effective hot product category data, and obtaining a matching consumption category;

[0069] S16, obtaining the user's territorial distance weight data, and obtaining the hot product matching recommendation index according to the sales data of the hot products in the matching consumer category combined with the territorial distance weight data, the effective product popularity index and the user's consumption intention data;

[0070] S17: Compare the hot product matching recommendation index with a preset matching recommendation index threshold to obtain a hot product recommendation status, and push the hot products that meet the preset requirements to the user's smart terminal for display.

[0071] It should be noted that different users will have different personalized needs. When determining the user's consumption intention, it is necessary to first generate a user consumption characteristic portrait that conforms to the user itself based on the user characteristic data, and then extract the user's consumption intention data to understand the user's latest consumption intention or demand, and then obtain the product heat index through analysis of product sales data and social media data. The product heat index reflects the popularity of the product to a certain extent. The effective product heat index is obtained by comparing the threshold of the product heat index. The effective product heat index reflects the popularity of the product or the degree of attention. The products corresponding to the effective product heat index have category classification. The effective hot product category data can be extracted and matched with the user's consumption intention data to obtain the matching consumption category. In the matching consumption category, the hot product matching recommendation index is obtained based on the sales data, the territorial distance weight data, the effective product heat index and the user's consumption intention data. After the hot product matching recommendation index is compared with the threshold, the hot products that meet the requirements are pushed to the smart terminal for display.

[0072] Please refer to Figure 2 , Figure 2 This is a flowchart of obtaining user consumption intention data of a method for intelligently recommending regional hot products provided by an embodiment of the present application. According to an embodiment of the present invention, obtaining user feature data, generating a user consumption feature portrait based on the user feature data, and extracting user consumption intention data based on the user consumption feature portrait specifically includes:

[0073] S21, obtaining user characteristic data, including user basic information data, historical shopping record data, and online shopping and browsing record data;

[0074] S22, generating a user consumption feature portrait based on the user's basic information data, historical shopping record data, and online shopping and browsing record data;

[0075] S23. Extracting user consumption intention data based on the user consumption feature portrait, including single product consumption budget data, consumption intention category data and consumption willingness index data.

[0076] It should be noted that users are different individuals, so analysis based on individual situations is more meaningful and can more accurately grasp user needs and obtain feature data, including user basic situation data, historical shopping record data, and online shopping browsing record data, where the user basic situation data includes the user's age, gender, income level, and interests and hobbies, and the historical shopping records include offline shopping record data and online shopping record data; by analyzing the user's basic situation data, historical shopping record data, and online shopping browsing record data, a user consumption feature portrait is generated according to a preset data processing method, and the preset data processing method is obtained after training with a large amount of historical data; based on the user consumption feature portrait, user consumption intention data is extracted, including single product consumption budget data, which refers to the budget amount for purchasing a certain product or a type of product, and consumption intention category data refers to data on which type of product the user is more inclined to buy. The category is the classification of the product, which can be customized according to the user's needs. In this embodiment, the product is classified according to its purpose, i.e., cultural and creative products, food, etc.; the consumption willingness index is data indicating the degree to which the user is willing to consume.

[0077] Please refer to Figure 3 , Figure 3 This is a flow chart of obtaining sales data and social media feature data of a method for intelligently recommending regional hot products provided by an embodiment of the present application. According to an embodiment of the present invention, obtaining sales information and social media feature information of each commodity in a preset area within a preset time period, and extracting sales data and social media feature data respectively, specifically includes:

[0078] S31, obtaining sales data of each commodity in a preset area within a preset time period, including sales volume data and sales amount data;

[0079] S32: Obtaining social media feature data of each product in a preset area within a preset time period, including the total number of social media searches, topic tag popularity data, and product support data.

[0080] It should be noted that some regional hot products are regional and time-sensitive. In order to better reflect the latest hot products, it is necessary to obtain the sales data of each product in the preset area and preset time period. The preset area and preset time can be set according to user needs. The sales data of the product includes sales volume data and sales revenue data. At the same time, the social media feature data of each product in the preset area and preset time period is obtained, including the total number of social media searches, topic tag popularity data and product support data. The number of social media searches refers to the total number of times the product is searched in social media, the topic tag popularity data refers to the popularity data of the product being discussed or paid attention to as a topic tag, and the product support data refers to the sum of the number of likes and reposts of the product in social media.

[0081] According to an embodiment of the present invention, calculating and obtaining a commodity popularity index according to the sales data and the social media feature data specifically includes:

[0082] Input the sales volume data, the total number of social media searches, the topic tag popularity data, and the product support volume data into a preset product popularity evaluation model for processing to obtain a product popularity index;

[0083] The calculation formula of the commodity heat index in the commodity heat evaluation model is:

[0084]

[0085] Among them, S r is the commodity popularity index, X l , S z , B r , Z s They are sales volume data, total number of social media searches, topic tag popularity data and product support data. α1 and α2 are preset feature coefficients (the feature coefficients are obtained by querying the intelligent recommendation platform for hot products in the preset area).

[0086] It should be noted that both sales data and social media feature data can reflect the popularity of the product to a certain extent. In order to comprehensively consider the two factors, it is necessary to calculate the product popularity index. The product popularity index comprehensively reflects the popularity of the product among the public or its popularity to a certain extent. The higher the product popularity index, the more popular the product is. The product popularity evaluation model is obtained by training a large amount of historical data, and the product popularity index can be calculated through this model.

[0087] Please refer to Figure 4 , Figure 4 The flowchart of obtaining an effective commodity heat index of a method for intelligently recommending regional hot commodities provided by an embodiment of the present application is as follows. According to an embodiment of the present invention, the comparison of the commodity heat index with a preset commodity heat index threshold to obtain an effective commodity heat index specifically includes:

[0088] S41, comparing the commodity heat index with a preset commodity heat index threshold to obtain a commodity heat index category, including a valid commodity heat index or an invalid commodity heat index;

[0089] S42: If the commodity heat index is greater than or equal to a preset commodity heat index threshold, the commodity heat index category is a valid commodity heat index and is marked as reserved;

[0090] S43. If the commodity heat index is less than a preset commodity heat index threshold, the commodity heat index category is invalid commodity heat index and is not counted.

[0091] It should be noted that after calculating the commodity heat index, some commodities have a relatively small heat index, which means that they are not popular with the public. In order to better recommend hot commodities, it is necessary to compare the commodity heat index with the preset commodity heat index threshold to obtain the commodity heat index category. The commodity heat index category includes two types: one is the effective commodity heat index, in which case the commodity belongs to the hot commodity; the other is the invalid commodity heat index, in which case the commodity heat index is relatively small, which means that the commodity is not concerned or popular with the public, and the effective commodity heat index needs to be recommended to the user; in this embodiment, the commodity heat index threshold is set to: [0, 0.6), which is an invalid commodity heat index; [0.6, 1], which is a valid commodity heat index; when the commodity heat index is 0.9, the commodity heat index category is a valid commodity heat index.

[0092] According to an embodiment of the present invention, obtaining the effective hot product category data corresponding to the effective product heat index, matching the user consumption intention data with the effective hot product category data, and obtaining the matching consumption category specifically includes:

[0093] Obtaining effective hot product category data corresponding to the effective product popularity index, and calculating the effective hot product category data with the consumer intention category data to obtain category matching index data;

[0094] Comparing the category matching index data with a preset category matching index threshold to obtain a category matching status, including a strong match and a weak match;

[0095] If the category matching index data is less than or equal to the preset category matching index threshold, the category matching status is a strong match, and the corresponding hot product category is marked as a matching consumption category;

[0096] If the category matching index data is greater than a preset category matching index threshold, the category matching status is a weak match, and the corresponding hot product category is marked as a non-matching consumer category.

[0097] It should be noted that in order to achieve more accurate intelligent selection of hot products, we must first match the user's consumption intention category data with the effective hot product category data to obtain the hot product category of the area that the user prefers. The products corresponding to the effective product heat index will have categories, marked as effective hot product category data. The effective hot product category data is assigned according to the product category according to the preset rules. Different effective hot product category data represent different product classifications. The more similar the hot product categories are, the closer the effective hot product category data are. The calculation formula for the category matching index data is:

[0098]

[0099] Among them, P pis the category matching index data, Y x is the effective hot product category data, X yy It is the data of consumption intention categories;

[0100] The category matching index data reflects to a certain extent the degree of matching between the effective hot product category data and the consumer intention category. In this embodiment, the category matching index threshold is set as follows: when the category matching index is less than or equal to 0.3, the category matching status is a strong match; when the category matching index is greater than 0.3, the category matching status is a weak match; when the category matching index data is 0.2, the category matching status is a strong match.

[0101] According to an embodiment of the present invention, the acquisition of the user's territorial distance weight data, and the processing of the sales data of the hot products in the matching consumer category in combination with the territorial distance weight data, the effective commodity popularity index and the user's consumption intention data to obtain the hot product matching recommendation index specifically include:

[0102] Obtain the user's geographical distance weight data, and calculate the average sales price data based on the sales volume data and sales revenue data of the hot products in the matching consumer category;

[0103] Input the average sales price data, territorial distance weight data, single product consumption budget data and consumption willingness index data into a preset hot product recommendation index evaluation model for processing to obtain a hot product matching recommendation index;

[0104] The calculation formula of the hot product matching recommendation index in the hot product recommendation index evaluation model is:

[0105]

[0106] Among them, T j is the hot product matching recommendation index, D y , X j , Q z , Y y They are single product consumption budget data, average sales price data, territorial distance weight data and consumption willingness index data. β1, β2 and β3 are preset characteristic coefficients (the characteristic coefficients are obtained by querying the intelligent recommendation platform for hot products in the preset area).

[0107] It should be noted that when the user has been in a certain place, the hot products in that area will basically be recommended. When the user's location changes, the system can recommend new hot products based on the user's location. Therefore, it is necessary to obtain the user's territorial distance weight data. The territorial distance weight data refers to the weight data of the impact of the distance between the user's permanent location and the current location on the hot product recommendation. After the user's location changes, he or she is more inclined to buy hot products in the new area, that is, the greater the weight; but when the user has been in an unchanged location, the territorial distance weight data will be small; dividing the sales data by the sales volume data will give the average selling price data. The more sufficient the budget, the stronger the intention to buy, and therefore the greater the need for recommendation; the hot product matching recommendation index reflects to a certain extent the degree of match between the recommended hot products and the hot products that the user prefers. The larger the hot product matching recommendation index, the more it should be recommended.

[0108] According to an embodiment of the present invention, comparing the hot product matching recommendation index with a preset matching recommendation index threshold to obtain a hot product recommendation status, and pushing hot products that meet the preset requirements to a user smart terminal and displaying them specifically includes:

[0109] Compare the hot product matching recommendation index with a preset matching recommendation index threshold to obtain a hot product recommendation status, including recommended hot products and non-recommended hot products;

[0110] If the hot product matching recommendation index is greater than or equal to the preset matching recommendation index threshold, the hot product recommendation status is recommended hot product;

[0111] If the hot product matching recommendation index is less than the preset matching recommendation index threshold, the hot product recommendation status is a non-recommended hot product;

[0112] The recommended hot products are sorted from large to small according to the hot product matching recommendation index and pushed to the user's smart terminal in order for display.

[0113] Among them, after obtaining the hot product matching recommendation index, not all hot products will be recommended. The hot product recommendation status can be obtained by comparing the hot product matching recommendation index with the preset matching recommendation index threshold, including recommended hot products and non-recommended hot products. After selecting the regional hot products with the recommended hot product status, they are sorted in descending order according to the hot product matching recommendation index, and pushed to the user's smart terminal for display, and the ones ranked in front will be recommended first; in this embodiment, the matching recommendation index threshold is set to: [0,0.7), which is a non-recommended hot product; [0.7, 1], which is a recommended hot product; when the hot product matching recommendation index is 0.9, the hot product recommendation status is a recommended hot product.

[0114] The present invention also discloses a system for intelligently recommending regional hot products, including a memory and a processor. The memory stores a program for intelligently recommending regional hot products. When the program for intelligently recommending regional hot products is executed by the processor, the following steps are implemented:

[0115] Obtain user feature data, generate user consumption feature portraits based on the user feature data, and extract user consumption intention data based on the user consumption feature portraits;

[0116] Obtain sales information and social media feature information of each product in a preset area within a preset time period, and extract sales data and social media feature data respectively;

[0117] Calculate the commodity popularity index based on the sales data and social media feature data;

[0118] Compare the commodity heat index with a preset commodity heat index threshold to obtain a valid commodity heat index;

[0119] Acquire effective hot product category data corresponding to the effective product popularity index, and match the user consumption intention data with the effective hot product category data to obtain a matching consumption category;

[0120] Obtain the user's territorial distance weight data, and obtain the hot product matching recommendation index based on the sales data of the hot products in the matching consumer category combined with the territorial distance weight data, the effective product popularity index and the user's consumption intention data;

[0121] The hot product matching recommendation index is compared with a preset matching recommendation index threshold to obtain a hot product recommendation status, and hot products that meet the preset requirements are pushed to the user's smart terminal and displayed.

[0122] It should be noted that different users will have different personalized needs. When determining the user's consumption intention, it is necessary to first generate a user consumption characteristic portrait that conforms to the user itself based on the user characteristic data, and then extract the user's consumption intention data to understand the user's latest consumption intention or demand, and then obtain the product heat index through analysis of product sales data and social media data. The product heat index reflects the popularity of the product to a certain extent. The effective product heat index is obtained by comparing the threshold of the product heat index. The effective product heat index reflects the popularity of the product or the degree of attention. The products corresponding to the effective product heat index have category classification. The effective hot product category data can be extracted and matched with the user's consumption intention data to obtain the matching consumption category. In the matching consumption category, the hot product matching recommendation index is obtained based on the sales data, the territorial distance weight data, the effective product heat index and the user's consumption intention data. After the hot product matching recommendation index is compared with the threshold, the hot products that meet the requirements are pushed to the smart terminal for display.

[0123] According to an embodiment of the present invention, the step of obtaining user characteristic data, generating a user consumption characteristic portrait according to the user characteristic data, and extracting user consumption intention data according to the user consumption characteristic portrait specifically includes:

[0124] Obtain user feature data, including basic user information, historical shopping records, and online shopping and browsing records;

[0125] Generate a user consumption feature profile based on the user's basic information data, historical shopping record data, and online shopping and browsing record data;

[0126] The user consumption intention data is extracted based on the user consumption characteristic portrait, including single product consumption budget data, consumption intention category data and consumption willingness index data.

[0127] It should be noted that users are different individuals, so analysis based on individual situations is more meaningful and can more accurately grasp user needs and obtain feature data, including user basic situation data, historical shopping record data, and online shopping browsing record data, where the user basic situation data includes the user's age, gender, income level, and interests and hobbies, and the historical shopping records include offline shopping record data and online shopping record data; by analyzing the user's basic situation data, historical shopping record data, and online shopping browsing record data, a user consumption feature portrait is generated according to a preset data processing method, and the preset data processing method is obtained after training with a large amount of historical data; based on the user consumption feature portrait, user consumption intention data is extracted, including single product consumption budget data, which refers to the budget amount for purchasing a certain product or a type of product, and consumption intention category data refers to data on which type of product the user is more inclined to buy. The category is the classification of the product, which can be customized according to the user's needs. In this embodiment, the product is classified according to its purpose, i.e., cultural and creative products, food, etc.; the consumption willingness index is data indicating the degree to which the user is willing to consume.

[0128] According to an embodiment of the present invention, the obtaining of sales information and social media feature information of each commodity in a preset area within a preset time period, and extracting the sales data and social media feature data respectively, specifically includes:

[0129] Obtain sales data of each commodity in a preset area within a preset time period, including sales volume data and sales amount data;

[0130] Get social media feature data for each product in a preset area within a preset time period, including the total number of social media searches, topic tag popularity data, and product support data.

[0131] It should be noted that some regional hot products are regional and time-sensitive. In order to better reflect the latest hot products, it is necessary to obtain the sales data of each product in the preset area and preset time period. The preset area and preset time can be set according to user needs. The sales data of the product includes sales volume data and sales revenue data. At the same time, the social media feature data of each product in the preset area and preset time period is obtained, including the total number of social media searches, topic tag popularity data and product support data. The number of social media searches refers to the total number of times the product is searched in social media, the topic tag popularity data refers to the popularity data of the product being discussed or paid attention to as a topic tag, and the product support data refers to the sum of the number of likes and reposts of the product in social media.

[0132] According to an embodiment of the present invention, calculating and obtaining a commodity popularity index according to the sales data and the social media feature data specifically includes:

[0133] Input the sales volume data, the total number of social media searches, the topic tag popularity data, and the product support volume data into a preset product popularity evaluation model for processing to obtain a product popularity index;

[0134] The calculation formula of the commodity heat index in the commodity heat evaluation model is:

[0135]

[0136] Among them, S r is the commodity popularity index, X l , S z , B r , Z s They are sales volume data, total number of social media searches, topic tag popularity data and product support data. α1 and α2 are preset feature coefficients (the feature coefficients are obtained by querying the intelligent recommendation platform for hot products in the preset area).

[0137] It should be noted that both sales data and social media feature data can reflect the popularity of the product to a certain extent. In order to comprehensively consider the two factors, it is necessary to calculate the product popularity index. The product popularity index comprehensively reflects the popularity of the product among the public or its popularity to a certain extent. The higher the product popularity index, the more popular the product is. The product popularity evaluation model is obtained by training a large amount of historical data, and the product popularity index can be calculated through this model.

[0138] According to an embodiment of the present invention, comparing the commodity heat index with a preset commodity heat index threshold to obtain a valid commodity heat index specifically includes:

[0139] Compare the commodity heat index with a preset commodity heat index threshold to obtain a commodity heat index category, including a valid commodity heat index or an invalid commodity heat index;

[0140] If the commodity heat index is greater than or equal to the preset commodity heat index threshold, the commodity heat index category is a valid commodity heat index and is marked as reserved;

[0141] If the commodity heat index is less than the preset commodity heat index threshold, the commodity heat index category is invalid commodity heat index and will not be counted.

[0142] It should be noted that after calculating the commodity heat index, some commodities have a relatively small heat index, which means that they are not popular with the public. In order to better recommend hot commodities, it is necessary to compare the commodity heat index with the preset commodity heat index threshold to obtain the commodity heat index category. The commodity heat index category includes two types: one is the effective commodity heat index, in which case the commodity belongs to the hot commodity; the other is the invalid commodity heat index, in which case the commodity heat index is relatively small, which means that the commodity is not concerned or popular with the public, and the effective commodity heat index needs to be recommended to the user; in this embodiment, the commodity heat index threshold is set to: [0, 0.6), which is an invalid commodity heat index; [0.6, 1], which is a valid commodity heat index; when the commodity heat index is 0.9, the commodity heat index category is a valid commodity heat index.

[0143] According to an embodiment of the present invention, obtaining the effective hot product category data corresponding to the effective product heat index, matching the user consumption intention data with the effective hot product category data, and obtaining the matching consumption category specifically includes:

[0144] Obtaining effective hot product category data corresponding to the effective product popularity index, and calculating the effective hot product category data with the consumer intention category data to obtain category matching index data;

[0145] Comparing the category matching index data with a preset category matching index threshold to obtain a category matching status, including a strong match and a weak match;

[0146] If the category matching index data is less than or equal to the preset category matching index threshold, the category matching status is a strong match, and the corresponding hot product category is marked as a matching consumption category;

[0147] If the category matching index data is greater than a preset category matching index threshold, the category matching status is a weak match, and the corresponding hot product category is marked as a non-matching consumer category.

[0148] It should be noted that in order to achieve more accurate intelligent selection of hot products, we must first match the user's consumption intention category data with the effective hot product category data to obtain the hot product category of the area that the user prefers. The products corresponding to the effective product heat index will have categories, marked as effective hot product category data. The effective hot product category data is assigned according to the product category according to the preset rules. Different effective hot product category data represent different product classifications. The more similar the hot product categories are, the closer the effective hot product category data are. The calculation formula for the category matching index data is:

[0149]

[0150] Among them, P p is the category matching index data, Y x is the effective hot product category data, X yy It is the data of consumption intention categories;

[0151] The category matching index data reflects to a certain extent the degree of matching between the effective hot product category data and the consumer intention category. In this embodiment, the category matching index threshold is set as follows: when the category matching index is less than or equal to 0.3, the category matching status is a strong match; when the category matching index is greater than 0.3, the category matching status is a weak match; when the category matching index data is 0.2, the category matching status is a strong match.

[0152] According to an embodiment of the present invention, the acquisition of the user's territorial distance weight data, and the processing of the sales data of the hot products in the matching consumer category in combination with the territorial distance weight data, the effective commodity popularity index and the user's consumption intention data to obtain the hot product matching recommendation index specifically include:

[0153] Obtain the user's geographical distance weight data, and calculate the average sales price data based on the sales volume data and sales revenue data of the hot products in the matching consumer category;

[0154] Input the average sales price data, territorial distance weight data, single product consumption budget data and consumption willingness index data into a preset hot product recommendation index evaluation model for processing to obtain a hot product matching recommendation index;

[0155] The calculation formula of the hot product matching recommendation index in the hot product recommendation index evaluation model is:

[0156]

[0157] Among them, T j is the hot product matching recommendation index, D y , X j , Q z , Y yThey are single product consumption budget data, average sales price data, territorial distance weight data and consumption willingness index data. β1, β2 and β3 are preset characteristic coefficients (the characteristic coefficients are obtained by querying the intelligent recommendation platform for hot products in the preset area).

[0158] It should be noted that when the user has been in a certain place, the hot products in that area will basically be recommended. When the user's location changes, the system can recommend new hot products based on the user's location. Therefore, it is necessary to obtain the user's territorial distance weight data. The territorial distance weight data refers to the weight data of the impact of the distance between the user's permanent location and the current location on the hot product recommendation. After the user's location changes, he or she is more inclined to buy hot products in the new area, that is, the greater the weight; but when the user has been in an unchanged location, the territorial distance weight data will be small; dividing the sales data by the sales volume data will give the average selling price data. The more sufficient the budget, the stronger the intention to buy, and therefore the greater the need for recommendation; the hot product matching recommendation index reflects to a certain extent the degree of match between the recommended hot products and the hot products that the user prefers. The larger the hot product matching recommendation index, the more it should be recommended.

[0159] According to an embodiment of the present invention, comparing the hot product matching recommendation index with a preset matching recommendation index threshold to obtain a hot product recommendation status, and pushing hot products that meet the preset requirements to a user smart terminal and displaying them specifically includes:

[0160] Compare the hot product matching recommendation index with a preset matching recommendation index threshold to obtain a hot product recommendation status, including recommended hot products and non-recommended hot products;

[0161] If the hot product matching recommendation index is greater than or equal to the preset matching recommendation index threshold, the hot product recommendation status is recommended hot product;

[0162] If the hot product matching recommendation index is less than the preset matching recommendation index threshold, the hot product recommendation status is a non-recommended hot product;

[0163] The recommended hot products are sorted from large to small according to the hot product matching recommendation index and pushed to the user's smart terminal in order for display.

[0164] Among them, after obtaining the hot product matching recommendation index, not all hot products will be recommended. The hot product recommendation status can be obtained by comparing the hot product matching recommendation index with the preset matching recommendation index threshold, including recommended hot products and non-recommended hot products. After selecting the regional hot products with the recommended hot product status, they are sorted in descending order according to the hot product matching recommendation index, and pushed to the user's smart terminal for display, and the ones ranked in front will be recommended first; in this embodiment, the matching recommendation index threshold is set to: [0,0.7), which is a non-recommended hot product; [0.7, 1], which is a recommended hot product; when the hot product matching recommendation index is 0.9, the hot product recommendation status is a recommended hot product.

[0165] A third aspect of the present invention provides a readable storage medium, which includes a program for a method for intelligently selecting regional hot products. When the program for intelligently selecting regional hot products is executed by a processor, the steps of a method for intelligently selecting regional hot products as described in any one of the above items are implemented.

[0166] The present invention discloses an intelligent recommendation method, system and medium for regional hot products. The method obtains user feature data and generates a user consumption feature portrait, then extracts user consumption intention data, obtains sales data and social media feature data of each product in a preset area within a preset time period, calculates a product heat index, compares the product heat index with a threshold to obtain an effective product heat index, matches the effective hot product category data with the user consumption intention data to obtain a matching consumption category, and then obtains a hot product matching recommendation index based on sales data, territorial distance weight data, the effective product heat index and the user consumption intention data, and then obtains a hot product recommendation status after comparing the hot product matching recommendation index with a threshold; thereby, faster and more accurate recommendation of regional hot products can be achieved through calculation and threshold comparison of the product heat index and the hot product matching recommendation index.

[0167] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as: multiple units or components can be combined, or can be integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the components shown or discussed can be through some interfaces, and the indirect coupling or communication connection of the devices or units can be electrical, mechanical or other forms.

[0168] The units described above as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units; they may be located in one place or distributed on multiple network units; some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0169] In addition, all functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may be separately used as a unit, or two or more units may be integrated into one unit; the above-mentioned integrated units may be implemented in the form of hardware or in the form of hardware plus software functional units.

[0170] Those skilled in the art can understand that: all or part of the steps of implementing the above method embodiments can be completed by hardware related to program instructions, and the aforementioned program can be stored in a readable storage medium. When the program is executed, it executes the steps of the above method embodiments; and the aforementioned storage medium includes: mobile storage devices, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), disks or optical disks, and other media that can store program codes.

[0171] Alternatively, if the above-mentioned integrated unit of the present invention is implemented in the form of a software function module and sold or used as an independent product, it can also be stored in a readable storage medium. Based on this understanding, the technical solution of the embodiment of the present invention can be essentially or partly reflected in the form of a software product that contributes to the prior art. The software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the methods described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as mobile storage devices, ROM, RAM, magnetic disks or optical disks.

Claims

1. An intelligent method for recommending regional hot products, characterized in that: include: Obtain user feature data, generate user consumption feature portraits based on the user feature data, and extract user consumption intention data based on the user consumption feature portraits; Obtain sales information and social media feature information of each product in a preset area within a preset time period, and extract sales data and social media feature data respectively; Calculate the commodity popularity index based on the sales data and social media feature data; Compare the commodity heat index with a preset commodity heat index threshold to obtain a valid commodity heat index; Acquire effective hot product category data corresponding to the effective product popularity index, and match the user consumption intention data with the effective hot product category data to obtain a matching consumption category; Obtain the user's territorial distance weight data, and obtain the hot product matching recommendation index based on the sales data of the hot products in the matching consumer category combined with the territorial distance weight data, the effective product popularity index and the user's consumption intention data; The hot product matching recommendation index is compared with a preset matching recommendation index threshold to obtain a hot product recommendation status, and hot products that meet the preset requirements are pushed to the user's smart terminal and displayed.

2. The intelligent recommendation method for regional hot products according to claim 1 is characterized in that: The obtaining of user characteristic data, generating a user consumption characteristic portrait according to the user characteristic data, and extracting user consumption intention data according to the user consumption characteristic portrait specifically includes: Obtain user feature data, including basic user information, historical shopping records, and online shopping and browsing records; Generate a user consumption feature profile based on the user's basic information data, historical shopping record data, and online shopping and browsing record data; The user consumption intention data is extracted based on the user consumption characteristic portrait, including single product consumption budget data, consumption intention category data and consumption willingness index data.

3. The intelligent recommendation method for regional hot products according to claim 2 is characterized in that: The obtaining of sales information and social media feature information of each commodity in a preset area within a preset time period, and extracting the sales data and social media feature data respectively, specifically includes: Obtain sales data of each commodity in a preset area within a preset time period, including sales volume data and sales amount data; Get social media feature data for each product in a preset area within a preset time period, including the total number of social media searches, topic tag popularity data, and product support data.

4. The intelligent recommendation method for regional hot products according to claim 3 is characterized in that: The calculating and obtaining the commodity popularity index according to the sales data and the social media feature data specifically includes: Input the sales volume data, the total number of social media searches, the topic tag popularity data, and the product support volume data into a preset product popularity evaluation model for processing to obtain a product popularity index; The calculation formula of the commodity heat index in the commodity heat evaluation model is: Among them, S r is the commodity popularity index, X l , S z , B r , Z s They are sales volume data, total number of social media searches, topic tag popularity data and product support data, and α1 and α2 are preset feature coefficients.

5. The intelligent recommendation method for regional hot products according to claim 4 is characterized in that: The comparing the commodity heat index with a preset commodity heat index threshold to obtain a valid commodity heat index specifically includes: Compare the commodity heat index with a preset commodity heat index threshold to obtain a commodity heat index category, including a valid commodity heat index or an invalid commodity heat index; If the commodity heat index is greater than or equal to the preset commodity heat index threshold, the commodity heat index category is a valid commodity heat index and is marked as reserved; If the commodity heat index is less than the preset commodity heat index threshold, the commodity heat index category is invalid commodity heat index and will not be counted.

6. The intelligent recommendation method for regional hot products according to claim 5 is characterized in that: The obtaining of effective hot product category data corresponding to the effective product heat index, and matching the user consumption intention data with the effective hot product category data to obtain a matching consumption category specifically includes: Obtaining effective hot product category data corresponding to the effective product popularity index, and calculating the effective hot product category data with the consumer intention category data to obtain category matching index data; Comparing the category matching index data with a preset category matching index threshold to obtain a category matching status, including a strong match and a weak match; If the category matching index data is less than or equal to the preset category matching index threshold, the category matching status is a strong match, and the corresponding hot product category is marked as a matching consumption category; If the category matching index data is greater than a preset category matching index threshold, the category matching status is a weak match, and the corresponding hot product category is marked as a non-matching consumer category.

7. The intelligent recommendation method for regional hot products according to claim 6 is characterized in that: The step of obtaining the user's territorial distance weight data and obtaining the hot product matching recommendation index according to the sales data of the hot products in the matching consumer category combined with the territorial distance weight data, the effective product popularity index and the user's consumption intention data specifically includes: Obtain the user's geographical distance weight data, and calculate the average sales price data based on the sales volume data and sales revenue data of the hot products in the matching consumer category; Input the average sales price data, territorial distance weight data, single product consumption budget data and consumption willingness index data into a preset hot product recommendation index evaluation model for processing to obtain a hot product matching recommendation index; The calculation formula of the hot product matching recommendation index in the hot product recommendation index evaluation model is: Among them, T j is the hot product matching recommendation index, D y , X j , Q z , Y y They are single product consumption budget data, average sales price data, territorial distance weight data and consumption willingness index data, and β1, β2 and β3 are preset characteristic coefficients.

8. The intelligent recommendation method for regional hot products according to claim 7 is characterized in that: The step of comparing the hot product matching recommendation index with a preset matching recommendation index threshold to obtain a hot product recommendation status, and pushing hot products that meet the preset requirements to the user's smart terminal for display specifically includes: Compare the hot product matching recommendation index with a preset matching recommendation index threshold to obtain a hot product recommendation status, including recommended hot products and non-recommended hot products; If the hot product matching recommendation index is greater than or equal to the preset matching recommendation index threshold, the hot product recommendation status is recommended hot product; If the hot product matching recommendation index is less than the preset matching recommendation index threshold, the hot product recommendation status is a non-recommended hot product; The recommended hot products are sorted from large to small according to the hot product matching recommendation index and pushed to the user's smart terminal in order for display.

9. An intelligent recommendation system for regional hot products, characterized in that: The system comprises a memory and a processor, wherein the memory comprises a method program for intelligently recommending regional hot products, and when the method program for intelligently recommending regional hot products is executed by the processor, the following steps are implemented: Obtain user feature data, generate user consumption feature portraits based on the user feature data, and extract user consumption intention data based on the user consumption feature portraits; Obtain sales information and social media feature information of each product in a preset area within a preset time period, and extract sales data and social media feature data respectively; Calculate the commodity popularity index based on the sales data and social media feature data; Compare the commodity heat index with a preset commodity heat index threshold to obtain a valid commodity heat index; Acquire effective hot product category data corresponding to the effective product popularity index, and match the user consumption intention data with the effective hot product category data to obtain a matching consumption category; Obtain the user's territorial distance weight data, and obtain the hot product matching recommendation index based on the sales data of the hot products in the matching consumer category combined with the territorial distance weight data, the effective product popularity index and the user's consumption intention data; The hot product matching recommendation index is compared with a preset matching recommendation index threshold to obtain a hot product recommendation status, and hot products that meet the preset requirements are pushed to the user's smart terminal and displayed.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a method program for intelligently recommending regional hot products. When the method program for intelligently recommending regional hot products is executed by a processor, the steps of the method for intelligently recommending regional hot products as described in any one of claims 1 to 8 are implemented.

Images