Evaluation and recommendation method and device for posting orders

Abstract

The present disclosure provides a method and device for recommending reviews and posting orders. The review and posting order recommendation device collects historical behavior data of users; constructs a long-term preference category pool, a hot-selling category pool, and an interactive category pool associated with the user based on the historical behavior data of the user; determines the recommendation weight of each category based on the appearance of each category in the long-term preference category pool, the hot-selling category pool, and the interactive category pool; selects a predetermined number of user preference categories in order of recommendation weight from large to small; scores the reviews and posting orders associated with the user's preference categories according to predetermined rules; and recommends the reviews and posting orders to the user in order of the scoring results from high to low. The present disclosure can recommend products that users are interested in to users, thereby effectively improving user experience, and effectively improving user stickiness and product conversion rate.

Description

Technical Field

[0001] The present disclosure relates to the field of information processing, and in particular to a method and device for evaluating and recommending orders. Background Art

[0002] In the related art, when users shop on e-commerce platforms, they will display and comment on the purchased products so that other users can provide corresponding references when purchasing the same or similar products on the e-commerce platform. The e-commerce platform will also recommend other users' display and comment content to a specific user, that is, provide a feed flow to the user. Summary of the invention

[0003] The inventors have discovered through research that in related technologies, when an e-commerce platform provides a feed stream to a user, it does not consider whether the content provided meets the needs of the user, which may cause the user to be uninterested in the recommended display and comment content.

[0004] Based on this, the present disclosure provides a review and order recommendation solution that can effectively improve user stickiness and product conversion rate.

[0005] According to a first aspect of an embodiment of the present disclosure, a method for recommending reviews and posting orders is provided, comprising: collecting historical behavior data of a user; constructing a long-term preference category pool, a hot-selling category pool, and an interactive category pool associated with the user based on the historical behavior data of the user; determining a recommendation weight of each category based on the appearance of each category in the long-term preference category pool, the hot-selling category pool, and the interactive category pool; selecting a predetermined number of the user's preference categories in order of the recommendation weights from large to small; scoring the reviews and posting orders associated with the user's preference categories according to predetermined rules; and recommending the reviews and posting orders to the user in order of the scoring results from high to low.

[0006] In some embodiments, constructing a long-term preference category pool associated with the user based on the user's historical behavior data includes: determining the user's first preference value for a preset category based on the user's category click volume within a first preset period; determining the user's second preference value for the preset category based on the user's purchase data within the first preset period; using the sum of the first preference value and the second preference value as the user's preference score for the preset category; and constructing the long-term preference category pool based on the preset categories and the corresponding preference scores.

[0007] In some embodiments, building a hot-selling category pool associated with the user based on the user's historical behavior data includes: building the user's hot-selling category pool based on at least one of the user's order volume, browsing data, attention data, or added purchase data within a second preset period, wherein the second preset period is greater than the first preset period.

[0008] In some embodiments, constructing an interactive category pool associated with the user based on the user's historical behavior data includes: calculating the user's interaction score for a preset category based on at least one of the user's click preview behavior, like behavior, reply behavior, or share behavior within a third preset period, wherein the third preset period is less than the first preset period; and constructing the user's interactive category pool based on the user's interaction score for the preset category.

[0009] In some embodiments, the interaction score Y is: Y = click preview behavior score × N1 + like behavior score × N2 + reply behavior score × N3 + sharing behavior score × N4; wherein, the total number of click preview behaviors, like behaviors, reply behaviors and sharing behaviors of the user within the third preset period is N, N1 is the percentage of the click preview behavior in the N, N2 is the percentage of the like behavior in the N, N3 is the percentage of the reply behavior in the N, and N4 is the percentage of the sharing behavior in the N.

[0010] In some embodiments, the recommendation weight W of a designated category is: W=U1X1+U2X2+U3X3; wherein, U1 is the appearance of the designated category in the long-term preference category pool. If the designated category appears in the long-term preference category pool, U1 takes 1, otherwise U1 takes 0, and X1 is the weight of the long-term preference category pool; U2 is the appearance of the designated category in the hot-selling category pool. If the designated category appears in the hot-selling category pool, U2 takes 1, otherwise U2 takes 0, and X2 is the weight of the hot-selling category pool; U3 is the appearance of the designated category in the interactive category pool. If the designated category appears in the interactive category pool, U2 takes 1, otherwise U2 takes 0, and X3 is the weight of the interactive category pool.

[0011] In some embodiments, X3>X1>X2.

[0012] According to a second aspect of an embodiment of the present disclosure, there is provided an evaluation and order recommendation device, comprising: a collection module, configured to collect historical behavior data of a user; a category pool construction module, configured to construct a long-term preference category pool, a hot-selling category pool and an interactive category pool associated with the user according to the historical behavior data of the user; a recommendation weight determination module, configured to determine the recommendation weight of each category according to the appearance of each category in the long-term preference category pool, the hot-selling category pool and the interactive category pool; a preference category selection module, configured to select a predetermined number of the user's preference categories in order of the recommendation weights from large to small; a scoring module, configured to score the evaluation and order associated with the user's preference categories according to predetermined rules; and a recommendation module, configured to recommend the evaluation and order to the user in order of the scoring results from high to low.

[0013] According to a third aspect of an embodiment of the present disclosure, there is provided a device for evaluating and recommending orders, including: a memory configured to store instructions; a processor coupled to the memory, the processor being configured to execute a method as described in any of the above embodiments based on the instructions stored in the memory.

[0014] According to a fourth aspect of an embodiment of the present disclosure, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions, and when the instructions are executed by a processor, the method involved in any of the above embodiments is implemented.

[0015] Other features and advantages of the present disclosure will become apparent from the following detailed description of exemplary embodiments of the present disclosure with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the embodiments of the present disclosure or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0017] Figure 1 A schematic diagram of a process of a method for recommending reviews and posting orders according to an embodiment of the present disclosure;

[0018] Figure 2 This is a structural schematic diagram of a device for reviewing and recommending orders according to an embodiment of the present disclosure;

[0019] Figure 3 This is a schematic diagram of the structure of a device for evaluating and recommending orders according to another embodiment of the present invention. DETAILED DESCRIPTION

[0020] The following will be combined with the drawings in the embodiments of the present disclosure to clearly and completely describe the technical solutions in the embodiments of the present disclosure. Obviously, the described embodiments are only part of the embodiments of the present disclosure, rather than all of the embodiments. The following description of at least one exemplary embodiment is actually only illustrative and is by no means intended to limit the present disclosure and its application or use. Based on the embodiments in the present disclosure, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present disclosure.

[0021] The relative arrangement of components and steps, the numerical expressions and numerical values ​​set forth in these embodiments do not limit the scope of the present disclosure unless specifically stated otherwise.

[0022] At the same time, it should be understood that for the convenience of description, the sizes of the various parts shown in the drawings are not drawn according to the actual proportional relationship.

[0023] Technologies, methods, and apparatus known to ordinary technicians in the relevant field may not be discussed in detail, but where appropriate, such technologies, methods, and apparatus should be considered part of the authorization specification.

[0024] In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not as limiting. Therefore, other examples of the exemplary embodiments may have different values.

[0025] It should be noted that like reference numerals and letters refer to similar items in the following figures, and therefore, once an item is defined in one figure, it need not be further discussed in subsequent figures.

[0026] Figure 1 The following is a flow chart of a method for evaluating and recommending orders according to an embodiment of the present disclosure. In some embodiments, the following steps of the method for evaluating and recommending orders are performed by a device for evaluating and recommending orders.

[0027] In step 101, historical behavior data of users is collected.

[0028] In step 102, a long-term preference category pool, a hot-selling category pool, and an interactive category pool associated with the user are constructed based on the user's historical behavior data.

[0029] In some embodiments, a first preference value of a user for a preset category is determined based on the number of category clicks by the user in a first preset period. A second preference value of a user for a preset category is determined based on the additional purchase data of the user in the first preset period. The sum of the first preference value and the second preference value is used as the user's preference score for the preset category, and a long-term preference category pool is constructed based on the preset categories and the corresponding preference scores.

[0030] In some embodiments, the first preset period is a period of approximately three months.

[0031] In some embodiments, the user's preference score for a preset category is calculated by using a TF (Term Frequency)-IDF (Inverse Document Frequency) algorithm.

[0032] For example, the first preference value of the category clicked by the user in the past three months is shown in formula (1) to formula (3):

[0033] tf=n / N (1)

[0034] Here, n is the number of clicks by the user on the specified category, and N is the total number of clicks by the user.

[0035] idf=log(d / D+1) (2)

[0036] Among them, d is the number of users who click on each category, and D is the total number of users.

[0037] TF-IDF= tf *idf (3)

[0038] That is, formula (1) to formula (3) can be used to calculate the first preference value of the category clicked by the user in the past three months. In addition, formula (1) to formula (3) can also be used to calculate the second preference value of the user in the past three months. Accordingly, in the process of calculating the second preference value, it is necessary to replace the data related to clicks with data related to add to cart.

[0039] Since the TF-IDF algorithm itself is not the inventive point of the present disclosure, it will not be described in detail here.

[0040] In some embodiments, a user's hot-selling category pool is constructed based on at least one of the user's order volume, browsing data, attention data, or added purchase data within a second preset period, where the second preset period is greater than the first preset period.

[0041] For example, the second preset period is a period of nearly half a year.

[0042] In some embodiments, the user's order volume, browsing data, attention data and added-to-purchase data in the second preset period are processed by utilizing the least squares method to calculate the category hot-selling ranking, and then build the user's hot-selling category pool.

[0043] In some embodiments, the user's interaction score for the preset category is calculated based on at least one of the user's click preview behavior, like behavior, reply behavior, or share behavior within a third preset period, wherein the third preset period is shorter than the first preset period. The user's interaction category pool is constructed based on the user's interaction score for the preset category.

[0044] For example, the third preset period is a time period of approximately 72 hours.

[0045] In some embodiments, the interaction score Y is:

[0046] Y = click preview behavior score × N1 + like behavior score × N2

[0047] +Reply behavior score × N3 +Share behavior score × N4 (4)

[0048] The total number of preview clicks, likes, replies, and shares of the user in the third preset period is N, N1 is the percentage of preview clicks in N, N2 is the percentage of likes in N, N3 is the percentage of replies in N, and N4 is the percentage of shares in N. That is, N1+N2+N3+N4=100%.

[0049] In step 103, the recommendation weight of each category is determined according to the appearance of each category in the long-term preference category pool, the hot-selling category pool, and the interactive category pool.

[0050] In some embodiments, the recommendation weight W of a specified category is:

[0051] W=U1X1+U2X2+ U3X3 (5)

[0052] Among them, U1 is the appearance of the specified category in the long-term preference category pool. If the specified category appears in the long-term preference category pool, U1 is 1, otherwise U1 is 0, and X1 is the weight of the long-term preference category pool.

[0053] U2 is the occurrence of the specified category in the hot-selling category pool. If the specified category appears in the hot-selling category pool, U2 is 1, otherwise U2 is 0, and X2 is the weight of the hot-selling category pool.

[0054] U3 is the appearance of the specified category in the interactive category pool. If the specified category appears in the interactive category pool, U2 takes 1, otherwise U2 takes 0, and X3 is the weight of the interactive category pool.

[0055] In some embodiments, X3>X1>X2.

[0056] In step 104, a predetermined number of user preference categories are selected in descending order of recommendation weights.

[0057] In step 105, the review posts associated with the user's preferred categories are scored according to predetermined rules.

[0058] It should be noted here that high-quality reviews and order posts are screened out through predetermined rules, that is, users can obtain more information about the products through the reviews and order posts, so as to make up for the deficiency of insufficient product information on the e-commerce platform.

[0059] In step 106, the review listings are recommended to the user in descending order of the rating results.

[0060] For example, by processing the user's historical behavior data, we found that the user is interested in two categories: mobile phones and tablets. By recommending high-quality reviews related to mobile phones and tablets to the user, we can effectively improve the user experience and effectively increase user stickiness and product conversion rate.

[0061] Figure 2 Schematic diagram of the structure of an evaluation and recommendation device according to an embodiment of the present disclosure. Figure 2 As shown, the evaluation and order recommendation device includes a collection module 21, a category pool construction module 22, a recommendation weight determination module 23, a preferred category selection module 24, a scoring module 25 and a recommendation module 26.

[0062] The collection module 21 is configured to collect historical behavior data of users.

[0063] The category pool building module 22 is configured to build a long-term preference category pool, a hot-selling category pool and an interactive category pool associated with the user according to the user's historical behavior data.

[0064] In some embodiments, the category pool building module 22 determines the user's first preference value for the preset category based on the user's category click volume in the first preset period. Determine the user's second preference value for the preset category based on the user's purchase data in the first preset period. The sum of the first preference value and the second preference value is used as the user's preference score for the preset category, and a long-term preference category pool is built based on the preset categories and the corresponding preference scores.

[0065] In some embodiments, the first preset period is a period of approximately three months.

[0066] In some embodiments, the user's preference score for the preset category is calculated by using the TF-IDF algorithm. For example, the user's preference score for the preset category is calculated using the above formulas (1) to (3).

[0067] Since the TF-IDF algorithm itself is not the inventive point of the present disclosure, it will not be described in detail here.

[0068] In some embodiments, the category pool construction module 22 constructs a user's hot-selling category pool based on at least one of the user's order volume, browsing data, attention data, or added purchase data within a second preset period, where the second preset period is greater than the first preset period.

[0069] For example, the second preset period is a period of nearly half a year.

[0070] In some embodiments, the category pool construction module 22 processes the user's order volume, browsing data, attention data and added purchase data in the second preset period by utilizing the least squares method to calculate the category popularity ranking and then construct the user's hot-selling category pool.

[0071] In some embodiments, the category pool building module 22 calculates the user's interaction score for the preset category based on at least one of the user's click preview behavior, like behavior, reply behavior, or share behavior within a third preset period, where the third preset period is shorter than the first preset period. The user's interactive category pool is built based on the user's interaction score for the preset category.

[0072] For example, the third preset period is a time period of approximately 72 hours.

[0073] In some embodiments, the interaction score Y is calculated using the above formula (4).

[0074] The recommendation weight determination module 23 is configured to determine the recommendation weight of each category according to the appearance of each category in the long-term preference category pool, the hot-selling category pool and the interactive category pool.

[0075] In some embodiments, the recommendation weight W of a specified category is calculated using the above formula (5).

[0076] The preference category selection module 24 is configured to select the preference categories of a predetermined number of users in descending order of recommendation weights.

[0077] The scoring module 25 is configured to score the review posts associated with the user's preferred categories according to predetermined rules.

[0078] The recommendation module 26 is configured to recommend the review and order to the user in descending order of the rating results.

[0079] Figure 3 FIG. 1 is a schematic diagram of a structure of a device for evaluating and recommending orders according to another embodiment of the present disclosure. Figure 3 As shown, the evaluation and recommendation device includes a memory 31 and a processor 32.

[0080] The memory 31 is used to store instructions. The processor 32 is coupled to the memory 31. The processor 32 is configured to execute the instructions stored in the memory to implement the following. Figure 1The method of any one of the embodiments.

[0081] like Figure 3 As shown, the evaluation and recommendation device also includes a communication interface 33 for information exchange with other devices. At the same time, the evaluation and recommendation device also includes a bus 34, and the processor 32, the communication interface 33, and the memory 31 communicate with each other through the bus 34.

[0082] The memory 31 may include a high-speed RAM memory, and may also include a non-volatile memory, such as at least one disk memory. The memory 31 may also be a memory array. The memory 31 may also be divided into blocks, and the blocks may be combined into virtual volumes according to certain rules.

[0083] In addition, the processor 32 may be a central processing unit (CPU), or may be an application specific integrated circuit (ASIC), or may be configured to implement one or more integrated circuits of the embodiments of the present disclosure.

[0084] The present disclosure also relates to a computer-readable storage medium, wherein the computer-readable storage medium stores computer instructions, which are executed by a processor to implement the following Figure 1 The method of any one of the embodiments.

[0085] In some embodiments, the functional unit module described above can be implemented as a general-purpose processor, a programmable logic controller (PLC), a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components or any appropriate combination thereof for performing the functions described in the present disclosure.

[0086] A person skilled in the art will understand that all or part of the steps to implement the above embodiments may be accomplished by hardware or by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, and the above-mentioned storage medium may be a read-only memory, a disk or an optical disk, etc.

[0087] The description of the present disclosure is given for the purpose of illustration and description, and is not intended to be exhaustive or to limit the present disclosure to the disclosed form. Many modifications and variations will be apparent to those of ordinary skill in the art. The embodiments are selected and described in order to better illustrate the principles and practical applications of the present disclosure, and to enable those of ordinary skill in the art to understand the present disclosure and thereby design various embodiments with various modifications suitable for specific uses.

Claims

1. A method for evaluating and recommending orders, comprising: Collect historical behavior data of users; A long-term preference category pool, a hot-selling category pool and an interactive category pool associated with the user are constructed according to the historical behavior data of the user, wherein the long-term preference category pool is associated with the category click volume and added-to-purchase data of the user in a first preset period, the hot-selling category pool is associated with at least one of the order volume, browsing data, attention data or added-to-purchase data of the user in a second preset period, and the interactive category pool is associated with at least one of the click preview behavior, like behavior, reply behavior or sharing behavior of the user in a third preset period, wherein the second preset period is greater than the first preset period, and the third preset period is less than the first preset period; Determine the recommendation weight of each category according to the appearance of each category in the long-term preferred category pool, the hot-selling category pool and the interactive category pool; Selecting a predetermined number of the user's preferred categories in descending order of the recommendation weights; Scoring the reviews and listings associated with the user's preferred categories according to predetermined rules; Recommending the review and order to the user in descending order of the rating results; Among them, the recommendation weight W of the designated category is: W=U1X1+U2X2+U3X3, among which, U1 is the appearance of the designated category in the long-term preference category pool. If the designated category appears in the long-term preference category pool, U1 takes 1, otherwise U1 takes 0, and X1 is the weight of the long-term preference category pool; U2 is the appearance of the designated category in the hot-selling category pool. If the designated category appears in the hot-selling category pool, U2 takes 1, otherwise U2 takes 0, and X2 is the weight of the hot-selling category pool; U3 is the appearance of the designated category in the interactive category pool. If the designated category appears in the interactive category pool, U2 takes 1, otherwise U2 takes 0, and X3 is the weight of the interactive category pool, X3>X1>X2.

2. The method according to claim 1, wherein: Building a long-term preference category pool associated with the user based on the user's historical behavior data includes: Determining a first preference value of the user for a preset category according to the category click volume of the user in a first preset period; Determining a second preference value of the user for the preset category according to the additional purchase data of the user in the first preset period; Using the sum of the first preference value and the second preference value as the user's preference score for the preset category; The long-term preference category pool is constructed according to the preset categories and the corresponding preference scores.

3. The method according to claim 2, wherein: Building an interactive category pool associated with the user according to the historical behavior data of the user includes: Calculating the user's interaction score for the preset category according to at least one of the user's click preview behavior, like behavior, reply behavior, or share behavior within a third preset period; An interactive category pool for the user is constructed according to the interaction score of the user for the preset categories.

4. The method according to claim 3, wherein: The interaction score Y is: Y = click preview behavior score × N1 + like behavior score × N2 + reply behavior score × N3 + share behavior score × N4 Among them, the total number of click preview behaviors, like behaviors, reply behaviors and sharing behaviors of the user within the third preset period is N, N1 is the percentage of click preview behaviors in the N, N2 is the percentage of like behaviors in the N, N3 is the percentage of reply behaviors in the N, and N4 is the percentage of sharing behaviors in the N.

5. A device for evaluating and recommending orders, comprising: A collection module, configured to collect historical behavior data of users; a category pool construction module, configured to construct a long-term preference category pool, a hot-selling category pool and an interactive category pool associated with the user according to the historical behavior data of the user, wherein the long-term preference category pool is associated with the category click volume and added-to-purchase data of the user in a first preset period, the hot-selling category pool is associated with at least one of the order volume, browsing data, attention data or added-to-purchase data of the user in a second preset period, and the interactive category pool is associated with at least one of the click preview behavior, like behavior, reply behavior or sharing behavior of the user in a third preset period, wherein the second preset period is greater than the first preset period, and the third preset period is less than the first preset period; The recommendation weight determination module is configured to determine the recommendation weight of each category according to the appearance of each category in the long-term preference category pool, the hot-selling category pool and the interactive category pool, wherein the recommendation weight W of the specified category is: W=U1X1+U2X2+U3X3, wherein U1 is the appearance of the specified category in the long-term preference category pool, if the specified category appears in the long-term preference category pool, U1 takes 1, otherwise U1 takes 0, and X1 is the weight of the long-term preference category pool; U2 is the appearance of the specified category in the hot-selling category pool, if the specified category appears in the hot-selling category pool, U2 takes 1, otherwise U2 takes 0, and X2 is the weight of the hot-selling category pool; U3 is the appearance of the specified category in the interactive category pool, if the specified category appears in the interactive category pool, U2 takes 1, otherwise U2 takes 0, and X3 is the weight of the interactive category pool, X3>X1>X2; A preference category selection module is configured to select a predetermined number of the user's preference categories in descending order of the recommendation weights; A scoring module is configured to score the review and order sharing associated with the user's preferred category according to a predetermined rule; The recommendation module is configured to recommend the evaluation and order sharing to the user in descending order of the rating results.

6. A device for evaluating and recommending orders, comprising: a memory configured to store instructions; A processor is coupled to the memory, and the processor is configured to execute the method according to any one of claims 1 to 4 based on instructions stored in the memory.

7. A computer-readable storage medium, wherein: The computer-readable storage medium stores computer instructions, and when the instructions are executed by a processor, the method according to any one of claims 1 to 4 is implemented.

Images