CN113689259A - Commodity personalized recommendation method and system based on user behaviors - Google Patents
Commodity personalized recommendation method and system based on user behaviors Download PDFInfo
- Publication number
- CN113689259A CN113689259A CN202110908381.3A CN202110908381A CN113689259A CN 113689259 A CN113689259 A CN 113689259A CN 202110908381 A CN202110908381 A CN 202110908381A CN 113689259 A CN113689259 A CN 113689259A
- Authority
- CN
- China
- Prior art keywords
- commodity
- user behavior
- user
- behavior data
- heat value
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- General Business, Economics & Management (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a commodity personalized recommendation method and system based on user behaviors, wherein the method comprises the following steps: acquiring user behavior data, wherein the user behavior data comprises first user behavior data and second user behavior data, the first user behavior data is behavior data generated by a user terminal on each commodity of the same commodity label, and the second user behavior data is behavior data generated by the user terminal on commodities of different commodity labels; acquiring a comprehensive heat value of each commodity of the same commodity label according to the first user behavior data and the time attenuation coefficient corresponding to the first user behavior data; acquiring a label heat value of each commodity label according to the second user behavior data and the time attenuation coefficient corresponding to the second user behavior data; and generating a commodity recommendation list according to the comprehensive heat value and the label heat value. The invention improves the accuracy and the pushing efficiency of commodity pushing.
Description
Technical Field
The invention relates to the technical field of commodity pushing, in particular to a commodity personalized recommendation method and system based on user behaviors.
Background
The user preferences for different commodities have certain differences, and the popularity of different commodities to the user also has differences. If the user finds that the page contains the commodities meeting the purchase demand of the user in the process of browsing the commodity page, the user can be motivated to purchase the products, and therefore the purchase intention of the user on the commodities is improved.
The intelligent recommendation algorithm is a key technology in the existing commodity pushing application field and is a mainstream scheme for predicting the purchasing intention and preference of consumers at present. The intelligent recommendation technology of the shopping website is developed relatively well at the present stage, and common intelligent recommendation algorithms include a collaborative filtering recommendation method, a content-based recommendation method, association rule-based recommendation, utility-based recommendation, knowledge-based recommendation, combined recommendation and the like.
However, the commodity pushing preferred by the user and the intelligent recommendation of the shopping website are greatly different and cannot be applied to all, and the accuracy and efficiency of commodity recommendation are to be further improved. Therefore, there is a need for a method and a system for personalized recommendation of commodities based on user behavior to solve the above problems.
Disclosure of Invention
Aiming at the problems in the prior art, the invention provides a commodity personalized recommendation method and system based on user behaviors.
The invention provides a commodity personalized recommendation method based on user behaviors, which comprises the following steps:
acquiring user behavior data, wherein the user behavior data comprises first user behavior data and second user behavior data, the first user behavior data is behavior data generated by a user terminal on each commodity of the same commodity label, and the second user behavior data is behavior data generated by the user terminal on commodities of different commodity labels;
acquiring a comprehensive heat value of each commodity of the same commodity label according to the first user behavior data and a time attenuation coefficient corresponding to the first user behavior data; acquiring a label heat value of each commodity label according to the second user behavior data and a time attenuation coefficient corresponding to the second user behavior data;
and generating a commodity recommendation list according to the comprehensive heat value and the label heat value.
According to the commodity personalized recommendation method based on the user behavior provided by the invention, before the user behavior data is acquired, the method further comprises the following steps:
setting a user behavior buried point and a user behavior type preset weight value based on a user behavior type so as to obtain user behavior data according to the user behavior buried point, wherein the user behavior buried point at least comprises entering a commodity detail page, successfully collecting commodities, canceling commodity collection, clicking commodity sharing, confirming commodity search success, confirming commodity arrival notification confirmation success, confirming that commodities are added into a shopping cart, clicking to purchase commodities immediately, clicking to submit commodity orders and successfully paying commodities.
According to the commodity personalized recommendation method based on the user behavior provided by the invention, the step of acquiring the user behavior data comprises the following steps:
acquiring the times of user behaviors generated by the user terminal to the commodity, the types of the user behaviors and the weight of the user behaviors based on the user behavior buried points and the preset weight values of the user behavior types;
and obtaining user behavior data according to the user behavior times, the user behavior types and the user behavior weights.
According to the commodity personalized recommendation method based on the user behaviors, the time attenuation coefficient is obtained based on the creation time of the buried point event corresponding to the commodity browsing history record of the user terminal and the commodity type.
According to the commodity personalized recommendation method based on the user behaviors, the comprehensive heat value is obtained by calculation according to the commodity type, the user behavior weight value corresponding to the user behavior burying point, the time attenuation coefficient, the user behavior statistical frequency and the user behavior burying point creation time.
According to the commodity personalized recommendation method based on the user behaviors, the label heat value is obtained by calculation according to the commodity label type, the user behavior weight value corresponding to the user behavior burying point, the user behavior frequency, the time attenuation coefficient and the user behavior burying point creation time.
According to the commodity personalized recommendation method based on the user behaviors, which is provided by the invention, the commodity recommendation list is generated according to the comprehensive heat value and the label heat value, and the method comprises the following steps:
sorting the comprehensive heat value, and obtaining a commodity sorting result based on each same commodity label according to the sorting result of the comprehensive heat value;
sorting the label heat value of each commodity label to obtain a commodity label heat value sorting result;
selecting target commodities from the commodity sorting results according to the commodity label heat value sorting results, and generating a commodity recommendation list according to the target commodities;
or determining pre-recommended commodities according to the commodity sorting result, and selecting a target commodity from the pre-recommended commodities according to the tag heat value sorting result to generate a commodity recommendation list.
The invention also provides a commodity personalized recommendation system based on user behaviors, which comprises the following steps:
the system comprises a buried point event acquisition module, a data processing module and a data processing module, wherein the buried point event acquisition module is used for acquiring user behavior data, the user behavior data comprises first user behavior data and second user behavior data, the first user behavior data is behavior data generated by a user terminal on each commodity of the same commodity label, and the second user behavior data is behavior data generated by the user terminal on commodities of different commodity labels;
the heat value acquisition module is used for acquiring a comprehensive heat value of each commodity of the same commodity label according to the first user behavior data and a time attenuation coefficient corresponding to the first user behavior data; acquiring a label heat value of each commodity label according to the second user behavior data and a time attenuation coefficient corresponding to the second user behavior data;
and the commodity recommendation module is used for generating a commodity recommendation list according to the comprehensive heat value and the label heat value.
The invention also provides an electronic device, which comprises a memory, a processor and a computer program stored on the memory and capable of running on the processor, wherein the processor executes the program to realize the steps of any one of the commodity personalized recommendation methods based on the user behaviors.
The present invention also provides a non-transitory computer readable storage medium, on which a computer program is stored, which, when being executed by a processor, implements the steps of the method for personalized recommendation of goods based on user behavior as described in any one of the above.
According to the commodity personalized recommendation method and system based on the user behaviors, the popularity value sequence of commodities under the commodity labels which are interested and preferred by the user is mastered according to the user behavior data, so that the commodities are recommended to different users, and the commodity pushing accuracy and the commodity pushing efficiency are improved.
Drawings
In order to more clearly illustrate the technical solutions of the present invention or the prior art, the drawings needed for the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and those skilled in the art can also obtain other drawings according to the drawings without creative efforts.
Fig. 1 is a schematic flow chart of a commodity personalized recommendation method based on user behaviors provided by the invention;
FIG. 2 is a schematic diagram of a product recommendation process based on user behavior and product labels according to the present invention;
FIG. 3 is a schematic structural diagram of a commodity personalized recommendation system based on user behaviors, provided by the invention;
fig. 4 is a schematic structural diagram of an electronic device provided in the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings, and it is obvious that the described embodiments are some, but not all embodiments of the present invention. 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 current commodity recommending method is mainly used for carrying out user portrait according to browsing records of a user on commodities of a shopping website and historical purchasing conditions. For example, similar products are recommended according to the user browsing records: when the user frequently shows beauty products such as skin care products and cosmetics in the historical purchase record, the user can be presumed to be a high-consumption female user when the user's historical order record contains female clothes and the unit price exceeds ten thousand yuan, or the user purchases products of high-end brand. Therefore, commodities of relatively high-end brands suitable for female application can be recommended for the user according to the current browsing requirements. Meanwhile, the target commodity recommending method can also recommend complementary products according to the browsing records of the user, for example, the user purchases a spectacle frame, and a store or a product capable of optometrically fitting a spectacle lens can be recommended to the user according to the complementary association of the commodities.
In the above commodity recommendation methods, common recommendation algorithms include a collaborative filtering recommendation method, a content-based recommendation method, an association rule-based recommendation method, a utility-based recommendation method, a knowledge-based recommendation method, a combined recommendation method, and the like.
However, the collaborative filtering recommendation method has the problem of expandability, the commodity recommendation quality depends on a historical data set, and the recommendation quality is poor when the system starts; in the content-based recommendation method, various complex attributes are not well processed, and a classifier needs to be constructed by enough data to establish a user portrait and needs to be based on a large amount of user behavior data; the recommendation method based on the association rule has the problems of difficult rule extraction, long time consumption and synonymity of product names, and the generated commodity recommendation result has low personalization degree; in the recommendation method based on the utility, a user must input a utility function, the recommendation is static, and the problems of poor flexibility and attribute overlapping exist; the knowledge-based recommendation method has the problems that knowledge is difficult to obtain and recommendation is static.
The method and the device predict the preference of the user for the content tag according to the operation behavior of the user on the commodity containing different commodity tags. Since a product may contain multiple tags, tagging different products with tags related to the product may increase the amount of basic data analyzed by the user, resulting in a lower deviation rate of recommendations, for example, a certain brand of refrigerator, with tags such as: tags such as a WIFI function, an energy saving, a large capacity, and a multifunctional refrigerator are provided, the preference degree of a user for different tags may be affected by a plurality of factors, and interested goods of the user may also change with time. Therefore, the invention reflects the preference degree of the user to the label commodity according to the behavior times and the behavior types of the user, and recommends the commodity for the user by integrating the behavior times, the behavior types, the time attenuation of the user behavior, the behavior weight, whether the commodity is a household electrical commodity and other factors. The invention is based on ranking the hot value of the commodities with the same label, and recommends the user according to the popularity of the commodities in the labels of the same commodities, thereby increasing the probability of purchasing the user. It should be noted that the present invention may collect behavior data through a user terminal used by a user, where the user terminal may be a personal computer, a mobile phone, a tablet computer, and the like, and the present invention is not limited to this specifically.
Fig. 1 is a schematic flow chart of a method for personalized recommendation of a commodity based on a user behavior according to the present invention, and as shown in fig. 1, the present invention provides a method for personalized recommendation of a commodity based on a user behavior, including:
In the invention, data generated by user behaviors are collected firstly, and the behaviors of different commodities are different due to different purchasing demands of each user. Therefore, according to the difference of the commodity operation of different users, the preference degree of the user for different commodities needs to be known, and commodity recommendation is provided for the user. The invention is explained by a certain APP running on a user terminal, and provides various purchasing plates for users in a mall page of the APP, wherein the purchasing plates comprise a plurality of related commodities such as refrigerators, washing machines, air conditioners, water heaters, color TV sets, kitchen appliances, living appliances, food and drink fresh products, intelligent products and the like. The different behaviors of the user on the operation of different commodities affect the favorite degree of the user on the commodities.
In an embodiment, before the obtaining the user behavior data, the method further includes:
setting a user behavior buried point and a user behavior type preset weight value based on a user behavior type so as to obtain user behavior data according to the user behavior buried point, wherein the user behavior buried point at least comprises entering a commodity detail page, successfully collecting commodities, canceling commodity collection, clicking commodity sharing, confirming commodity search success, confirming commodity arrival notification confirmation success, confirming that commodities are added into a shopping cart, clicking to purchase commodities immediately, clicking to submit commodity orders and successfully paying commodities.
In the invention, behavior access and point burying of a user mainly comprise entering a commodity detail page, successfully collecting commodities, canceling commodity collection, clicking commodity sharing, confirming that commodity searching is successful, informing and confirming that commodity arrival is successful, confirming that commodities are added into a shopping cart, clicking to immediately purchase commodities, clicking to submit commodity orders, successfully paying commodities and the like. According to the preference degree of the user to the commodity, different weight values are given, it needs to be noted that in a subsequent algorithm for calculating the preference score of the commodity label generated by the user behavior (namely when the label heat value is calculated), when the user purchases a certain commodity, the user considers that the same commodity cannot be continuously purchased in a short term, and the label of the commodity completing the purchasing behavior is set to be zero, so that the preference score of the commodity label is recalculated for the user.
Further, on the basis of the foregoing embodiment, the acquiring user behavior data includes: acquiring the times of user behaviors generated by the user terminal to the commodity, the types of the user behaviors and the weight of the user behaviors based on the user behavior buried points and the preset weight values of the user behavior types; and obtaining user behavior data according to the user behavior times, the user behavior types and the user behavior weights.
In the present invention, after a buried point for collecting user behaviors is set, user behavior data is constructed for each user behavior type, user behavior frequency and user behavior weight, where table 1 is a weight value of a user behavior type preset according to a service requirement, and can refer to table 1:
TABLE 1
| Type of user behavior | Degree of interest | User behavior weighted value (w) |
| Detailed page of incoming commodity | Is low in | +1 |
| Success of collecting commodities | In | +5 |
| Canceling the collection of commodities | In | -5 |
| Click to share merchandise | In | +10 |
| Confirming success of a search for a commodity | In | +10 |
| Commodity arrival notification confirmation success | In | +20 |
| Confirming the addition of a good to a shopping cart | In | +20 |
| Click to purchase an item immediately | Height of | +50 |
| Success of commodity payment | Is low in | 0 |
Respectively acquiring first user behavior data and second user behavior data based on user behavior weights preset in table 1, specifically, performing data statistics on the first user behavior data by collecting behaviors of a user on a commodity under the same commodity label, for example, the corresponding commodity label is a commodity of a refrigerator, and multiple brands (such as brand a, brand B and brand C) exist, and the user generates different behaviors through a user terminal of the user, wherein the refrigerator of the brand a is viewed for 5 times, shared for 2 times, added to a shopping cart and the like; the refrigerator of brand B is checked for detailed pages of 1 commodity, 1 share is clicked, the commodity is already collected and the like; see 2 detailed pages of items for brand C refrigerator, click 2 shares, and click 1 to buy items immediately, etc. By acquiring the first user behavior data of the user, the behavior differences of the user under the same commodity label of the refrigerator and different brands can be obtained through statistics.
Further, for the second user behavior data, statistics is carried out by collecting behavior data generated when the user browses commodities with different commodity labels. When a user browses an APP in a shopping mall, browsed interested commodity objects cannot be visually obtained, and the user may browse commodities (such as a refrigerator, an air conditioner and a washing machine) with different commodity labels and finally buy the air conditioner, so that behavior data of the commodities with different commodity labels needs to be generated by combining the user, and the preference of the user is more accurately analyzed. For example, a user browses in commodities labeled by three commodities, namely a refrigerator, an air conditioner and a washing machine, through a user terminal, and based on the data statistics on the commodity behaviors, user behavior data generated by the commodities of the user under the three commodity labels, namely second user behavior data, is obtained, for example, the user views the commodities labeled by the commodity label as the refrigerator for 5 times, clicks 2 shares, joins 3 commodities, enters a shopping cart, and the like; and (3) checking the detailed page of the commodity for 1 time, clicking 1 sharing and 1 commodity with the commodity label collected for the air conditioner on the commodity with the commodity label as the air conditioner, and the like. It should be noted that, in the present invention, for the behavior type of the successful payment of the goods, the user behavior burial point is only used when calculating the tag heat values of different goods tags of the user.
102, acquiring a comprehensive heat value of each commodity of the same commodity label according to the first user behavior data and a time attenuation coefficient corresponding to the first user behavior data; and acquiring a label heat value of each commodity label according to the second user behavior data and a time attenuation coefficient corresponding to the second user behavior data, wherein the time attenuation coefficient is acquired based on the embedding point event creation time and the commodity type corresponding to the commodity browsing history record of the user terminal.
In the invention, based on the obtained user behavior data generated by the user on different commodities of the same commodity label, the comprehensive heat value of each commodity of the same commodity label is calculated according to the formula of the time attenuation coefficient and the comprehensive heat value. Specifically, the formula of the time attenuation coefficient is as follows:
H(t)＝exp[-a(t-tj)]；
wherein H (t) represents the time attenuation coefficient of the product at the t-th day, a represents the cooling coefficient, and tjRepresenting the creation time of the jth buried event.
In the invention, the time attenuation coefficient refers to that the preference value generated by the user behavior is attenuated along with the time according to a certain attenuation coefficient. The coefficient is an exponential function with e as base, tjRepresenting the creation time of the jth buried event, the present invention decays in days. a is the so-called "gravity factor" (i.e., cooling coefficient), i.e., the greater the value, the faster the heat of the product will fall. Specifically, when the user browses the item for a long time ago, the influence of the item browsed for a long time ago on recommendation gradually decreases and the influence on the heat value gradually decreases as time changes. Therefore, if a considerable period of time has elapsed since the last generation of the same user behavior for different behavior types, it is determined that the influence of the user behavior on the heat value gradually decreases according to the product browsing history of the user terminal.
The comprehensive heat value is obtained by calculation according to the commodity type, the user behavior weight value corresponding to the user behavior burying point, the time attenuation coefficient, the user behavior statistical frequency and the user behavior burying point establishing time, and the formula is as follows:
wherein Q isuiRepresents the integrated calorific value, W, of the i-th productijThe weight value of the user behavior corresponding to the jth embedded point event generated by the user terminal on the ith commodity is represented; a represents the cooling coefficient, the invention divides the commodities into household commodities and non-household commodities according to the commodity types, wherein, the invention takes the cooling coefficient of the household commodities as0.33, the value of the non-household goods is 0.7;
represents a time attenuation coefficient; t represents the current time in days; t is tjRepresents the creation time, N, of the jth buried eventijAnd the statistical times of the user behaviors corresponding to the jth buried point event generated by different user terminals on the ith commodity are shown.
In the invention, different users generate behavior times statistics of different behaviors (such as click-in, praise, collection, forwarding, comment and the like) on the commodities with the same label, and the behavior times statistics are derived from a buried event at the front end of a page.
On the basis of the above embodiment, the tag heat value is obtained by calculation according to the type of the commodity tag, the user behavior weight value corresponding to the user behavior burying point, the user behavior frequency, the time attenuation coefficient, and the user behavior burying point creation time, and the formula is as follows:
wherein Q isukIndicates a label heat value, W, of a k-th merchandise labelkjThe weight value of the user behavior corresponding to the jth embedded point event generated by the user terminal based on the kth commodity label is represented; n is a radical ofkjRepresenting the behavior times of the user behavior corresponding to the jth embedded point event generated in the kth commodity label; q. q.skA weighting coefficient indicating a kth type commodity label, for example, the present invention sets a weight coefficient of 1.1 for a buried point event in which the commodity label is a household type commodity; a represents the cooling coefficient of the molten steel,
represents a time attenuation coefficient; t represents the current time in days; t is tjRepresenting the creation time of the jth buried event.
And 103, generating a commodity recommendation list according to the comprehensive heat value and the label heat value.
In the invention, the commodities are sorted based on the comprehensive heat value; and sorting the different commodity labels based on the label heat value, and further generating a personalized commodity recommendation list for the user according to the two sorting results.
According to the commodity personalized recommendation method based on the user behaviors, the popularity value sequence of commodities under the commodity labels which are interested and preferred by the user is mastered according to the user behavior data, so that the commodities are recommended to different users, and the commodity pushing accuracy and the commodity pushing efficiency are improved.
On the basis of the above embodiment, the generating a product recommendation list according to the comprehensive heat value and the tag heat value includes:
sorting the comprehensive heat value, and obtaining a commodity sorting result based on each same commodity label according to the sorting result of the comprehensive heat value;
sorting the label heat value of each commodity label to obtain a commodity label heat value sorting result;
selecting target commodities from the commodity sorting results according to the commodity label heat value sorting results, and generating a commodity recommendation list according to the target commodities;
or determining pre-recommended commodities according to the commodity sorting result, and selecting a target commodity from the pre-recommended commodities according to the tag heat value sorting result to generate a commodity recommendation list.
Fig. 2 is a schematic diagram of a commodity recommendation process based on user behaviors and commodity labels according to the present invention, and as shown in fig. 2, a user behavior of all commodities already put on shelves in an APP mall used by a user terminal is dotted, and a commodity comprehensive heat value of the same commodity label is calculated for an operation behavior of the user on the commodity (entering a detail page, successfully collecting, canceling collection, clicking to share, confirming search success, confirming goods arrival notification confirmation success, confirming adding a shopping cart, clicking to immediately purchase, and successfully paying), so that commodity ordering of the same commodity label is performed according to the calculated comprehensive heat value.
Meanwhile, logic calculation is carried out on the user behavior buried points of all the commodities carrying different commodity labels, and the heat ranking calculation of each commodity label is carried out according to the preference label of the user, namely the ordering is carried out according to the label heat value of each commodity label. In the calculation of the heat, the attenuation coefficients of the household appliance and the non-household appliance are distinguished, and since the label of the household appliance type product is important in the entire recommendation calculation, the corresponding attenuation coefficient is small.
And further recommending hot commodities under the label according to the calculated preference degree of the user to a certain label and by combining the sequencing result of the comprehensive hot value of the commodities, and recommending the commodities with the hot value to the user according to the sequencing of the hot values from high to low. The method can also be used for generating pre-recommended commodities for a user according to the calculated preference degree of the user to a certain commodity, then combining the sorting results of the label heat value carried by the commodities, sorting the commodities with the heat value from high to low from the pre-recommended commodities according to the label heat value, for example, firstly obtaining the sorting results of the commodities with the multifunctional electric appliance labels according to the comprehensive heat value of the commodities, then sorting the commodities according to the commodity labels carried by the commodities, such as a refrigerator, a washing machine, an air conditioner and the like, through the label heat value, and recommending the commodities to the user according to the sequence from high to low. In the present invention, as shown in fig. 2, a recommendation list is generated based on the comprehensive popularity of the product and the popularity of the tag, and the product contents are sorted according to the recommendation list, so as to generate a current real-time recommended content, which can be used as a reference recommended content in a next time period, and can provide product recommendation more efficiently and accurately when a user performs subsequent product recommendation.
The commodity personalized recommendation method based on the user behaviors can be applied to various website platforms or online shopping application programs, can accurately master the interest preference labels of the users, positions commodities contained in related label types according to the labels, and recommends commodities of different users according to the popularity value sequence of the commodities under the labels, so that thousands of people are realized, the purchase requirements of the users are better met, and the users have better use experience. Meanwhile, algorithm design and calculation are carried out based on various comprehensive factors such as different specific gravity degrees and time attenuation coefficients brought by historical behaviors and different behaviors, and the accuracy of commodity pushing is improved. The invention pushes the commodities interested by the customer according to the preference of the user, improves the buying desire of the user and further can improve the economic benefit.
Fig. 3 is a schematic structural diagram of a commodity personalized recommendation system based on user behaviors, and as shown in fig. 3, the invention provides a commodity personalized recommendation system based on user behaviors, which includes a buried point event acquisition module 301, a heat value acquisition module 302 and a commodity recommendation module 303, where the buried point event acquisition module 301 is configured to acquire user behavior data, the user behavior data includes first user behavior data and second user behavior data, the first user behavior data is behavior data generated by a user terminal for each commodity of the same commodity label, and the second user behavior data is behavior data generated by the user terminal for commodities of different commodity labels; the heat value obtaining module 302 is configured to obtain a comprehensive heat value of each product of the same product label according to the first user behavior data and a time attenuation coefficient corresponding to the first user behavior data; acquiring a label heat value of each commodity label according to the second user behavior data and a time attenuation coefficient corresponding to the second user behavior data; the commodity recommending module 303 is configured to generate a commodity recommending list according to the comprehensive heat value and the label heat value.
According to the commodity personalized recommendation system based on the user behaviors, the popularity value sequence of commodities under the commodity labels which are interested and preferred by the user is mastered according to the user behavior data, so that the commodities are recommended to different users, and the commodity pushing accuracy and the commodity pushing efficiency are improved.
The system provided by the present invention is used for executing the above method embodiments, and for the specific processes and details, reference is made to the above embodiments, which are not described herein again.
Fig. 4 is a schematic structural diagram of an electronic device provided in the present invention, and as shown in fig. 4, the electronic device may include: a processor (processor)401, a communication interface (communication interface)402, a memory (memory)403 and a communication bus 404, wherein the processor 401, the communication interface 402 and the memory 403 complete communication with each other through the communication bus 404. The processor 401 may call logic instructions in the memory 403 to execute a method for personalized recommendation of goods based on user behavior, the method comprising: acquiring user behavior data, wherein the user behavior data comprises first user behavior data and second user behavior data, the first user behavior data is behavior data generated by a user terminal on each commodity of the same commodity label, and the second user behavior data is behavior data generated by the user terminal on commodities of different commodity labels; acquiring a comprehensive heat value of each commodity of the same commodity label according to the first user behavior data and a time attenuation coefficient corresponding to the first user behavior data; acquiring a label heat value of each commodity label according to the second user behavior data and a time attenuation coefficient corresponding to the second user behavior data; and generating a commodity recommendation list according to the comprehensive heat value and the label heat value.
In addition, the logic instructions in the memory 403 may be implemented in the form of software functional units and stored in a computer readable storage medium when the software functional units are sold or used as independent products. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: various media capable of storing program codes, such as a usb disk, a removable hard disk, a Read-only memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk.
In another aspect, the present invention further provides a computer program product, which includes a computer program stored on a non-transitory computer-readable storage medium, the computer program including program instructions, when the program instructions are executed by a computer, the computer being capable of executing the commodity personalized recommendation method based on user behavior provided by the above methods, the method including: acquiring user behavior data, wherein the user behavior data comprises first user behavior data and second user behavior data, the first user behavior data is behavior data generated by a user terminal on each commodity of the same commodity label, and the second user behavior data is behavior data generated by the user terminal on commodities of different commodity labels; acquiring a comprehensive heat value of each commodity of the same commodity label according to the first user behavior data and a time attenuation coefficient corresponding to the first user behavior data; acquiring a label heat value of each commodity label according to the second user behavior data and a time attenuation coefficient corresponding to the second user behavior data; and generating a commodity recommendation list according to the comprehensive heat value and the label heat value.
In yet another aspect, the present invention further provides a non-transitory computer-readable storage medium, on which a computer program is stored, where the computer program is implemented by a processor to execute the method for personalized recommendation of goods based on user behavior provided in the foregoing embodiments, and the method includes: acquiring user behavior data, wherein the user behavior data comprises first user behavior data and second user behavior data, the first user behavior data is behavior data generated by a user terminal on each commodity of the same commodity label, and the second user behavior data is behavior data generated by the user terminal on commodities of different commodity labels; acquiring a comprehensive heat value of each commodity of the same commodity label according to the first user behavior data and a time attenuation coefficient corresponding to the first user behavior data; acquiring a label heat value of each commodity label according to the second user behavior data and a time attenuation coefficient corresponding to the second user behavior data; and generating a commodity recommendation list according to the comprehensive heat value and the label heat value.
The above-described embodiments of the apparatus are merely illustrative, and the units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment. One of ordinary skill in the art can understand and implement it without inventive effort.
Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware. With this understanding in mind, the above-described technical solutions may be embodied in the form of a software product, which can be stored in a computer-readable storage medium such as ROM/RAM, magnetic disk, optical disk, etc., and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the methods described in the embodiments or some parts of the embodiments.
Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (10)
1. A commodity personalized recommendation method based on user behaviors is characterized by comprising the following steps:
acquiring user behavior data, wherein the user behavior data comprises first user behavior data and second user behavior data, the first user behavior data is behavior data generated by a user terminal on each commodity of the same commodity label, and the second user behavior data is behavior data generated by the user terminal on commodities of different commodity labels;
acquiring a comprehensive heat value of each commodity of the same commodity label according to the first user behavior data and a time attenuation coefficient corresponding to the first user behavior data; acquiring a label heat value of each commodity label according to the second user behavior data and a time attenuation coefficient corresponding to the second user behavior data;
and generating a commodity recommendation list according to the comprehensive heat value and the label heat value.
2. The method for personalized recommendation of goods based on user behavior according to claim 1, wherein before said obtaining of user behavior data, said method further comprises:
setting a user behavior buried point and a user behavior type preset weight value based on a user behavior type so as to obtain user behavior data according to the user behavior buried point, wherein the user behavior buried point at least comprises entering a commodity detail page, successfully collecting commodities, canceling commodity collection, clicking commodity sharing, confirming commodity search success, confirming commodity arrival notification confirmation success, confirming that commodities are added into a shopping cart, clicking to purchase commodities immediately, clicking to submit commodity orders and successfully paying commodities.
3. The method for personalized recommendation of goods based on user behavior according to claim 2, wherein the obtaining of user behavior data comprises:
acquiring the times of user behaviors generated by the user terminal to the commodity, the types of the user behaviors and the weight of the user behaviors based on the user behavior buried points and the preset weight values of the user behavior types;
and obtaining user behavior data according to the user behavior times, the user behavior types and the user behavior weights.
4. The commodity personalized recommendation method based on user behaviors as claimed in claim 1, wherein the time attenuation coefficient is obtained based on a buried event creation time and a commodity type corresponding to a commodity browsing history of a user terminal.
5. The user behavior-based commodity personalized recommendation method according to claim 1, wherein the comprehensive heat value is calculated according to a commodity type, a user behavior weight value corresponding to a user behavior burying point, a time attenuation coefficient, a user behavior statistical number and a user behavior burying point creation time.
6. The user behavior-based commodity personalized recommendation method according to claim 1, wherein the tag heat value is calculated according to a commodity tag type, a user behavior weight value corresponding to a user behavior burying point, a user behavior frequency, a time attenuation coefficient and a user behavior burying point creation time.
7. The method for personalized recommendation of goods based on user behavior according to claim 1, wherein said generating of recommendation list of goods according to said comprehensive heat value and said tag heat value comprises:
sorting the comprehensive heat value, and obtaining a commodity sorting result based on each same commodity label according to the sorting result of the comprehensive heat value;
sorting the label heat value of each commodity label to obtain a commodity label heat value sorting result;
selecting target commodities from the commodity sorting results according to the commodity label heat value sorting results, and generating a commodity recommendation list according to the target commodities;
or determining pre-recommended commodities according to the commodity sorting result, and selecting a target commodity from the pre-recommended commodities according to the tag heat value sorting result to generate a commodity recommendation list.
8. A commodity personalized recommendation system based on user behaviors is characterized by comprising:
the system comprises a buried point event acquisition module, a data processing module and a data processing module, wherein the buried point event acquisition module is used for acquiring user behavior data, the user behavior data comprises first user behavior data and second user behavior data, the first user behavior data is behavior data generated by a user terminal on each commodity of the same commodity label, and the second user behavior data is behavior data generated by the user terminal on commodities of different commodity labels;
the heat value acquisition module is used for acquiring a comprehensive heat value of each commodity of the same commodity label according to the first user behavior data and a time attenuation coefficient corresponding to the first user behavior data; acquiring a label heat value of each commodity label according to the second user behavior data and a time attenuation coefficient corresponding to the second user behavior data;
and the commodity recommendation module is used for generating a commodity recommendation list according to the comprehensive heat value and the label heat value.
9. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the steps of the method for personalized recommendation of goods based on user behavior according to any one of claims 1 to 7 when executing the computer program.
10. A non-transitory computer readable storage medium, on which a computer program is stored, wherein the computer program, when being executed by a processor, implements the steps of the method for personalized recommendation of goods based on user behavior according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110908381.3A CN113689259A (en) | 2021-08-09 | 2021-08-09 | Commodity personalized recommendation method and system based on user behaviors |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110908381.3A CN113689259A (en) | 2021-08-09 | 2021-08-09 | Commodity personalized recommendation method and system based on user behaviors |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113689259A true CN113689259A (en) | 2021-11-23 |
Family
ID=78579202
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110908381.3A Pending CN113689259A (en) | 2021-08-09 | 2021-08-09 | Commodity personalized recommendation method and system based on user behaviors |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113689259A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114820142A (en) * | 2022-06-29 | 2022-07-29 | 国能(北京)商务网络有限公司 | Commodity information recommendation method facing to B-end purchasing user |
| CN116227808A (en) * | 2022-12-08 | 2023-06-06 | 艾米丁(杭州)互联网科技有限公司 | Intelligent operation management control system for laundry |
| CN116546091A (en) * | 2023-07-07 | 2023-08-04 | 深圳市四格互联信息技术有限公司 | Recommendation method, device, equipment and storage medium of streaming content |
| CN117114823A (en) * | 2023-10-25 | 2023-11-24 | 长沙识达科技有限公司 | Internet-based data mining method and system |
-
2021
- 2021-08-09 CN CN202110908381.3A patent/CN113689259A/en active Pending
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114820142A (en) * | 2022-06-29 | 2022-07-29 | 国能(北京)商务网络有限公司 | Commodity information recommendation method facing to B-end purchasing user |
| CN114820142B (en) * | 2022-06-29 | 2022-09-16 | 国能(北京)商务网络有限公司 | Commodity information recommendation method for B-side purchasing user |
| CN116227808A (en) * | 2022-12-08 | 2023-06-06 | 艾米丁(杭州)互联网科技有限公司 | Intelligent operation management control system for laundry |
| CN116546091A (en) * | 2023-07-07 | 2023-08-04 | 深圳市四格互联信息技术有限公司 | Recommendation method, device, equipment and storage medium of streaming content |
| CN116546091B (en) * | 2023-07-07 | 2023-11-28 | 深圳市四格互联信息技术有限公司 | Recommendation method, device, equipment and storage medium of streaming content |
| CN117114823A (en) * | 2023-10-25 | 2023-11-24 | 长沙识达科技有限公司 | Internet-based data mining method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103164804B (en) | The information-pushing method of a kind of personalization and device | |
| CN113689259A (en) | Commodity personalized recommendation method and system based on user behaviors | |
| CN110569432B (en) | Commodity sequence calculating method, commodity sequence calculating device, computer equipment and storage medium | |
| CN112837118B (en) | Commodity recommendation method and device for enterprise users | |
| CN106469382A (en) | Idle merchandise items information processing method and device | |
| CN106708821A (en) | User personalized shopping behavior-based commodity recommendation method | |
| JP2019512809A (en) | Method and apparatus for time division recommendation for service target | |
| JPWO2016157424A1 (en) | Information processing apparatus, information processing method, and information processing program | |
| CN111738805B (en) | Behavior log-based search recommendation model generation method, device and storage medium | |
| CN112435067A (en) | Intelligent advertisement putting method and system for cross-e-commerce platform and social platform | |
| JP6976207B2 (en) | Information processing equipment, information processing methods, and programs | |
| CN111899047A (en) | Resource recommendation method and device, computer equipment and computer-readable storage medium | |
| JP5824602B1 (en) | Information processing apparatus, information processing method, and information processing program | |
| CN116894709A (en) | Advertisement commodity recommendation method and system | |
| CN111914188A (en) | Method, system, device and storage medium for selecting recommendation target user | |
| CN114581175A (en) | Commodity pushing method and device, storage medium and electronic equipment | |
| CN111767459A (en) | Item recommendation method and device | |
| Aprilianti et al. | Implementation of weighted parallel hybrid recommender systems for e-commerce in indonesia | |
| CN114880560A (en) | Content recommendation method and device, storage medium and electronic device | |
| CN111429214A (en) | Transaction data-based buyer and seller matching method and device | |
| CN110570271A (en) | information recommendation method and device, electronic equipment and readable storage medium | |
| CN110751537A (en) | Online and offline intelligent marketing method and device | |
| CN112132660B (en) | Commodity recommendation method, system, equipment and storage medium | |
| JP2020013447A (en) | Determination device, determination method, and determination program | |
| CN111027351A (en) | Offline commodity recommendation method and device and electronic equipment |
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 |