CN112801761A

CN112801761A - Commodity recommendation method, computing device and computer-readable storage medium

Info

Publication number: CN112801761A
Application number: CN202110391679.1A
Authority: CN
Inventors: 孙盼盼; 张诗韵
Original assignee: Zhongzhi Guanaitong Shanghai Technology Co ltd; Zhongzhi Aiyoutong Nanjing Information Technology Co ltd
Current assignee: Zhongzhi Guanaitong Shanghai Technology Co ltd; Zhongzhi Aiyoutong Nanjing Information Technology Co ltd
Priority date: 2021-04-13
Filing date: 2021-04-13
Publication date: 2021-05-14

Abstract

The invention provides a commodity recommendation method, a computing device and a computer-readable storage medium. The method comprises the following steps: determining an organization label of the user based on the user identification information, wherein the organization label is used for indicating an organization to which the user belongs; determining whether the user has personal pattern data based on the user identification information, the personal pattern data being generated based on historical operational behavior of the user; if the user is determined not to have the personal pattern data, determining a first recommendation value of each of the plurality of commodities as a recommendation value of the commodity based on the organization pattern data of the associated organization associated with the organization tag; and if it is determined that the user has personal pattern data, determining a first recommendation value for each of a plurality of items based on the organization pattern data of the associated organization associated with the organization tag, determining a second recommendation value for the item based on the personal pattern data, and determining a recommendation value for the item based on the first recommendation value and the second recommendation value.

Description

Commodity recommendation method, computing device and computer-readable storage medium

Technical Field

The present invention relates generally to the field of computer software, and more particularly, to a merchandise recommendation method, a computing device, and a computer-readable storage medium.

Background

With the continuous development of networks, more and more users meet shopping demands through the e-commerce system. After a user conducts browsing, shopping (adding to a shopping cart) or purchasing operation on a specific commodity in the e-commerce system, the e-commerce system can record historical operation behaviors of the user and determine mode data of the user based on the historical operation behaviors.

When a user first accesses the e-commerce system, the e-commerce system will typically present a default merchandise page generated based on a default recommendation algorithm of the system for the user, and all of the merchandise presented in the default merchandise page and the ordering thereof are the same for all first-accessed users.

However, the uniform cold start method does not well meet the personalized demands of different users. In particular, in an e-commerce system for an enterprise user, the enterprise user may preset individual accounts for a plurality of employees thereof in advance for the employees thereof to search and purchase appropriate goods in the e-commerce system, respectively, and the goods of interest of the employees in the e-commerce system may have a certain correlation. For example, in an e-commerce system that provides supply chain procurement services for employees of a business, supply chain items of interest to employees of the same business or similar businesses may have some similarity. For another example, in an e-commerce system providing a physical goods or service goods purchase service for employees of a business, the physical goods or service goods that are of interest to employees of the same business or similar businesses may have certain similarities (e.g., employees of the same type of chemical industry may be interested in the same labor insurance goods or physical goods), and so on.

In this case, when recommending goods to the user, if the organization information of the user can be combined, the accuracy of recommending goods will be greatly improved.

Disclosure of Invention

In view of the above problems, the present invention provides a product recommendation scheme, in which by using organization pattern data of an organization related to a user (including an organization to which the user belongs or similar organizations) to perform product recommendation for the user, accuracy of product recommendation can be improved, and particularly in the case of a cold start of the user, recommended products can be made to better meet user requirements.

According to an aspect of the present invention, there is provided a commodity recommendation method. The method comprises the following steps: determining an organization label of a user based on user identification information, the organization label being used for indicating an organization to which the user belongs; determining whether the user has personal pattern data based on the user identification information, the personal pattern data being generated based on historical operational behavior of the user; if the user is determined not to have the personal pattern data, determining a first recommendation value of each of a plurality of commodities as a recommendation value of the commodity based on organization pattern data of an associated organization associated with the organization tag; and if it is determined that the user has personal pattern data, determining a first recommendation value for each of a plurality of items based on organization pattern data of an associated organization associated with the organization tag, determining a second recommendation value for the item based on the personal pattern data, and determining a recommendation value for the item based on the first and second recommendation values.

According to another aspect of the invention, a computing device is provided. The computing device includes: at least one processor; and at least one memory coupled to the at least one processor and storing instructions for execution by the at least one processor, the instructions when executed by the at least one processor causing the computing device to perform steps according to the above-described method.

According to yet another aspect of the present invention, a computer-readable storage medium is provided, having stored thereon computer program code, which when executed performs the method as described above.

In some embodiments, the organization pattern data includes a recommended category coefficient for the associated organization and a recommended brand coefficient for the associated organization, wherein determining a first recommendation value for each of a plurality of items based on the organization pattern data for the associated organization associated with the organization tag includes: determining an organization category recommendation value of the commodity based on the recommendation category coefficient and the organization category recommendation basis coefficient of the commodity in the associated organization; determining an organization brand recommendation value of the commodity based on the recommended brand coefficient and the organization brand recommendation basis coefficient of the commodity in the associated organization; and determining a first recommendation value for the good based on the organization category recommendation value, the category weight, the organization brand recommendation value, and the brand weight for the good.

In some embodiments, the organization pattern data further comprises a white list category coefficient for the associated organization and a white list brand coefficient for the associated organization, wherein determining a first recommendation value for each of a plurality of items based on the organization pattern data for the associated organization associated with the organization tag comprises: determining an organization category recommendation value for the commodity based on the recommendation category coefficient for the commodity in the associated organization, the white list category coefficient in the associated organization, and the organization category recommendation basis coefficient; and determining an organization brand recommendation value for the good based on the recommended brand coefficient for the associated organization, the whitelist brand coefficient for the associated organization, and the organization brand recommendation value for the good.

In some embodiments, the personal pattern data includes a high frequency category coefficient for the user, a high frequency brand coefficient for the user, wherein determining the second recommendation value for the good based on the personal pattern data includes: determining a personal category recommendation value of the commodity based on the high-frequency category coefficient and the personal category recommendation basis coefficient of the user; determining a personal brand recommendation value for the good based on the high frequency brand coefficient and the personal brand recommendation base coefficient of the user; and determining a second recommendation value for the good based on the personal category recommendation value, the category weight, the personal brand recommendation value, and the brand weight for the good.

In some embodiments, the organization pattern data further includes an organization price range for the associated organization, the personal pattern data further includes a personal price range for the user, and the method further includes: determining a recommended price range for the user based on the organization price range for the associated organization and the personal price range for the user; determining whether the price of the commodity is in a recommended price range of the user; if the price of the commodity is in the recommended price range of the user, updating the recommended value of the commodity based on the first price weight; and if the price of the commodity is not in the recommended price range of the user, updating the recommended value of the commodity based on a second price weight, wherein the second price weight is lower than the first price weight.

In some embodiments, determining the recommended price range for the user based on the organizational price range for the associated organization and the personal price range for the user comprises: and taking a union of the organization price range of the associated organization and the personal price range of the user to determine the recommended price range of the user.

In some embodiments, the method further comprises: determining a default recommendation weight of the commodity, wherein the default recommendation weight is based on a default recommendation algorithm of a commodity recommendation system; and updating the recommended value of the commodity based on the default recommended weight of the commodity.

In some embodiments, the method further comprises: sorting the plurality of commodities based on the recommended values of the plurality of commodities to display the plurality of commodities to the user.

In some embodiments, wherein the associated organization comprises an organization similar to an organization to which the user belongs, and the method further comprises: obtaining organization information of a plurality of candidate organizations from an organization database, wherein the organization information comprises at least one of organization properties, industry types, organization sizes and places of the candidate organizations; comparing the tissue information of the plurality of candidate tissues with the tissue information of the tissue to which the user belongs respectively to determine a similarity coefficient between each of the plurality of candidate tissues and the tissue to which the user belongs; and selecting one candidate tissue with the highest similarity coefficient from the plurality of candidate tissues as the associated tissue.

In some embodiments, determining a similarity coefficient for each of the plurality of candidate tissues to the tissue to which the user belongs comprises: setting a first similarity value for the candidate tissue based on whether the tissue property of the candidate tissue is the same as the tissue property of the tissue to which the user belongs; setting a second similarity value for the candidate organization based on whether the industry type of the candidate organization is the same as the industry type of the organization to which the user belongs; setting a third similarity value for the candidate tissue based on whether the tissue scale of the candidate tissue is the same as the tissue scale of the tissue to which the user belongs; setting a fourth similarity value for the candidate organization based on whether the region where the candidate organization is located is the same as the region where the organization the user belongs to; and determining a similarity coefficient between the candidate tissue and the tissue to which the user belongs based on the first similarity value, the second similarity value, the third similarity value, and the fourth similarity value.

Drawings

The invention will be better understood and other objects, details, features and advantages thereof will become more apparent from the following description of specific embodiments of the invention given with reference to the accompanying drawings.

Fig. 1 shows a schematic diagram of a system for implementing an article recommendation method according to an embodiment of the invention.

FIG. 2 illustrates a flow diagram of an item recommendation method according to some embodiments of the invention.

FIG. 3 shows a flowchart of steps for determining a recommendation value for each of a plurality of items based on organizational schema data of an associated organization associated with an organizational tag, according to one embodiment of the invention.

FIG. 4 shows a flowchart of steps for determining a recommendation value for an item based on organization pattern data for an associated organization associated with an organization tag and personal pattern data for a user, according to one embodiment of the invention.

FIG. 5 shows a flowchart of substeps of determining a second recommendation value for an item based on a user's personal pattern data, according to one embodiment of the invention.

FIG. 6 shows a flowchart of steps for adjusting the recommendation value for an item, according to one embodiment of the invention.

FIG. 7 shows a flowchart of steps for determining an associated organization associated with a user's organization tag, in accordance with one embodiment of the invention.

FIG. 8 illustrates a block diagram of a computing device suitable for implementing embodiments of the present invention.

Detailed Description

Preferred embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. While the preferred embodiments of the present invention are shown in the drawings, it should be understood that the present invention may be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art.

In the following description, for the purposes of illustrating various inventive embodiments, certain specific details are set forth in order to provide a thorough understanding of the various inventive embodiments. One skilled in the relevant art will recognize, however, that the embodiments may be practiced without one or more of the specific details. In other instances, well-known devices, structures and techniques associated with this application may not be shown or described in detail to avoid unnecessarily obscuring the description of the embodiments.

Throughout the specification and claims, the word "comprise" and variations thereof, such as "comprises" and "comprising," are to be understood as an open, inclusive meaning, i.e., as being interpreted to mean "including, but not limited to," unless the context requires otherwise.

Reference throughout this specification to "one embodiment" or "some embodiments" means that a particular feature, structure, or characteristic described in connection with the embodiments is included in at least one embodiment. Thus, the appearances of the phrases "in one embodiment" or "in some embodiments" in various places throughout this specification are not necessarily all referring to the same embodiment. Furthermore, the particular features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.

Furthermore, the terms first, second and the like used in the description and the claims are used for distinguishing objects for clarity, and do not limit the size, other order and the like of the described objects.

Fig. 1 shows a schematic diagram of a system 1 for implementing a merchandise recommendation method according to an embodiment of the invention. As shown in fig. 1, the system 1 may include a user terminal 10, a computing device 20, a server 30, and a network 40. User terminal 10, computing device 20, and server 30 may interact with data via network 40. Here, each user terminal 10 may be a mobile or fixed terminal of a user, such as a mobile phone, a tablet computer, a desktop computer, or the like. A user is here an individual user (e.g. an employee of an organization, such as a business) belonging to the organization. The user terminal 10 may communicate with a server 30 of the e-commerce enterprise, for example, through an e-commerce application or a particular search engine installed thereon, to send information to the server 30 and/or receive information from the server 30. The computing device 20 may perform corresponding operations based on data from the user terminal 10 and/or the server 30. The computing device 20 may include at least one processor 210 and at least one memory 220 coupled to the at least one processor 210, the memory 220 having stored therein instructions 230 executable by the at least one processor 210, the instructions 230, when executed by the at least one processor 210, performing at least a portion of the method 100 as described below. Note that herein, computing device 20 may be part of server 30 or may be separate from server 30. The specific structure of computing device 20 or server 30 may be described, for example, in connection with FIG. 8, below.

FIG. 2 illustrates a flow diagram of an item recommendation method 100 according to some embodiments of the invention. The method 100 may be performed, for example, by the computing device 20 or the server 30 in the system 1 shown in fig. 1. The method 100 is described below in conjunction with fig. 1-8, with an example being performed in the computing device 20. The item recommendation method 100 is for ranking a plurality of items to be presented to a user according to the user's organizational schema data. Wherein the plurality of items may be generated based on a default recommendation algorithm of the item recommendation system.

As shown in fig. 2, method 100 includes step 110, where computing device 20 may determine an organization tag for the user based on the user identification information. The organization tag is used to indicate the organization to which the user belongs.

As previously mentioned, a user herein refers to an individual user (e.g., an employee of an organization) belonging to the organization (e.g., the business). Thus, the user identification information may be generated when each individual user registers with the server 30, and includes business identification information (also referred to as a business tag) and personal identification information (also referred to as a personal tag). Alternatively, the user identification information may be personal identification information (e.g., a mobile phone number, an email address, etc.) used when the individual user registers with the server 30, and the server 30 may store in advance user information (e.g., preset by the enterprise for its respective employees) associated with the enterprise, so that the computing device 20 may obtain organization identification information of an organization to which the user belongs from the server 30 according to the personal identification information of the user. The user information may be stored in a personal pattern database as shown in table 1 below, or may be stored in a dedicated user database. The user information may include, for example, the user's name, gender, age, organization to which the user belongs, and the like.

Next, at step 120, computing device 20 may determine whether the user has personal pattern data based on the user identification information of step 110. Here, the personal pattern data of the user is generated based on the historical operation behavior of the user.

The computing device 20 or the server 30 may have a corresponding personal pattern database for recording corresponding operational data for users who have generated historical operational behavior in the system 1. When each user first generates a commodity operation behavior (e.g., browsing, buying, purchasing, etc.) in the system 1, a record may be generated for the user in the personal pattern database to record at least one of the category, brand, operation time, commodity price range, etc. of the operation object (i.e., browsing, buying, purchasing, etc.) of the user. In one embodiment, the record may be updated to include the category, brand, operation time, commodity price range, etc. of the new operation object each time the user generates a new commodity operation behavior in the system 1. Alternatively, in other embodiments, the category, brand, commodity price range, and the like of the operation object of the user may be counted each time the user generates a new commodity operation behavior in the system 1, and only the high-frequency category and brand are saved in the record. Table 1 shows an exemplary representation of personal pattern data according to such an embodiment.

A piece of personal pattern data of user 111 is listed in table 1, including a high frequency purchase category ("3C number" and "daily department item"), a high frequency purchase brand ("hua ye", "san xing" and "white cat"), a high frequency browsing category ("men's clothing" and "sports equipment"), a high frequency browsing brand ("lie in"), and a price range (here, the price range may be a browsing price or a purchase price, or may be a union of a browsing price and a purchase price, hereinafter also referred to as a personal price range). Personal pattern data generated based on browsing behavior and purchasing behavior of the user is exemplarily shown in table 1, and only several (e.g., 1-3, which can be set by the server 30) categories and brands for high-frequency purchase and browsing are exemplarily listed in each item of data (high-frequency purchase category, high-frequency purchase brand, high-frequency browsing category, high-frequency browsing brand).

The present invention will be described below by taking the personal pattern data shown in table 1 as an example. However, those skilled in the art will appreciate that the user's personal pattern data is not limited to the specific items and content listed in Table 1, but may contain more or fewer items or content. For example, table 1 may also include the user high-frequency purchase category or brand, or may also include the latest operation time of the user, or may include only the user high-frequency purchase category or brand and not the high-frequency browsing category or brand, and these variations of the personal pattern data do not affect the intended protection scope of the present invention.

For each user, when the user first generates a commodity operation behavior in the system 1, a piece of user pattern data as shown in table 1 is created for the user. Thus, computing device 20 may determine whether personal pattern data for the user is present in system 1 simply by looking up in the personal pattern database according to the user identification information.

If it is determined that the user does not have personal pattern data ("NO" of decision step 120), then, at step 130, computing device 20 may determine a recommendation value for each of the plurality of items based on the organization pattern data for the associated organization associated with the organization tag.

As described above, when a user first generates a commodity operation behavior in the system 1, a piece of user pattern data as shown in table 1 is created for the user. Thus, the computing device 20 determines that the absence of personal pattern data for the user indicates that the user has not previously generated historical operations in the system 1, that is, that the user needs to be cold started.

Unlike the conventional discordant cold start based on the default recommendation algorithm, in the present invention, the recommendation value for each item can be determined based on the organization pattern data of the user's associated organization without the personal pattern data of the user. Here, in order to distinguish from a recommendation value determined based on the personal pattern data of the user as described below, a recommendation value determined based on the organization pattern data of the associated organization is referred to as a first recommendation value, and a recommendation value determined based on the personal pattern data of the user is referred to as a second recommendation value. Depending on whether the user's personal pattern data is present, the finally determined recommendation value for the good will include only the first recommendation value or a combination of the first and second recommendation values.

Similar to the personal pattern data, the computing device 20 or the server 30 may have a corresponding organization pattern database for recording corresponding operational data for an organization that has generated historical operational behavior in the system 1. The organization pattern data may be generated based on historical operational behavior of the organization (e.g., individual users in the organization). When each organization has a user to generate a commodity operation behavior (such as browse, buy, purchase, etc.) in the system 1 for the first time, a record can be generated for the organization in the organization pattern database to record at least one of the category, brand, operation time, commodity price range, etc. of the operation object (i.e. the browsed, bought, or purchased commodity) of the organization. In one embodiment, the record may be updated to include the category, brand, operation time, commodity price range, etc. of the new operation object each time a user in the organization generates a new commodity operation behavior in the system 1. Alternatively, in other embodiments, each time a user in an organization generates a new commodity operation behavior in the system 1, the category, the brand, the commodity price range, and the like of the operation object of the organization may be counted, and only the high-frequency category and the brand are saved in the record. Table 2 shows an exemplary representation of organization scheme data according to such an embodiment.

A piece of organization pattern data of the organization 222 is listed in table 2, including a recommendation category, a recommendation brand, and a price range (here, the price range may be a browsing price or a purchasing price, or a union of the browsing price and the purchasing price, and is also referred to as an organization price range hereinafter). The organization pattern data generated based on the historical operation behaviors (e.g., historical purchasing behaviors) of the users in the organization is exemplarily shown in table 2, and only several categories (e.g., 1-3, which can be set by the server 30) with the largest number of purchasers and brands are exemplarily listed in each item of data (recommended category, recommended brand).

In one embodiment, the number of people E who buy a certain category of goods and the number of people F who browse the category of goods with high frequency of users of an organization may be obtained based on the personal pattern data as shown in table 1, and the recommended value U for the category may be calculated. For example, the recommended value U for the category may be expressed as U = (E +0.8F)/N, N being the total number of people for the organization. The recommendation value U is calculated for all categories of items purchased and/or viewed by users in the organization, and several categories (e.g., 1-3) with the largest recommendation value U are selected as the recommendation categories in table 2.

Similarly, the number of persons G who buy a certain brand of goods and the number of persons H who browse the brand of goods with high frequency of users of the organization may be acquired based on the personal pattern data as shown in table 1, and the recommended value V of the brand may be calculated. For example, the recommended value V for the brand may be represented as V = (G +0.8H)/N, N being the total number of people for the organization. The recommendation value V is calculated for all brands of merchandise purchased and/or viewed by users in the organization, and a number of brands (e.g., 1-3) with the largest recommendation value V are selected as recommended brands in Table 2.

Alternatively, the organizational pattern data may also include recommended categories and recommended brands for users of different genders. For example, Table 3 shows another exemplary representation of organization pattern data.

In one embodiment, the number of men E1 and women E2 who buy a certain category of commodities at high frequency and the number of men F1 and women F2 who browse the category of commodities at high frequency may be acquired based on the user of the organization for the individual pattern data as shown in table 1, and the male recommendation value U1 and the female recommendation value U2 for the category are calculated. For example, the category-purpose male recommendation value U1 may be represented as U1= (E1 +0.8F1)/N1, N1 is the number of men in the organization, and the category-purpose female recommendation value U2 may be represented as U2= (E2 +0.8F2)/N2, and N2 is the number of women in the organization. The male recommendation value U1 and the female recommendation value U2 are calculated for all categories of merchandise purchased and/or viewed by users in the organization, respectively, and several categories (e.g., 1-3) with the largest male recommendation value U1 are selected as the male recommendation categories in table 3, and several categories (e.g., 1-3) with the largest female recommendation value U2 are selected as the female recommendation categories in table 3.

Similarly, the number of men G1 and women G2 who have a high frequency of purchasing a certain brand of merchandise and the number of men H1 and women H2 who have a high frequency of browsing the brand of merchandise may be acquired based on the individual pattern data as shown in table 1, and the male recommendation value V1 and the female recommendation value V2 of the brand may be calculated. For example, the recommended value for the brand, V1, may be represented by V1= (G1 +0.8H1)/N1, N1 is the number of men in the organization, and the recommended value for the brand, V2, may be represented by V2= (G2 +0.8H2)/N2, and N2 is the number of women in the organization. The male recommendation value V1 and the female recommendation value V2 are calculated for all brands of merchandise purchased and/or browsed by users in the organization, respectively, and several brands (e.g., 1-3) with the largest male recommendation value V1 are selected as the male recommendation brands in table 3, and several brands (e.g., 1-3) with the largest female recommendation value V2 are selected as the female recommendation brands in table 3.

In some embodiments, the associated organization associated with the user's organization tag includes the organization to which the user indicated by the organization tag belongs. In this case, the recommended value of each commodity in the case of cold start may be determined for the user based on the organization pattern data of the organization to which the user belongs. For example, taking tables 1 and 2 as an example, suppose that the organization to which user 111 belongs is 222, and the organization pattern corresponding to the organization is 1, the organization pattern data of the organization includes a recommended category (3C number) and a recommended category coefficient (B1) corresponding to the recommended category, and recommended brands (samsung, hua) and a recommended brand coefficient (D1) corresponding to each recommended brand.

In other embodiments, the associated organization associated with the organization tag of the user includes an organization similar to the organization to which the user belongs. For example, an organization similar in at least one respect (organization nature, industry type, organization scale, territory in which the user is located, etc.) to which the user belongs. In this case, the recommendation value for each item in the case of cold start may be determined for the user based on the organization pattern data of the organization associated with the organization to which the user belongs. When the organization to which the user belongs has not generated the historical operation behavior of the individual user in the system 1 or has generated the historical operation behavior too little, the organization pattern data of the similar associated organization may be used to determine the recommendation value of the commodity. This is equivalent to a situation where the entire organization (e.g., enterprise) is completely cold-started.

FIG. 3 shows a flowchart of step 130 of determining a recommendation value for each of a plurality of items based on organizational schema data of an associated organization associated with an organizational tag, according to one embodiment of the invention.

As shown in FIG. 3, step 130 may include a substep 132 in which computing device 20 may determine an organization category recommendation value R1 for each item based on the recommendation category coefficient B1 and the organization category recommendation basis coefficient A1 for the item in the associated organization.

The organization category recommendation basis coefficient a1 may be determined based on the total number of users under the organization S1 and the number of users under the organization S2 who have generated historical operating behavior (i.e., the number of users who have generated a piece of user pattern data as shown in table 1). For example, the organization category recommendation base coefficient a1 may be determined based on the following equation (1):

A1=0.5 *S2/S1 （1）。

the recommendation category coefficient B1 for the item in the associated organization may be based on the ranking of the recommendation categories in the organization pattern data. For example, in the case where a plurality of recommendation categories are included in the organization pattern data, the plurality of recommendation categories may be sorted according to the number of purchases by the users within the organization, so that each recommendation category may be given a different recommendation category coefficient B1, so that the commodities belonging to the corresponding recommendation category have a corresponding recommendation category coefficient B1. For example, in the case where a plurality of recommendation categories are included, the recommendation category coefficient B1 of the commodities in the plurality of recommendation categories may be set to be sequentially decreased, for example, to be 0.8, 0.6, 0.4 … …, respectively, while the recommendation category coefficient B1 of the commodities other than the recommendation categories may be set to be 0. In the case where only one recommendation category is included, the recommendation category coefficient B1 for the items in the recommendation category may be set to 0.5, while the recommendation category coefficient B1 for the items not belonging to the recommendation category may be set to 0.

Further, in some embodiments, as shown in table 3, male recommendation categories and female recommendation categories distinguished by user gender are also included in the organizational schema data. In this case, the recommended category corresponding to the user gender and the recommended category coefficient B1 thereof may also be acquired from the organization pattern data according to the user gender. That is, even if the commodities belong to the same category, there may be different recommendation category coefficients B1 for male users and female users.

In one embodiment, the organization category recommendation value R1 for the good may be determined as R1= a1 × B1.

In sub-step 134, the computing device 20 may determine an organization brand recommendation value Q1 for the good based on the recommended brand coefficient D1 and the organization brand recommendation base coefficient C1 for the good in the associated organization.

Similar to the organization category recommendation basis coefficient a1, the organization brand recommendation basis coefficient C1 may also be determined based on the total number of users S1 under the organization and the number of users S2 under the organization who have generated historical operating behavior (i.e., the number of users who have generated a piece of user pattern data shown in table 1). For example, the organization brand recommendation base coefficient C1 may be determined based on the following equation (2):

C1=0.5 *S2/S1 （2）。

similar to the recommendation category coefficient B1, the recommended brand coefficient D1 for the good in the associated organization may be based on an ordering of recommended brands in the organization pattern data. For example, in the case where a plurality of recommended brands are included in the organization pattern data, the plurality of recommended brands may be sorted according to the number of purchases by the users within the organization, so that each recommended brand may be given a different recommended brand coefficient D1, and thus, the items belonging to the corresponding recommended brand have a corresponding recommended brand coefficient D1. For example, in a case where a plurality of recommended brands are included, the recommended brand coefficients D1 of the commodities in the plurality of recommended brands may be set to be sequentially decremented, for example, to be 0.8, 0.6, and 0.4 … …, respectively, while the recommended brand coefficient D1 of the commodities outside the recommended brand may be set to be 0. In the case where only one recommended brand is included, the recommended brand coefficient D1 of the items in the recommended brand may be set to 0.5, while the recommended brand coefficient D1 of the items not belonging to the recommended brand may be set to 0.

Further, in some embodiments, as shown in table 3, male and female recommended brands distinguished by user gender are also included in the organizational pattern data. In this case, the recommended brand corresponding to the user's gender and its recommended brand coefficient D1 may also be acquired from the organization pattern data according to the user's gender. That is, even if the commodities belong to the same brand, there may be different recommended brand coefficients D1 for male users and female users.

In one embodiment, the organization brand recommendation Q1 for the good may be determined as Q1= C1 by D1.

In sub-step 136, computing device 20 may determine an organization category recommendation value R1 for the item, a category weight w corresponding to the item category based on sub-step 132_rThe organizational brand recommendation Q1 for the good and the brand weight w corresponding to the brand of the good determined in substep 134_qA first recommended value W1 for the good is determined.

For example, the first recommended value W1 may be represented as W1= W_r *R1+ w_q *Q1。

In some embodiments, the organization pattern data may also include whitelist categories and whitelist category coefficients B3 and whitelist brands and whitelist brand coefficients D3. In this case, in sub-step 132, the computing device 20 may determine an organization category recommendation value R1 for the good based on the recommendation category coefficient B1 for the good in the associated organization, the whitelist category coefficient B3 in the associated organization, and the organization category recommendation basis coefficient a 1.

The white list category may be derived based on historical data of the entire system 1 or may be derived from historical data of a given organization. For example, categories for which the user rating is above a given threshold (e.g., 4.5 out of 5 points) and the return rate is below a given threshold (e.g., 1%) over a longer period of time may be determined as white list categories. The organization pattern data for an organization may include one or more whitelist categories to which the whitelist category coefficients B3 for items belonging to the whitelist categories may be set to the same value, for example, to 0.8. Preferably, the white list category coefficient B3 for the same category may be set to be greater than the recommended category coefficient B1.

In this case, the tissue category recommendation value R1 may be determined as R1= a1 × B1+ a1 × B3.

Similarly, in sub-step 134, the computing device 20 may determine an organization brand recommendation value, Q1, for the good based on the associated organization's recommended brand coefficient, D1, the associated organization's whitelist brand coefficient, D3, and the organization brand recommendation base coefficient, C1.

The white list brands may be derived based on historical data of the entire system 1 or may be derived from historical data of a given organization. For example, brands that have a user rating above a given threshold (e.g., 4.5 out of 5 points) and a return rate below a given threshold (e.g., 1%) over a longer period of time may be determined to be whitelisted brands. The organization pattern data for an organization may include one or more whitelisted brands whose whitelisted brand coefficients D3 may be set to the same value, for example, to 0.8. Preferably, the white list brand coefficient D3 for the same brand may be set to be greater than or equal to the recommended brand coefficient D1.

In this case, the tissue brand recommendation Q1 may be determined as Q1= C1 × D1+ C1 × D3.

In step 120, if it is determined that there is no personal pattern data of the user, a first recommendation value W1 for each commodity may be determined as a recommendation value W for the commodity based on organization pattern data of an associated organization, i.e., W = W1, in step 130.

On the other hand, if it is determined in step 120 that the personal pattern data for the user exists ("yes" determination in step 120), then, in step 140, computing device 20 may determine a recommended value W for the good based on the organization pattern data for the associated organization associated with the organization tag and the personal pattern data for the user.

FIG. 4 shows a flowchart of step 140 of determining a recommendation value for an item based on organization pattern data of an associated organization associated with an organization tag and personal pattern data of a user, according to one embodiment of the invention.

As shown in fig. 4, step 140 may include a substep 142 in which computing device 130 may determine a first recommendation value W1 for each of a plurality of items based on organization pattern data for an associated organization associated with the organization tag. Here, the method for determining the first recommended value W1 of the commodity in the sub-step 142 is the same as the method for determining the recommended value of the commodity in the step 130 described above with reference to fig. 3, and therefore, the description thereof is omitted.

In sub-step 144, computing device 20 may determine a second recommendation value W2 for the item based on the user's personal pattern data. Here, the personal pattern data of the user includes, for example, as shown in table 1, a high-frequency purchase category and/or a high-frequency browsing category (hereinafter collectively referred to as a high-frequency category) and a high-frequency purchase brand and/or a high-frequency browsing brand (hereinafter collectively referred to as a high-frequency brand). The method of determining the second recommendation value W2 for the good based on the user's personal pattern data is similar to the method of determining the first recommendation value W1 based on the organization pattern data of the associated organization as described in connection with fig. 3, as described below.

FIG. 5 shows a flowchart of sub-step 144 of determining a second recommendation value for an item based on the user's personal pattern data, in accordance with one embodiment of the present invention.

As shown in fig. 5, in sub-step 1442 of sub-step 144, computing device 20 may determine a personal category recommendation value R2 for the item based on the user's high frequency category coefficient B2 and the personal category recommendation base coefficient a 2.

The personal category recommendation basis coefficient a2 may be set to a reference value, for example, to 1. Thus, the organization category recommendation basis coefficient a1=0.5 × S2/S1 is always less than or equal to half of the individual category recommendation basis coefficient a2, so that the influence of the organization category is less than (typically less than half) the influence of the individual category in the final product recommendation. The method is more suitable for the actual situation that personal historical data can reflect user preferences.

The user's high frequency category coefficient B2 may be derived based on the ordering of the high frequency categories in the personal pattern data. For example, in the case where a plurality of high frequency categories (high frequency purchase categories and/or high frequency browsing categories) are included in the personal pattern data, the plurality of high frequency categories may be sorted according to the number of times of purchase or browsing by the user, so that a different high frequency category coefficient B2 may be assigned to each high frequency category, so that the goods belonging to the corresponding high frequency category have a corresponding high frequency category coefficient B2. For example, in the case where at least one high frequency category is included, the high frequency category coefficients B2 of the commodities belonging to these high frequency categories may be set to be sequentially decreased, for example, to be 1.0, 0.9, 0.8 … …, respectively, while the high frequency category coefficients B2 of the commodities other than the high frequency categories are set to be 0. Further, in the case where both the high-frequency purchase category and the high-frequency browsing category are included in the personal pattern data, a high-frequency category coefficient B2 may be set separately for both, and the coefficient B2 for the high-frequency purchase category may be set larger than the coefficient B2 for the high-frequency browsing category.

In one embodiment, the personal recommendation value R2 for the item of merchandise may be determined as R2= a2 × B2.

In sub-step 1444, computing device 20 may determine a personal brand recommendation value, Q2, for the good based on the user's high frequency brand coefficient, D2, and personal brand recommendation base coefficient, C2.

Similar to the organization brand recommendation basis coefficient C1, the individual brand recommendation basis coefficient C2 may be set to a reference value, for example, to 1. Thus, the organization brand recommendation base coefficient C1=0.5 × S2/S1 is always less than or equal to half of the individual brand recommendation base coefficient C2, so that the influence of the organization brand is less than the influence of the individual brand (typically less than half) in the final product recommendation. The method is more suitable for the actual situation that personal historical data can reflect user preferences.

Similar to the recommended brand coefficient D1, the user's high frequency brand coefficient D2 may be derived based on the ranking of the high frequency brands in the personal pattern data. For example, in the case where a plurality of high-frequency brands (high-frequency purchase brands and/or high-frequency browsing brands) are included in the personal pattern data, the plurality of high-frequency brands may be sorted according to the number of times of purchase or browsing by the user, so that each high-frequency brand may be given a different high-frequency brand coefficient D2. For example, in the case of including at least one high-frequency brand, the high-frequency brand coefficients D2 of the high-frequency brands may be set to be sequentially decreased, for example, to be 1.0, 0.9, and 0.8 … …, respectively, while the high-frequency brand coefficient D2 of the goods outside the high-frequency brand may be set to be 0. Further, in the case where both the high-frequency purchase brand and the high-frequency browsing brand are included in the personal pattern data, the high-frequency brand coefficient D2 may be set separately for both, and the coefficient D2 of the high-frequency purchase brand may be set larger than the coefficient D2 of the high-frequency browsing category.

In one embodiment, the personal brand recommendation Q2 for the good may be determined as Q2= C2 × D2.

In sub-step 1446, computing device 20 may determine a personal category recommendation value R2, a category weight (w) corresponding to the category of merchandise, based on sub-step 1442_r) Personal brand recommendation Q2 determined in sub-step 1444 and brand weight (w) corresponding to the brand of merchandise_q) A second recommended value W2 for the good is determined.

For example, the second recommended value W2 may be represented as W2= W_r *R2+ w_q *Q2。

Continuing with FIG. 4, at substep 146, computing device 20 may determine a recommended value W for the item based on the first recommended value W1 determined at substep 142 and the second recommended value W2 determined at substep 144.

For example, the recommended value W for the good may be expressed as:

W=W1+W2=( w_r *R1+ w_q *Q1) + (w_r *R2+ w_q *Q2)

= w_r *(R1+R2)+ w_q *(Q1+Q2)。

thus, the organization category based recommendation value R1, the personal category recommendation value R2, and the category weight w may also be used_rThe obtained recommendation value is called the category recommendation value R of the commodity, namely

R= w_r *(R1+R2)。

Similarly, organization-based brand recommendations Q1, personal brand recommendations Q2, and brand weights w_qThe obtained recommendation value is called brand recommendation value Q of the commodity, namely

Q= w_q *(Q1+Q2)。

Thus, the resultant recommended value W for each commodity may be represented as W = R + Q.

In some embodiments, the impact of price factors and other factors on the recommended value of the good may also be considered. Specifically, as shown in table 1, a personal price range of each user may also be included in the personal pattern data of the user. In one embodiment, the personal price range may be a price range for goods that the user has viewed and/or purchased. For example, in Table 1, user 111 has a personal price range of 159 to 3266 dollars. Similarly, as shown in tables 2 and 3, an organization price range for each organization may also be included in the organization pattern data for that organization. The organization price range may be determined based on the price ranges of items purchased by all users (employees) in the organization who have generated historical purchasing behavior. In one embodiment, the lowest value of the items purchased by the employee may be directly used as the lower limit of the organization price range, and the highest value of the items purchased by the employee may be used as the upper limit of the organization price range. In another embodiment, the organizational price range may be chosen according to a normal distribution over the price ranges of the items purchased by all employees who have generated historical purchasing behavior.

In this case, the method 100 may further include the step of adjusting the recommended value for each item according to the organization price range of the associated organization and the personal price range of the user.

As shown in FIG. 6, at step 152, computing device 20 may determine a recommended price range for the user based on the organization price range of the associated organization and the personal price range of the user.

In one embodiment, the organization price ranges of the associated organization as shown in Table 2 or Table 3 and the personal price ranges of the user as shown in Table 1 may be taken together to determine the recommended price range for the user. For example, in the example shown in tables 1 and 2, the recommended price range for user 111 may be determined to be 100 to 3800 dollars.

At step 154, computing device 20 may determine whether the price of an item is within the recommended price range for the user.

If it is determined that the price of the item is within the recommended price range for the user ("yes" determination of step 154), then, at step 156, computing device 20 may base the first price weight w_p1The recommended value W for the commodity is updated.

The updated recommended value W 'of the commodity may be represented as W' = W_p1*W。

On the other hand, if the price of the item is not within the recommended price range for the user ("NO" of the determination of step 154), then, at step 158, computing device 20 may weight w based on the second price_p2The recommended value W for the commodity is updated.

The updated recommended value W 'of the commodity may be represented as W' = W_p2*W。

Here, the second lattice weight w_p2Should be lower than the first price weight w_p1。

Further, as previously described, the plurality of items may be generated by a default recommendation algorithm of the system 1, in which different default recommendation weights may be assigned to different items. For example, in many e-commerce recommendation systems, items in the 3C digital category are given a higher default recommendation weight so that by default items in the 3C digital category are placed more often in front.

In this case, in some embodiments, the computing device 20 may also determine a default recommendation weight for each item and update the recommendation value W for the item based on the default recommendation weight for the item.

After determining or updating the recommended values for the plurality of items as described above, the computing device 20 may rank the items according to the recommended values to rank the plurality of items based on the recommended values for the plurality of items to present the items to the user. Thus, even in the case of cold start, the ranking of the items viewed by the user is determined according to the organizational scheme of the organization to which the user belongs, rather than the uniform default ranking, which shows that it can be more suitable for the user's personal needs.

As previously described, the associated organization may include the organization to which the user indicated by the organization tag belongs, and may also include an organization that is similar in at least one respect to the organization to which the user belongs. The former may be applicable to the case of cold start only of the current user, and the latter may be applicable to the case of complete cold start of the entire organization to which the current user belongs.

For the latter, the method 100 should also include the step of determining the associated organization. FIG. 7 shows a flowchart of steps for determining an associated organization associated with a user's organization tag, in accordance with one embodiment of the invention.

As shown in fig. 7, at step 162, computing device 20 may obtain tissue information for a plurality of candidate tissues from a tissue database. The organization information may include at least one of organization properties, industry type, organization size, and locality of each candidate organization.

Computing device 20 or server 30 may have a corresponding organization database for storing organization information for its various organizational users (e.g., businesses). Table 4 shows an example of organization information in the organization database according to an embodiment of the present invention.

As shown in table 4, the organization information may include the nature of the organization (e.g., nationally owned and nationally owned stock enterprises, civil enterprises, scientific institutions, higher institutions, etc.), the type of industry (e.g., manufacturing, planting, mining, etc.), the size of the organization (e.g., large, medium, small, micro, etc.), and the locale of the organization.

At step 164, computing device 20 compares the tissue information of the plurality of candidate tissues with the tissue information of the tissue to which the user belongs, respectively, to determine a similarity coefficient Y between each of the plurality of candidate tissues and the tissue to which the user belongs.

More specifically, for each candidate tissue, a first similarity value Y1 may be set for the candidate tissue based on whether the tissue property of the candidate tissue is the same as the tissue property of the tissue to which the user belongs. For example, if the tissue property of the candidate tissue is the same as the tissue property of the tissue to which the user belongs, Y1=1, and conversely, Y1= 0.

Further, a second similarity value Y2 may be set for the candidate organization based on whether the industry type of the candidate organization is the same as the industry type of the organization to which the user belongs. For example, if the industry type of the candidate organization is the same as the industry type of the organization to which the user belongs, Y2=1, and conversely, Y2= 0.

Further, a third similarity value Y3 may be set for the candidate tissue based on whether the tissue scale of the candidate tissue is the same as the tissue scale of the tissue to which the user belongs. For example, if the tissue size of the candidate tissue is the same as the tissue size of the tissue to which the user belongs, Y3=1, and conversely, Y3= 0.

Further, the fourth similarity value Y4 may be set for the candidate organization based on whether the region in which the candidate organization is located is the same as the region in which the organization to which the user belongs. For example, if the region in which the candidate tissue is located is the same as the region in which the tissue to which the user belongs, Y4=1, and conversely, Y4= 0.

Finally, a similarity coefficient Y between the candidate tissue and the tissue to which the user belongs may be determined based on the first similarity value Y1, the second similarity value Y2, the third similarity value Y3, and the fourth similarity value Y4.

For example, in one embodiment, Y = Y1+ Y2+ Y3+ Y4.

In another embodiment, various organization information may be given different weights. For example, in general, the influence of the size of the tissue and the region where the tissue is located on whether the tissue is similar to each other is larger, and therefore, the weights of the first similarity value Y1 and the second similarity value Y2 may be set to be smaller than the third similarity value Y3 and the fourth similarity value Y4.

Alternatively, the above-described Y1, Y2, Y3 and Y4 may be set to different values directly upon judging that the corresponding tissue information is the same.

Continuing with FIG. 7, at step 166, computing device 20 may select, from the plurality of candidate tissues obtained at step 162, the one candidate tissue having the highest similarity coefficient as the associated tissue similar to the tissue to which the user belongs.

For example, as shown in Table 4, assuming that organization 444 is not currently having organization pattern data (e.g., any user that has not yet had organization 444 has generated historical operational behavior in system 1), then organizations 222 and 333 may be selected from the organization database as candidate organizations, and a similarity coefficient between organization 444 and organizations 222 and 333 is calculated, respectively. As shown in table 4, the similarity coefficient Y =3 between the tissue 444 and the tissue 222, and the similarity coefficient Y =1 between the tissue 444 and the tissue 333, and therefore, the tissue 333 may be selected as the associated tissue of the tissue 444, and the tissue mode 2 of the tissue 333 may be selected as the tissue mode when the user in the tissue 444 first accesses the system 1.

FIG. 8 illustrates a block diagram of a computing device 800 suitable for implementing embodiments of the present invention. Computing device 800 may be, for example, computing device 20 or server 30 in system 1 as described above.

As shown in fig. 8, computing device 800 may include one or more Central Processing Units (CPUs) 810 (only one of which is schematically shown) that may perform various suitable actions and processes in accordance with computer program instructions stored in a Read Only Memory (ROM) 820 or loaded from a storage unit 880 into a Random Access Memory (RAM) 830. In the RAM 830, various programs and data required for the operation of the computing device 800 may also be stored. The CPU 810, ROM 820, and RAM 830 are connected to each other by a bus 840. An input/output (I/O) interface 850 is also connected to bus 840.

A number of components in computing device 800 are connected to I/O interface 850, including: an input unit 860 such as a keyboard, a mouse, and the like; an output unit 870 such as various types of displays, speakers, and the like; a storage unit 880 such as a magnetic disk, optical disk, or the like; and a communication unit 890 such as a network card, modem, wireless communication transceiver, or the like. The communication unit 890 allows the computing device 800 to exchange information/data with other devices via a computer network, such as the internet, and/or various telecommunications networks.

The method 100 described above may be performed, for example, by the CPU 810 of one or more computing devices 800. For example, in some embodiments, method 100 may be implemented as a computer software program tangibly embodied in a machine-readable medium, such as storage unit 880. In some embodiments, some or all of the computer program can be loaded onto and/or installed onto computing device 800 via ROM 820 and/or communications unit 890. When loaded into RAM 830 and executed by CPU 810, the computer program may perform one or more of the operations of method 100 described above. Further, the communication unit 890 may support wired or wireless communication functions.

Those skilled in the art will appreciate that the computing device 800 illustrated in FIG. 8 is merely illustrative. In some embodiments, computing device 800 may contain more or fewer components than shown in FIG. 8.

By using the scheme of the invention, the commodity recommendation is carried out for the user by using the organization mode data of the organization related to the user, the commodity recommendation accuracy can be improved, and especially under the condition of cold start of the user, the recommended commodity can better meet the user requirements.

The merchandise recommendation method 100 and the computing device 800 that may be used to implement the method 100 according to the present invention are described above in connection with the drawings. However, it will be appreciated by those skilled in the art that the performance of the steps of the method 100 is not limited to the order shown in the figures and described above, but may be performed in any other reasonable order. Further, the computing device 800 also need not include all of the components shown in FIG. 8, it may include only some of the components necessary to perform the functions described in the present disclosure, and the manner in which these components are connected is not limited to the form shown in the figures.

The present invention may be methods, apparatus, systems and/or computer program products. The computer program product may include a computer-readable storage medium having computer-readable program instructions embodied therein for carrying out aspects of the present invention.

In one or more exemplary designs, the functions described herein may be implemented in hardware, software, firmware, or any combination thereof. For example, if implemented in software, the functions may be stored on or transmitted over as one or more instructions or code on a computer-readable medium.

The units of the apparatus disclosed herein may be implemented using discrete hardware components, or may be integrally implemented on a single hardware component, such as a processor. For example, the various illustrative logical blocks, modules, and circuits described in connection with the invention may be implemented or performed with a general purpose processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other programmable logic device, discrete gate or transistor logic, discrete hardware components, or any combination thereof designed to perform the functions described herein.

Those of skill would further appreciate that the various illustrative logical blocks, modules, circuits, and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, computer software, or combinations of both.

The previous description of the invention is provided to enable any person skilled in the art to make or use the invention. Various modifications to the disclosure will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other variations without departing from the spirit or scope of the disclosure. Thus, the present invention is not intended to be limited to the examples and designs described herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method of merchandise recommendation, comprising:

determining an organization label of a user based on user identification information, the organization label being used for indicating an organization to which the user belongs;

determining whether the user has personal pattern data based on the user identification information, the personal pattern data being generated based on historical operational behavior of the user;

if the user is determined not to have the personal pattern data, determining a first recommendation value of each of a plurality of commodities as a recommendation value of the commodity based on organization pattern data of an associated organization associated with the organization tag; and

if it is determined that the user has personal pattern data, a first recommendation value for each of a plurality of items is determined based on organization pattern data for an associated organization associated with the organization tag, a second recommendation value for the item is determined based on the personal pattern data, and a recommendation value for the item is determined based on the first and second recommendation values.

2. The method of claim 1, wherein the organization pattern data comprises a recommended category coefficient for the associated organization and a recommended brand coefficient for the associated organization, wherein determining a first recommendation value for each of a plurality of items based on the organization pattern data for the associated organization associated with the organization tag comprises:

determining an organization category recommendation value of the commodity based on the recommendation category coefficient and the organization category recommendation basis coefficient of the commodity in the associated organization;

determining an organization brand recommendation value of the commodity based on the recommended brand coefficient and the organization brand recommendation basis coefficient of the commodity in the associated organization; and

determining a first recommendation value for the good based on the organization category recommendation value, the category weight, the organization brand recommendation value, and the brand weight for the good.

3. The method of claim 2, wherein the organization pattern data further comprises a whitelist category coefficient for the associated organization and a whitelist brand coefficient for the associated organization, wherein determining a first recommendation value for each of a plurality of items based on the organization pattern data for the associated organization associated with the organization tag comprises:

determining an organization category recommendation value for the commodity based on the recommendation category coefficient for the commodity in the associated organization, the white list category coefficient in the associated organization, and the organization category recommendation basis coefficient; and

determining an organization brand recommendation value for the good based on the recommended brand coefficient for the associated organization, the whitelist brand coefficient for the associated organization, and the organization brand recommendation value for the good.

4. The method of any of claims 1-3, wherein the personal pattern data includes a high frequency category coefficient for the user, a high frequency brand coefficient for the user, wherein determining a second recommended value for the good based on the personal pattern data includes:

determining a personal category recommendation value of the commodity based on the high-frequency category coefficient and the personal category recommendation basis coefficient of the user;

determining a personal brand recommendation value for the good based on the high frequency brand coefficient and the personal brand recommendation base coefficient of the user; and

determining a second recommendation value for the good based on the personal category recommendation value, the category weight, the personal brand recommendation value, and the brand weight for the good.

5. The method of claim 1, wherein the organization pattern data further comprises an organization price range for the associated organization, the personal pattern data further comprises a personal price range for the user, and the method further comprises:

determining a recommended price range for the user based on the organization price range for the associated organization and the personal price range for the user;

determining whether the price of the commodity is in a recommended price range of the user;

if the price of the commodity is in the recommended price range of the user, updating the recommended value of the commodity based on the first price weight; and

if the price of the commodity is not in the recommended price range of the user, updating the recommended value of the commodity based on a second price weight, wherein the second price weight is lower than the first price weight.

6. The method of claim 5, wherein determining the recommended price range for the user based on the organizational price range for the associated organization and the personal price range for the user comprises:

and taking a union of the organization price range of the associated organization and the personal price range of the user to determine the recommended price range of the user.

7. The method of claim 5, further comprising:

determining a default recommendation weight of the commodity, wherein the default recommendation weight is based on a default recommendation algorithm of a commodity recommendation system; and

and updating the recommended value of the commodity based on the default recommended weight of the commodity.

8. The method of claim 1, further comprising:

sorting the plurality of commodities based on the recommended values of the plurality of commodities to display the plurality of commodities to the user.

9. The method of claim 1, wherein the associated organization comprises an organization similar to an organization to which the user belongs, and the method further comprises:

obtaining organization information of a plurality of candidate organizations from an organization database, wherein the organization information comprises at least one of organization properties, industry types, organization sizes and places of the candidate organizations;

comparing the tissue information of the plurality of candidate tissues with the tissue information of the tissue to which the user belongs respectively to determine a similarity coefficient between each of the plurality of candidate tissues and the tissue to which the user belongs; and

selecting one candidate tissue with the highest similarity coefficient from the plurality of candidate tissues as the associated tissue.

10. The method of claim 9, wherein determining a similarity coefficient for each of the plurality of candidate tissues to the tissue to which the user belongs comprises:

setting a first similarity value for the candidate tissue based on whether the tissue property of the candidate tissue is the same as the tissue property of the tissue to which the user belongs;

setting a second similarity value for the candidate organization based on whether the industry type of the candidate organization is the same as the industry type of the organization to which the user belongs;

setting a third similarity value for the candidate tissue based on whether the tissue scale of the candidate tissue is the same as the tissue scale of the tissue to which the user belongs;

setting a fourth similarity value for the candidate organization based on whether the region where the candidate organization is located is the same as the region where the organization the user belongs to; and

determining a similarity coefficient between the candidate tissue and a tissue to which the user belongs based on the first, second, third, and fourth similarity values.

11. A computing device, comprising:

at least one processor; and

at least one memory coupled to the at least one processor and storing instructions for execution by the at least one processor, the instructions when executed by the at least one processor causing the computing device to perform the steps of the method of any of claims 1-10.

12. A computer readable storage medium having stored thereon computer program code which, when executed, performs the method of any of claims 1 to 10.