CN111127152A - Commodity recommendation method, device and equipment based on user preference prediction and readable medium - Google Patents
Commodity recommendation method, device and equipment based on user preference prediction and readable medium Download PDFInfo
- Publication number
- CN111127152A CN111127152A CN201911339787.3A CN201911339787A CN111127152A CN 111127152 A CN111127152 A CN 111127152A CN 201911339787 A CN201911339787 A CN 201911339787A CN 111127152 A CN111127152 A CN 111127152A
- Authority
- CN
- China
- Prior art keywords
- target
- preference
- historical
- target user
- determining
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0631—Item recommendations
Landscapes
- Business, Economics & Management (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The embodiment of the invention discloses a commodity recommendation method, a device, equipment and a readable medium based on user preference prediction, wherein the method comprises the following steps: historical shopping data of a target user is obtained, and historical preference degrees of the target user for a plurality of target commodities are determined according to the historical shopping data; acquiring a shopping request sent by the target user, and determining the current shopping data of the target user according to the shopping request; determining the target preference degree of the target user for each target commodity according to the current shopping data and the historical preference degree; and determining a target recommended commodity corresponding to the target user according to the target preference degree. The invention improves the accuracy of user preference prediction and commodity recommendation.
Description
Technical Field
The invention relates to the technical field of big data processing, in particular to a commodity recommendation method, device and equipment based on user preference prediction and a readable medium.
Background
With the development of economy, the popularization of internet application and the improvement of the living standard of people, shopping behaviors based on various large network e-commerce platforms are widely popularized. Based on the characteristic that the online sales platform is different from online physical shopping, in order to obtain rich commodity information and good shopping experience under the condition that a user cannot contact a commodity entity, each large e-commerce platform can recommend commodities which may be interested to the user.
In the existing e-commerce platform, commodity recommendation for a user is generally carried out on similar commodities based on historical purchase data of the user, but shopping preferences of the user are constantly changed, and the accuracy of recommendation only by means of the historical data is not high, so that the shopping experience of the user is poor, and the marketing effect of the e-commerce platform is poor.
Disclosure of Invention
In view of the above, it is necessary to provide a product recommendation method, device, computer device and readable medium based on user preference prediction.
A commodity recommendation method based on user preference prediction is characterized by comprising the following steps:
historical shopping data of a target user is obtained, and historical preference degrees of the target user for a plurality of target commodities are determined according to the historical shopping data;
acquiring a shopping request sent by the target user, and determining the current shopping data of the target user according to the shopping request;
determining the target preference degree of the target user for each target commodity according to the current shopping data and the historical preference degree;
and determining a target recommended commodity corresponding to the target user according to the target preference degree.
Wherein the determining the historical preference of the target user for a plurality of target commodities according to the historical shopping data comprises:
determining a historical operation type of the target user for each target commodity according to the historical shopping data, wherein the historical operation type comprises the following steps: not clicked, browsed, added to a shopping cart, purchased;
and determining the historical preference of the target user for each target commodity according to a preset preference weight value corresponding to the historical operation type.
The obtaining of the current shopping data of the target user and the determining of the target preference degree of the target user for each target commodity according to the current shopping data and the historical preference degree comprise:
determining the total number of target commodities browsed by the target user currently and the current purchase times of each target commodity according to the current shopping data;
determining the current preference degree of the target user for each target commodity according to the total number of the target commodities and the current purchasing times;
and determining the target preference degree of the target user for each target commodity according to the current preference degree and the historical preference degree.
The determining the target preference degree of the target user for each target commodity according to the current preference degree and the historical preference degree comprises the following steps:
acquiring the ending time point of the latest historical shopping behavior of the target user and the starting time point of the current shopping behavior;
determining a historical influence weight value of the historical preference of the target user for each target commodity according to the ending time point and the starting time point;
and determining the target preference degree of the target user for each target commodity according to the historical influence weight value, the historical preference degree and the current preference degree.
Further, the determining the historical influence weight of the historical preference of the target user for each target commodity according to the ending time point and the starting time point comprises:
determining a shopping time interval of the target user according to the end time point and the start time point;
determining a preference change state of the target user according to a comparison result of the shopping time interval and a preset time interval threshold, wherein the preference change state comprises a preference stable period, a preference forgetting period and a complete forgetting period;
determining the historical impact weight according to the preference change state.
The determining the target recommended commodity corresponding to the target user according to the target preference of the target user for each target commodity comprises:
acquiring the total number of users of the target user and the total number of commodities of the target commodity;
determining preference similarity among the target users according to the total number of the users, the total number of the commodities and the target preference;
and determining the target recommended commodity corresponding to the target user according to the preference similarity.
Wherein, the determining the target recommended commodity corresponding to the target user according to the preference similarity comprises:
determining other users with preference similarity larger than a preset similarity threshold value with the target user as reference users of the target user according to the preference similarity;
acquiring a target preference degree of the reference user for each target commodity as a reference preference degree, and acquiring target commodities, other than the purchased target commodities, of which the historical operation types corresponding to the target user are to be recommended as commodities to be recommended;
and determining a target recommended commodity corresponding to the target user in the commodities to be recommended according to the reference preference degree.
An article recommendation apparatus based on user preference prediction, the apparatus comprising:
a first acquisition unit: the system comprises a data acquisition module, a data processing module and a display module, wherein the data acquisition module is used for acquiring historical shopping data of a target user and determining historical preference of the target user for a plurality of target commodities according to the historical shopping data;
a second acquisition unit: the system comprises a shopping server, a shopping server and a shopping server, wherein the shopping server is used for acquiring a shopping request sent by a target user and determining current shopping data of the target user according to the shopping request;
a first determination unit: the target preference degree of the target user for each target commodity is determined according to the current shopping data and the historical preference degree;
a second determination unit: and the target recommendation commodity corresponding to the target user is determined according to the target preference.
A computer device comprising a memory and a processor, the memory storing a computer program that, when executed by the processor, causes the processor to perform the steps of:
historical shopping data of a target user is obtained, and historical preference degrees of the target user for a plurality of target commodities are determined according to the historical shopping data;
acquiring a shopping request sent by the target user, and determining the current shopping data of the target user according to the shopping request;
determining the target preference degree of the target user for each target commodity according to the current shopping data and the historical preference degree;
and determining a target recommended commodity corresponding to the target user according to the target preference degree.
A computer-readable storage medium storing a computer program which, when executed by a processor, causes the processor to perform the steps of:
historical shopping data of a target user is obtained, and historical preference degrees of the target user for a plurality of target commodities are determined according to the historical shopping data;
acquiring a shopping request sent by the target user, and determining the current shopping data of the target user according to the shopping request;
determining the target preference degree of the target user for each target commodity according to the current shopping data and the historical preference degree;
and determining a target recommended commodity corresponding to the target user according to the target preference degree.
In the embodiment of the invention, historical shopping data of a target user is firstly acquired, and the historical preference of the target user for a plurality of target commodities is determined according to the historical shopping data. And after the historical preference degree is determined, acquiring a shopping request sent by the target user in real time, and determining the current shopping data of the target user according to the shopping request. And finally, determining the target preference of the target user for each target commodity according to the current shopping data and the historical preference, and determining the target recommended commodity corresponding to the target user according to the target preference.
Therefore, compared with the prior art that the current preference of the user is predicted only according to the historical shopping data of the user, and the decline and the migration of the preference of the user are ignored along with the change of time, the invention combines the historical preference and the current preference and refers to the shopping data of other users which are similar to the preference of the target user to recommend commodities to the target user. The accuracy of the preference prediction of the user is improved, and the accuracy and the efficiency of commodity recommendation based on the preference prediction are indirectly improved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
Wherein:
FIG. 1 shows a flow diagram of a method for merchandise recommendation based on user preference prediction in one embodiment;
FIG. 2 illustrates a flow diagram for determining historical preferences of the target user for a plurality of target items in one embodiment;
FIG. 3 illustrates a flow diagram for determining a target preference of the target user for each of the target items in one embodiment;
FIG. 4 illustrates a flow diagram for determining a target preference for each of the target items for the target user based on the current preference and historical preferences, in one embodiment;
FIG. 5 illustrates a flow diagram for determining a historical impact weight value for the target user's historical preference for each target good in one embodiment;
FIG. 6 is a flow diagram that illustrates the determination of a target recommended good corresponding to the target user based on target preferences, under an embodiment;
FIG. 7 is a flow diagram that illustrates the determination of a target recommended good for the target user based on the preference similarity, in one embodiment;
FIG. 8 is a block diagram showing a configuration of an article recommending apparatus based on user preference prediction in one embodiment;
FIG. 9 is a diagram illustrating an internal structure of a computer device in one embodiment.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The invention provides a commodity recommendation method based on user preference prediction.
Referring to fig. 1, an embodiment of the present invention provides a commodity recommendation method based on user preference prediction.
FIG. 1 shows a flow diagram of a method for merchandise recommendation based on user preference prediction in one embodiment. The commodity recommendation method based on user preference prediction in the present invention at least includes steps S1022 to S1028 shown in fig. 1, which are described in detail as follows:
in step S1022, historical shopping data of a target user is acquired, and historical preference degrees of the target user for a plurality of target commodities are determined according to the historical shopping data.
Wherein, step S1022 may also include at least steps S1032-S1034 shown in fig. 2. This is explained below with reference to fig. 2. FIG. 2 illustrates a flow diagram for determining historical preferences of the target user for a plurality of target items, in one embodiment.
In step S1032, determining a historical operation type of the target user for each target product according to the historical shopping data, where the historical operation type includes: not clicked, viewed, added to shopping cart, purchased.
There may first be a number of preset items waiting to be selected and recommended, in an alternative embodiment, A, B, C, D, E, F and item G, respectively. For the target product to be recommended, the operations performed by the user "Alice" in the past 48 hours may be: the commodity a is purchased, the commodity B, the commodity C, and the commodity D are added to the shopping cart, the commodity E and the commodity F are browsed, and the commodity G is not clicked.
It is noted that it is contemplated that clicking, browsing, joining a shopping cart, and ultimately purchasing may occur simultaneously in the consumption process for a good and in a sequential, progressive order in time. That is, a consumer generally clicks a certain commodity, enters an information display page of the commodity for browsing, and then can choose to add a shopping cart or directly purchase the commodity, or quit the information display page of the commodity after browsing, and does not perform the next operation.
Therefore, when one to-be-recommended commodity corresponds to more than one historical operation type, the type closest to the consumption behavior end point (namely purchase) in the time sequence of the commodity is selected.
In step S1034, the historical preference of the target user for each target product is determined according to a preset preference weight value corresponding to a historical operation type.
It is easy to understand that in the actual online shopping process, a consumer firstly clicks a commodity which is interested by the consumer to check and compare commodity information, and after basically determining a commodity which is preferred by the consumer and has a high purchase probability, the commodity is generally added into a shopping cart so as to be checked out and purchased, and under normal conditions, the commodity paid and purchased by the consumer is the largest preference of the commodity in a current commodity library to be recommended.
Therefore, the preference weighting values corresponding to the above history operation types may be arranged as follows: purchased > shopping cart joined > browsed > not clicked. In an optional embodiment, the history preference degree of the target user for each target product may be determined according to the number of times of execution of the history final operation on the target product by the target user and a weight value corresponding to the operation.
In step S1024, a shopping request sent by the target user is obtained, and the current shopping data of the target user is determined according to the shopping request.
For example, a user "Alice" may start clicking and browsing (e.g., in a manner of sliding up and down) a commodity list after entering a commodity display interface in a preset shopping system.
The commodities existing in the current commodity list may be commodities P1, P2, P3, P4, P5, and P6, respectively, from top to bottom. And the user "Alice" browses and views in the detail pages of the 5 items clicked respectively while sliding from P1 to P5 in turn, and finally adds the items P3, P4 to the shopping cart and purchases item P2.
In step S1026, a target preference degree of the target user for each target product is determined according to the current shopping data and the historical preference degree.
Specifically, step S1026 may further include steps S1042 to S1046 shown in fig. 3. FIG. 3 illustrates a flow diagram for determining a target preference of the target user for each of the target items in one embodiment.
In step S1042, the total number of target products currently browsed by the target user and the current purchase frequency of each target product are determined according to the current shopping data.
In connection with the example in step S1024, the total number of the commodities browsed by the target user "Alice" is 5, and the current purchase times for the commodities P1, P2, P3, P4, and P5 are 0, 1, 0, and 0, respectively.
In step S1044, determining a current preference of the target user for each target product according to the total number of the target products and the current number of purchases.
Specifically, the calculation of the current preference degree of the target user for each target commodity may be performed according to the following process:
the difference value of the purchase times between the commodity with the largest purchase time and the commodity with the smallest purchase time in all the target commodities is obtained as denominator according to the three categories of 'purchased', 'added shopping cart' and 'browsed', the difference value of the historical average purchase times of the three categories of commodities minus the purchase times of the commodity with the smallest purchase time is used as numerator, three numerical values are determined, and the sum of the three numerical values is used as the current preference degree.
In step S1046, determining a target preference degree of the target user for each target commodity according to the current preference degree and the historical preference degree.
Considering that the time interval between the current shopping behavior and the end of the historical shopping behavior of the user may be greatly different, the time interval between the operation of logging in the online shopping website every time is short, and is minutes, and is days, even weeks or months. It is easy to understand that the user's preference for goods is also based on the principle of human cognition and memory, i.e. it is also in line with the human memory forgetting curve. That is, over time, the user's preference for a particular thing decays proportionally over time before the next occurrence of cognitive enhancing stimuli occurs.
The consumer preference for goods as applied to a shopping scenario is such that the user may be able to obtain/learn new preferences that are more stable over a shorter time interval (perhaps 3-7 days, taking into account the human memory threshold and the length of the short-term memory storage period), i.e. the user's liking or disliking of the target goods will remain unchanged for a period of time. When a long time (for example, 7 days or more) has elapsed, the user's memory of each target product begins to decline, that is, the like/dislike degree of the product gradually approaches toward neutral. After a longer transition, the user's preferences have substantially completed the preference neutralization, i.e., are in a no preference state.
In summary, in order to correct the determined target preference degrees for the respective target commodities based on the relationship between the preferences of people and time, step S1046 may further include steps S1052 to S1056 shown in fig. 4. FIG. 4 illustrates a flow diagram for determining the target preference of the target user for each of the target items based on the current preference and historical preferences, in one embodiment.
In step S1052, an end time point of the last historical shopping behavior of the target user and a start time point of the current shopping behavior are acquired.
Specifically, the user "Alice" may have exited the online shopping website last time in 12/2019 at 16 days 20:15:08, while the current shopping behavior starts in 2019 at 12/19 days 14:15:50 when the user "Alice" logs in the online shopping website.
In step S1054, a history influence weight value of the history preference of the target user for each target product is determined according to the ending time point and the starting time point.
It is easy to understand that, here, the end time point and the start time point are obtained to determine the forgetting period of the user for the online shopping website, so as to determine the magnitude of the transition and the attenuation degree of the preference according to the length of the forgetting period. Thus, step S1054 may again include steps S1062-S1066 shown in FIG. 5. FIG. 5 illustrates a flow diagram for determining a historical impact weight value for the target user's historical preference for each target item, in one embodiment.
In step S1062, a shopping time interval of the target user is determined according to the ending time point and the starting time point.
According to the example in step S1052, it can be calculated that the shopping time interval of the user "Alice" is 66 hours.
In step S1064, a preference change state of the target user is determined according to a comparison result between the shopping time interval and a preset time interval threshold, where the preference change state includes a preference stable period, a preference forgetting period, and a complete forgetting period.
Two duration thresholds may first be set to define the period of time that the degree of migration of the user's preferences (i.e., the preference change state) is in. If the preference variation shape of the user with the shopping time interval of less than or equal to 48 hours (the first duration threshold) is determined as the preference stable period, correspondingly, the preference variation shape of the user with the shopping time interval of less than or equal to 48 hours is determined as the preference forgetting period. Thus, the preference change state of the user "Alice" in the foregoing step is determined as the preference forgetting period.
At the same time, a second duration threshold, such as 168 hours (i.e., 7 days), may be further determined, and when the shopping interval is greater than the memory limit threshold, the user is considered to have been in a complete forgetting period, during which the user's preference level for each target item is all zeroed.
In step S1066, the historical impact weight is determined according to the preference change state.
In the preference stabilization period, the historical influence weight corresponding to the historical preference of the target user on each target commodity can be 1, that is, the preference of the user does not change.
In the preference forgetting period, the history influence weight corresponding to the history preference of the target user for each target commodity may be 0.5 (i.e., less than 1), that is, the preference of the user for each target commodity is gradually attenuated on the basis of the history preference.
Correspondingly, if the user is in the complete forgetting period, the influence of the user on the historical preference of each target commodity is weak, so the historical influence weight corresponding to the historical preference of the target user on each target commodity at the period can be set to be 0.
In step S1056, a target preference degree of the target user for each target product is determined according to the historical influence weight value, the historical preference degree, and the current preference degree.
Specifically, the target preference of the target user for each target commodity may be calculated according to the following process:
in the case where the history influence weight value of the target user is 0 (i.e., the user is in the complete forgetting period), the target preference degree for each target item is the current preference degree.
And in the case that the historical influence weight value of the target user is not 0 (namely the user is in a preference forgetting period or a preference stabilizing period), the target preference of each target commodity is the sum of the product of the historical influence weight value and the historical preference plus the product of the current preference and 1 minus the historical influence weight value.
In step S1028, a target recommended item corresponding to the target user is determined according to the target preference degree.
Finally, this step may also include steps S1072-S1076 shown in FIG. 6. FIG. 6 shows a flow diagram for determining a target recommended good corresponding to the target user according to a target preference in one embodiment.
In step S1072, the total number of users of the target user and the total number of products of the target product are acquired.
Specifically, there may be m target users and n target goods. Here, each target user is considered as a point in the space, and the target preference of each target user for each commodity is considered as a vector in the space.
In step S1074, a preference similarity between the target users is determined according to the total number of users, the total number of products, and the target preference.
The collaborative filtering idea is adopted, a large amount of user preference data of a large amount of commodities are used as an analysis basis, and not only the current and historical shopping data of a certain user are analyzed, but also the similarity is analyzed according to the preference of a large amount of users, so that recommendation is carried out.
The specific process of determining the preference similarity between users may be performed as follows:
assuming that preference similarity between a user "Alice" and a user "Bob" is calculated, a commodity set in which the user "Alice" and the user "Bob" have operations in common is obtained first, and average preference of the user "Alice" and the user "Bob" for all commodities in the commodity set and target preference of the user "Alice" and the user "Bob" for a specific commodity K in the commodity set are determined. And determining the preference similarity of the two users according to the deviation degree of the target preference of the two users for each commodity in the commodity set from the average preference.
In step S1076, the target recommended product corresponding to the target user is determined according to the preference similarity.
Step S1076 may in turn include steps S1082-S1086 shown in fig. 7. FIG. 7 is a flowchart illustrating determining a target recommended goods corresponding to the target user according to the preference similarity in one embodiment.
In step S1082, determining, according to the preference similarity, other users whose preference similarity to the target user is greater than a preset similarity threshold as reference users of the target user.
In step S1084, the target preference degree of the reference user for each target product is obtained as a reference preference degree, and a target product other than the purchased target product is obtained as a product to be recommended, where the historical operation type corresponding to the target user is obtained.
If the reference users of the user 'Alice' are determined to be the user 'Bob' and the user 'Chris', the target commodity which is not browsed or purchased by the current target user 'Alice' is taken as the commodity to be recommended.
In step S1086, a target recommended product corresponding to the target user is determined in the to-be-recommended product according to the reference preference degree.
Specifically, the average preference degrees of the reference users "Bob" and "Chris" and the target user "Alice" for the to-be-recommended goods, the preference similarities between "Bob" and "Chris" and the target user "Alice", respectively, and the target preference degrees of "Bob" and "Chris" for each to-be-recommended goods are obtained, and the target recommended goods corresponding to "Alice" are finally determined from the recommended goods according to the parameters.
FIG. 8 is a block diagram showing a configuration of an article recommendation device based on user preference prediction in one embodiment.
Referring to fig. 8, an article recommendation device 1090 based on user preference prediction according to an embodiment of the present invention includes: a first acquiring unit 1092, a second connecting unit 1094, a first determining unit 1096, a second determining unit 1098.
Wherein, the first obtaining unit 1092: the system comprises a data acquisition module, a data processing module and a display module, wherein the data acquisition module is used for acquiring historical shopping data of a target user and determining historical preference of the target user for a plurality of target commodities according to the historical shopping data;
the second acquiring unit 1094: the system comprises a shopping server, a shopping server and a shopping server, wherein the shopping server is used for acquiring a shopping request sent by a target user and determining current shopping data of the target user according to the shopping request;
the first determination unit 1096: the target preference degree of the target user for each target commodity is determined according to the current shopping data and the historical preference degree;
the second determination unit 1098: and the target recommendation commodity corresponding to the target user is determined according to the target preference.
FIG. 9 is a diagram illustrating an internal structure of a computer device in one embodiment. The computer device may specifically be a terminal, and may also be a server. As shown in fig. 9, the computer device includes a processor, a memory, and a communication module, a processing module, a presentation module, which are connected by a system bus. Wherein the memory includes a non-volatile storage medium and an internal memory. The non-volatile storage medium of the computer device stores an operating system and may also store a computer program that, when executed by the processor, causes the processor to implement the present user preference prediction based item recommendation method. The internal memory may also have stored therein a computer program that, when executed by the processor, causes the processor to perform the method of merchandise recommendation based on the user preference prediction. Those skilled in the art will appreciate that the architecture shown in fig. 9 is merely a block diagram of some of the structures associated with the disclosed aspects and is not intended to limit the computing devices to which the disclosed aspects apply, as particular computing devices may include more or less components than those shown, or may combine certain components, or have a different arrangement of components.
In one embodiment, a computer device is proposed, comprising a memory and a processor, the memory storing a computer program which, when executed by the processor, causes the processor to perform the steps of:
historical shopping data of a target user is obtained, and historical preference degrees of the target user for a plurality of target commodities are determined according to the historical shopping data; acquiring a shopping request sent by the target user, and determining the current shopping data of the target user according to the shopping request; determining the target preference degree of the target user for each target commodity according to the current shopping data and the historical preference degree; determining a target recommended commodity corresponding to the target user according to the target preference degree
In one embodiment, a computer-readable storage medium is proposed, in which a computer program is stored which, when executed by a processor, causes the processor to carry out the steps of:
historical shopping data of a target user is obtained, and historical preference degrees of the target user for a plurality of target commodities are determined according to the historical shopping data; acquiring a shopping request sent by the target user, and determining the current shopping data of the target user according to the shopping request; determining the target preference degree of the target user for each target commodity according to the current shopping data and the historical preference degree; determining a target recommended commodity corresponding to the target user according to the target preference degree
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by a computer program, which can be stored in a non-volatile computer-readable storage medium, and can include the processes of the embodiments of the methods described above when the program is executed. Any reference to memory, storage, database, or other medium used in the embodiments provided herein may include non-volatile and/or volatile memory, among others. Non-volatile memory can include read-only memory (ROM), Programmable ROM (PROM), Electrically Programmable ROM (EPROM), Electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDRSDRAM), Enhanced SDRAM (ESDRAM), Synchronous Link DRAM (SLDRAM), Rambus Direct RAM (RDRAM), direct bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).
The technical features of the above embodiments can be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the above embodiments are not described, but should be considered as the scope of the present specification as long as there is no contradiction between the combinations of the technical features.
The above-mentioned embodiments only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the present application. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present patent shall be subject to the appended claims.
Claims (10)
1. A commodity recommendation method based on user preference prediction is characterized by comprising the following steps:
historical shopping data of a target user is obtained, and historical preference degrees of the target user for a plurality of target commodities are determined according to the historical shopping data;
acquiring a shopping request sent by the target user, and determining the current shopping data of the target user according to the shopping request;
determining the target preference degree of the target user for each target commodity according to the current shopping data and the historical preference degree;
and determining a target recommended commodity corresponding to the target user according to the target preference degree.
2. The method of claim 1, wherein said determining historical preferences of said target user for a plurality of target items from said historical shopping data comprises:
determining a historical operation type of the target user for each target commodity according to the historical shopping data, wherein the historical operation type comprises the following steps: not clicked, browsed, added to a shopping cart, purchased;
and determining the historical preference of the target user for each target commodity according to a preset preference weight value corresponding to the historical operation type.
3. The method of claim 1, wherein the obtaining current shopping data of the target user and determining the target preference of the target user for each target commodity according to the current shopping data and the historical preference comprises:
determining the total number of target commodities browsed by the target user currently and the current purchase times of each target commodity according to the current shopping data;
determining the current preference degree of the target user for each target commodity according to the total number of the target commodities and the current purchasing times;
and determining the target preference degree of the target user for each target commodity according to the current preference degree and the historical preference degree.
4. The method of claim 3, wherein determining the target preference of the target user for each of the target items according to the current preference and the historical preference comprises:
acquiring the ending time point of the latest historical shopping behavior of the target user and the starting time point of the current shopping behavior;
determining a historical influence weight value of the historical preference of the target user for each target commodity according to the ending time point and the starting time point;
and determining the target preference degree of the target user for each target commodity according to the historical influence weight value, the historical preference degree and the current preference degree.
5. The method of claim 4, wherein determining the historical impact weight of the target user's historical preference for each target item as a function of the ending time point and the starting time point comprises:
determining a shopping time interval of the target user according to the end time point and the start time point;
determining a preference change state of the target user according to a comparison result of the shopping time interval and a preset time interval threshold, wherein the preference change state comprises a preference stable period, a preference forgetting period and a complete forgetting period;
determining the historical impact weight according to the preference change state.
6. The method according to claim 1, wherein the determining the target recommended product corresponding to the target user according to the target preference of the target user for each target product comprises:
acquiring the total number of users of the target user and the total number of commodities of the target commodity;
determining preference similarity among the target users according to the total number of the users, the total number of the commodities and the target preference;
and determining the target recommended commodity corresponding to the target user according to the preference similarity.
7. The method according to claim 6, wherein the determining the target recommended goods corresponding to the target user according to the preference similarity comprises:
determining other users with preference similarity larger than a preset similarity threshold value with the target user as reference users of the target user according to the preference similarity;
acquiring a target preference degree of the reference user for each target commodity as a reference preference degree, and acquiring target commodities, other than the purchased target commodities, of which the historical operation types corresponding to the target user are to be recommended as commodities to be recommended;
and determining a target recommended commodity corresponding to the target user in the commodities to be recommended according to the reference preference degree.
8. An article recommendation apparatus based on user preference prediction, the apparatus comprising:
a first acquisition unit: the system comprises a data acquisition module, a data processing module and a display module, wherein the data acquisition module is used for acquiring historical shopping data of a target user and determining historical preference of the target user for a plurality of target commodities according to the historical shopping data;
a second acquisition unit: the system comprises a shopping server, a shopping server and a shopping server, wherein the shopping server is used for acquiring a shopping request sent by a target user and determining current shopping data of the target user according to the shopping request;
a first determination unit: the target preference degree of the target user for each target commodity is determined according to the current shopping data and the historical preference degree;
a second determination unit: and the target recommendation commodity corresponding to the target user is determined according to the target preference.
9. A readable storage medium storing a computer program which, when executed by a processor, causes the processor to perform the steps of the method according to any one of claims 1 to 7.
10. A computer device comprising a memory and a processor, the memory storing a computer program that, when executed by the processor, causes the processor to perform the steps of the method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911339787.3A CN111127152A (en) | 2019-12-23 | 2019-12-23 | Commodity recommendation method, device and equipment based on user preference prediction and readable medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911339787.3A CN111127152A (en) | 2019-12-23 | 2019-12-23 | Commodity recommendation method, device and equipment based on user preference prediction and readable medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111127152A true CN111127152A (en) | 2020-05-08 |
Family
ID=70501276
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911339787.3A Pending CN111127152A (en) | 2019-12-23 | 2019-12-23 | Commodity recommendation method, device and equipment based on user preference prediction and readable medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111127152A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111681086A (en) * | 2020-06-16 | 2020-09-18 | 上海风秩科技有限公司 | Commodity recommendation method and device, computer equipment and readable storage medium |
| CN111815351A (en) * | 2020-05-29 | 2020-10-23 | 杭州览众数据科技有限公司 | Cooperative filtering and association rule-based clothing recommendation method |
| CN112231576A (en) * | 2020-11-05 | 2021-01-15 | 中国联合网络通信集团有限公司 | Target user determination method and device |
| CN112446764A (en) * | 2020-11-30 | 2021-03-05 | 广州三七互娱科技有限公司 | Game commodity recommendation method and device and electronic equipment |
| CN113609375A (en) * | 2021-06-21 | 2021-11-05 | 青岛海尔科技有限公司 | Content recommendation method and device, storage medium and electronic device |
| CN115496566A (en) * | 2022-11-16 | 2022-12-20 | 九州好礼(山东)电商科技有限公司 | Regional specialty recommendation method and system based on big data |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108648049A (en) * | 2018-05-03 | 2018-10-12 | 中国科学技术大学 | A kind of sequence of recommendation method based on user behavior difference modeling |
| CN109214893A (en) * | 2018-08-31 | 2019-01-15 | 深圳春沐源控股有限公司 | Method of Commodity Recommendation, recommender system and computer installation |
| CN109559208A (en) * | 2019-01-04 | 2019-04-02 | 平安科技（深圳）有限公司 | A kind of information recommendation method, server and computer-readable medium |
| CN110570279A (en) * | 2019-09-04 | 2019-12-13 | 深圳创新奇智科技有限公司 | Strategic recommendation method and device based on real-time user behavior |
-
2019
- 2019-12-23 CN CN201911339787.3A patent/CN111127152A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108648049A (en) * | 2018-05-03 | 2018-10-12 | 中国科学技术大学 | A kind of sequence of recommendation method based on user behavior difference modeling |
| CN109214893A (en) * | 2018-08-31 | 2019-01-15 | 深圳春沐源控股有限公司 | Method of Commodity Recommendation, recommender system and computer installation |
| CN109559208A (en) * | 2019-01-04 | 2019-04-02 | 平安科技（深圳）有限公司 | A kind of information recommendation method, server and computer-readable medium |
| CN110570279A (en) * | 2019-09-04 | 2019-12-13 | 深圳创新奇智科技有限公司 | Strategic recommendation method and device based on real-time user behavior |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111815351A (en) * | 2020-05-29 | 2020-10-23 | 杭州览众数据科技有限公司 | Cooperative filtering and association rule-based clothing recommendation method |
| CN111681086A (en) * | 2020-06-16 | 2020-09-18 | 上海风秩科技有限公司 | Commodity recommendation method and device, computer equipment and readable storage medium |
| CN112231576A (en) * | 2020-11-05 | 2021-01-15 | 中国联合网络通信集团有限公司 | Target user determination method and device |
| CN112231576B (en) * | 2020-11-05 | 2023-06-06 | 中国联合网络通信集团有限公司 | Target user determination method and device |
| CN112446764A (en) * | 2020-11-30 | 2021-03-05 | 广州三七互娱科技有限公司 | Game commodity recommendation method and device and electronic equipment |
| CN113609375A (en) * | 2021-06-21 | 2021-11-05 | 青岛海尔科技有限公司 | Content recommendation method and device, storage medium and electronic device |
| CN115496566A (en) * | 2022-11-16 | 2022-12-20 | 九州好礼(山东)电商科技有限公司 | Regional specialty recommendation method and system based on big data |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111127152A (en) | Commodity recommendation method, device and equipment based on user preference prediction and readable medium | |
| CN109559208B (en) | Information recommendation method, server and computer readable medium | |
| WO2020048084A1 (en) | Resource recommendation method and apparatus, computer device, and computer-readable storage medium | |
| CN110413870B (en) | Commodity recommendation method and device and server | |
| CN108959644B (en) | Search ranking method and device, computer equipment and storage medium | |
| US9298809B2 (en) | System, apparatus, and method for recommending items | |
| TWI539305B (en) | Personalized information push method and device | |
| CN110827114A (en) | Commodity recommendation method and device | |
| US8386329B1 (en) | Social network-based recommendation | |
| CN107423430B (en) | Data processing method, device and computer readable storage medium | |
| WO2011019485A1 (en) | Method and system of web page content filtering | |
| CN111506849A (en) | Page generation method and device | |
| CN110163703B (en) | Classification model establishing method, file pushing method and server | |
| CN111159570B (en) | Information recommendation method and server | |
| US8793236B2 (en) | Method and apparatus using historical influence for success attribution in network site activity | |
| WO2020221022A1 (en) | Service object recommendation method | |
| CN112149003B (en) | Commodity community recommendation method and device and computer equipment | |
| CN112132660B (en) | Commodity recommendation method, system, equipment and storage medium | |
| CN111461827B (en) | Push method and device for product evaluation information | |
| KR20160118644A (en) | System and Method for recommending item, and device thereof | |
| CN113379516A (en) | Recommended product determination method and device | |
| CN111680213B (en) | Information recommendation method, data processing method and device | |
| JP6424194B2 (en) | Information processing apparatus, information processing method, and information processing program | |
| CN110889748B (en) | Store platform product recommendation method, store platform product recommendation device, computer equipment and storage medium | |
| CN114722266A (en) | Questionnaire pushing method and device, electronic equipment and readable storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |