CN116629973A - Commodity recommendation method and system based on neural network - Google Patents

Commodity recommendation method and system based on neural network Download PDF

Info

Publication number
CN116629973A
CN116629973A CN202310647342.1A CN202310647342A CN116629973A CN 116629973 A CN116629973 A CN 116629973A CN 202310647342 A CN202310647342 A CN 202310647342A CN 116629973 A CN116629973 A CN 116629973A
Authority
CN
China
Prior art keywords
commodity
user
neural network
information
purchase
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
Application number
CN202310647342.1A
Other languages
Chinese (zh)
Inventor
岳国庆
凤鹏飞
张继山
郑春
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Anhui Sanlian University
Original Assignee
Anhui Sanlian University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Anhui Sanlian University filed Critical Anhui Sanlian University
Priority to CN202310647342.1A priority Critical patent/CN116629973A/en
Publication of CN116629973A publication Critical patent/CN116629973A/en
Pending legal-status Critical Current

Links

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
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • 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
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising
    • G06Q30/0241—Advertisements
    • G06Q30/0251—Targeted advertisements
    • G06Q30/0255—Targeted advertisements based on user history
    • 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
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising
    • G06Q30/0241—Advertisements
    • G06Q30/0251—Targeted advertisements
    • G06Q30/0264—Targeted advertisements based upon schedule

Landscapes

  • Business, Economics & Management (AREA)
  • Engineering & Computer Science (AREA)
  • Accounting & Taxation (AREA)
  • Finance (AREA)
  • Strategic Management (AREA)
  • Physics & Mathematics (AREA)
  • Development Economics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Marketing (AREA)
  • Economics (AREA)
  • General Business, Economics & Management (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Game Theory and Decision Science (AREA)
  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Biomedical Technology (AREA)
  • Biophysics (AREA)
  • Computational Linguistics (AREA)
  • Data Mining & Analysis (AREA)
  • Evolutionary Computation (AREA)
  • General Health & Medical Sciences (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The application is applicable to the technical field of commodity recommendation, and particularly relates to a commodity recommendation method and system based on a neural network, wherein the method comprises the following steps: acquiring commodity purchase data; counting commodities purchased by each user and commodity types according to the user information, and screening to obtain individual users; acquiring commodity purchase data corresponding to all individual users, analyzing each commodity based on a neural network, and determining the service cycle of each commodity; and acquiring historical order information of the user to be recommended, generating a commodity pushing frequency scheme, and executing. According to the commodity recommending method based on the neural network, the attribute of the commodity is analyzed through the neural network, so that the type of the commodity is determined, and then the purchase frequency or the purchase interval of each commodity is determined, so that pushing is reduced or avoided in the time of the purchase interval, the effect of avoiding overlapping and useless pushing is achieved, and the effectiveness of commodity pushing is improved.

Description

Commodity recommendation method and system based on neural network
Technical Field
The application belongs to the technical field of commodity recommendation, and particularly relates to a commodity recommendation method and system based on a neural network.
Background
The artificial neural network is an algorithm mathematical model which simulates the behavior characteristics of the animal neural network and performs distributed parallel information processing. The network relies on the complexity of the system and achieves the purpose of processing information by adjusting the relationship of the interconnection among a large number of nodes.
The artificial neural network is a technical reproduction of the biological neural network in a certain simplified sense, and is mainly used for constructing a practical artificial neural network model according to the principle of the biological neural network and the actual application requirement, designing a corresponding learning algorithm, simulating certain intelligent activities of the human brain and then realizing the artificial neural network in the technical way so as to solve the actual problem.
In the current commodity recommendation process, the commodity recommendation is mainly performed according to the purchasing and browsing conditions of commodities, the intelligent degree is high, and the problem of repeated recommendation easily occurs.
Disclosure of Invention
The embodiment of the application aims to provide a commodity recommending method based on a neural network, and aims to solve the problems that the prior art recommends according to the purchasing and browsing conditions of commodities, the intelligent degree is too high, and repeated recommends are easy to occur.
The embodiment of the application is realized in such a way that the commodity recommending method based on the neural network comprises the following steps:
acquiring commodity purchase data, wherein the commodity purchase data comprises purchase user information and purchase time information;
counting commodities purchased by each user and commodity types according to the user information, and screening to obtain individual users;
acquiring commodity purchase data corresponding to all individual users, analyzing each commodity based on a neural network, and determining the service cycle of each commodity;
and acquiring historical order information of the user to be recommended, generating a commodity pushing frequency scheme, and executing.
Preferably, the step of counting the commodities purchased by each user and the commodity types according to the user information and screening to obtain the individual user specifically includes:
user information is called one by one, and commodity order data in a preset time period are obtained according to the user information;
removing users with the quantity of the single order purchased commodities exceeding a first preset value to obtain primary screening users;
counting the use frequency of commodities purchased by the primary screening users, and eliminating the primary screening users with the use frequency exceeding a second preset value to obtain individual users.
Preferably, the step of collecting commodity purchase data corresponding to all individual users, analyzing each commodity based on the neural network, and determining the service cycle of each commodity specifically includes:
training the neural network through a preset order data set;
collecting commodity purchase data corresponding to all individual users, and extracting commodity models and order time contained in the commodity purchase data;
and taking the commodity model and the order time as input, and analyzing by using the trained neural network to determine the service cycle of each commodity.
Preferably, the step of obtaining historical order information of the user to be recommended, generating a commodity pushing frequency scheme, and executing specifically includes:
acquiring historical order information of a user to be recommended, and extracting commodity information;
inquiring the corresponding use period according to commodity information, and determining a period allowance;
and generating a commodity pushing frequency scheme according to the period allowance, and executing.
Preferably, when the period allowance is larger than a preset value, commodity pushing is not performed.
Preferably, the commodities are divided according to the use time and pushed according to the use time.
Another object of an embodiment of the present application is to provide a commodity recommendation system based on a neural network, the system including:
the data acquisition module is used for acquiring commodity purchase data, wherein the commodity purchase data comprises purchase user information and purchase time information;
the user screening module is used for counting commodities purchased by each user and commodity types according to the user information, and screening to obtain individual users;
the period analysis module is used for collecting commodity purchase data corresponding to all individual users, analyzing all commodities based on the neural network and determining the service period of each commodity;
and the commodity recommending module is used for acquiring historical order information of the user to be recommended, generating a commodity pushing frequency scheme and executing the commodity pushing frequency scheme.
Preferably, the user screening module includes:
the order extraction unit is used for calling the user information one by one and presetting commodity order data in a time period according to the user information;
the first screening unit is used for eliminating users with the quantity of the single order purchased commodities exceeding a first preset value to obtain primary screening users;
and the second screening unit is used for counting the use frequency of the commodities purchased by the primary screening users, eliminating the primary screening users with the use frequency exceeding a second preset value, and obtaining the individual users.
Preferably, the period analysis module includes:
the neural network training unit is used for training the neural network through a preset order data set;
the data extraction unit is used for collecting commodity purchase data corresponding to all individual users and extracting commodity models and order time contained in the commodity purchase data;
and the commodity analysis unit is used for taking commodity model and order time as input, analyzing by utilizing the trained neural network and determining the service cycle of each commodity.
Preferably, the commodity recommendation module includes:
the information extraction unit is used for acquiring historical order information of the user to be recommended and extracting commodity information;
the residual calculating unit is used for inquiring the corresponding use period according to the commodity information and determining the period residual;
and the commodity pushing unit is used for generating a commodity pushing frequency scheme according to the period allowance and executing the commodity pushing frequency scheme.
According to the commodity recommendation method based on the neural network, the attribute of the commodity is analyzed through the neural network, so that the type of the commodity is determined, and then the purchase frequency or the purchase interval of each commodity is determined, so that pushing is reduced or avoided in the time of the purchase interval, the effect of avoiding overlapping and useless pushing is achieved, and the effectiveness of commodity pushing is improved.
Drawings
Fig. 1 is a flowchart of a commodity recommendation method based on a neural network according to an embodiment of the present application;
FIG. 2 is a flowchart showing steps for counting commodities purchased by each user and commodity types according to user information and screening to obtain individual users according to the embodiment of the present application;
FIG. 3 is a flowchart of the steps for acquiring commodity purchase data corresponding to all individual users, analyzing each commodity based on a neural network, and determining the service cycle of each commodity according to the embodiment of the present application;
FIG. 4 is a flowchart of steps performed to obtain historical order information of a user to be recommended, generate a commodity pushing frequency scheme, and execute the method;
FIG. 5 is a schematic diagram of a commodity recommendation system based on a neural network according to an embodiment of the present application;
fig. 6 is a schematic diagram of a user screening module according to an embodiment of the present application;
FIG. 7 is a schematic diagram of a cycle analysis module according to an embodiment of the present application;
fig. 8 is a schematic diagram of a commodity recommendation module according to an embodiment of the present application.
Detailed Description
The present application will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present application more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the application.
The artificial neural network is a technical reproduction of the biological neural network in a certain simplified sense, and is mainly used for constructing a practical artificial neural network model according to the principle of the biological neural network and the actual application requirement, designing a corresponding learning algorithm, simulating certain intelligent activities of the human brain and then realizing the artificial neural network in the technical way so as to solve the actual problem. In the current commodity recommendation process, the commodity recommendation is mainly performed according to the purchasing and browsing conditions of commodities, the intelligent degree is high, and the problem of repeated recommendation easily occurs.
According to the application, the attribute of the commodity is analyzed through the neural network, so that the type of the commodity is determined, and then the purchase frequency or the purchase interval of each commodity is determined, so that pushing is reduced or avoided in the time of the purchase interval, the effects of avoiding overlapping and useless pushing are realized, and the effectiveness of commodity pushing is improved.
As shown in fig. 1, a flowchart of a commodity recommendation method based on a neural network according to an embodiment of the present application is provided, where the method includes:
s100, commodity purchase data is acquired, wherein the commodity purchase data comprises purchase user information and purchase time information.
In this step, commodity purchase data is obtained, all the historical data in the platform are called, or the historical data can be called according to a time period, for example, the current moment is taken as a reference, the time of year is traced back, all the generated historical data in the period are called, and commodity purchase data is obtained, wherein the commodity purchase data at least comprises user information and purchase time information, for example, the user information comprises the name of the user and a mobile phone number, and the purchase time information is the time when the product purchased by the user is signed.
And S200, counting commodities purchased by each user and commodity types according to the user information, and screening to obtain individual users.
In this step, the commodity purchased by each user and the commodity type are counted according to the user information, specifically, all the users are counted, the commodity situation purchased by each user in the preset time period and the corresponding commodity type are counted one by one, whether the commodity purchased by the user is used or not is determined according to the number of purchased commodities and the commodity use time, when the number of purchased commodities exceeds the preset value, the user is indicated to be possibly wholesale, therefore, the basis for analyzing the use time of each commodity by all the data of the user is unreliable, the user is removed, and finally, the individual user is obtained through screening.
And S300, acquiring commodity purchase data corresponding to all individual users, analyzing all commodities based on a neural network, and determining the service cycle of each commodity.
In this step, commodity purchase data corresponding to all individual users are collected, all commodities purchased by the individual users in a preset time period are counted, the corresponding commodity purchase data comprise purchase time, and therefore the neural network is trained by utilizing a preset data set, so that the neural network can analyze the service time of all the commodities to determine the service period of each commodity, the average service time of each commodity can be counted and used as the service period of the commodity, the mode in the service time can be used as the service period of the commodity, the labels can be set for the individual users according to the condition that the individual users purchase the commodity, and the service periods of the same commodity corresponding to different crowds can be determined according to the distribution condition of the labels, for example, the A user and the B user have corresponding labels, the service periods of the same product are regarded as the same, when the labels are determined, the labels are generated according to the types of the purchased commodities, particularly, the labels are generated according to the attributes of the commodities, for example, the child labels are generated according to the sizes of mother and infant products, and the clothes, and the like, and the users are classified according to the labels.
S400, acquiring historical order information of a user to be recommended, generating a commodity pushing frequency scheme, and executing.
In the step, historical order information of a user to be recommended is obtained, order information of the user needing commodity pushing is obtained, the time of the user receiving the commodity can be determined according to the historical order information, so that the time of the user buying the commodity next time is estimated according to the time of receiving the commodity and the service cycle of the commodity, the pushing frequency is gradually increased before the time of buying the commodity next time is reached, for example, the first last pushing is 5 times, the second last pushing is 4 times, the third last pushing is 3 times, and the like, pushing is not carried out when the time from buying the commodity next time is more than 5 days, and therefore invalid pushing is avoided, and the pushing effectiveness is influenced.
As shown in fig. 2, as a preferred embodiment of the present application, the step of counting the commodities purchased by each user and the commodity types according to the user information and screening to obtain the individual user specifically includes:
s201, calling the user information one by one, and presetting commodity order data in a time period according to the user information.
In this step, user information is called one by one, when user information is determined, a user corresponding to the same mobile phone number or the same real name information is regarded as a user, all order data corresponding to the user are combined, for example, a certain user generates a plurality of orders through two mobile phone numbers respectively, and the plurality of orders are regarded as data of the user, so that the orders are screened according to time, and commodity order data in a corresponding time period are obtained.
S202, removing users with the quantity of the single order purchased commodities exceeding a first preset value, and obtaining primary screening users.
In this step, users who purchase goods in a single order with a quantity exceeding a first preset value are removed, and for users who purchase goods in a single order with a quantity exceeding a preset value, the purpose of purchasing the products is not to use the products themselves, so that the users cannot be used for analyzing the use time of the goods.
And S203, counting the use frequency of the commodities purchased by the primary screening users, and eliminating the primary screening users with the use frequency exceeding a second preset value to obtain individual users.
In this step, the frequency of use of the commodity purchased by the primary screening user is counted to determine the frequency of use of a certain commodity by the user, for example, the single use of a certain commodity is 5ml, the total capacity is 100ml, the specification states that the single use of a single day is not more than 10ml, if the user finishes using and re-purchases within three days, the specification does not use the commodity according to the specification, and therefore the commodity cannot be used as a basis for analyzing the use time of the commodity, and the rest can be regarded as a personal user through twice screening.
As shown in fig. 3, as a preferred embodiment of the present application, the step of collecting commodity purchase data corresponding to all individual users, analyzing each commodity based on a neural network, and determining a service period of each commodity specifically includes:
s301, training the neural network through a preset order data set.
In this step, training is performed using a preset order data set, in which input data and result data of training are included, and the input data is input into the neural network model to complete training.
S302, commodity purchase data corresponding to all individual users are collected, and commodity models and order time contained in the commodity purchase data are extracted.
In this step, commodity purchase data corresponding to all individual users are collected, and in order to determine the use time of commodities, commodity models and order time contained in the commodity purchase data need to be extracted, and the time of using commodities by each user can be estimated according to the order time.
S303, taking the commodity model and the order time as input, and analyzing by using the trained neural network to determine the service cycle of each commodity.
In the step, the commodity model and the order time are used as input, and the trained neural network is utilized to analyze so as to determine the distribution condition of the use time corresponding to each commodity model, thereby determining one use time as the use period of the commodity, and accordingly determining the use period of all commodities.
As shown in fig. 4, as a preferred embodiment of the present application, the steps of obtaining historical order information of a user to be recommended, generating a commodity pushing frequency scheme, and executing specifically include:
s401, acquiring historical order information of a user to be recommended, and extracting commodity information.
In the step, historical order information of the user to be recommended is obtained, the model of the commodity purchased by the user and the corresponding purchase time are recorded in the historical order information of the user, and the use time of the commodity is determined according to the time signed by the user.
S402, inquiring the corresponding use period according to commodity information, and determining the period allowance.
S403, generating a commodity pushing frequency scheme according to the period allowance, and executing the commodity pushing frequency scheme.
In the step, inquiring the corresponding service period according to commodity information, determining the period allowance, if the service period of a certain commodity is 10 days, if the user signs for 7 days, determining the pushing times of each day in the rest 7 days, and pushing according to the pushing times; the commodities are divided according to the use time, and pushed according to the use time.
As shown in fig. 5, a commodity recommendation system based on a neural network according to an embodiment of the present application includes:
the data acquisition module 100 is configured to acquire commodity purchase data, where the commodity purchase data includes purchase user information and purchase time information.
In the system, the data acquisition module 100 acquires commodity purchase data, and invokes all the historical data in the platform according to a time period, for example, the current moment is taken as a reference, and the historical data generated in the period is traced back for one year, so that commodity purchase data is obtained, the commodity purchase data at least comprises user information and purchase time information, for example, the user information comprises the name of the user and a mobile phone number, and the purchase time information is the time when the product purchased by the user is signed.
And the user screening module 200 is used for counting the commodities purchased by each user and the commodity types according to the user information, and screening to obtain the individual users.
In the system, the user screening module 200 counts the commodities purchased by each user and the commodity types according to the user information, specifically counts all the users, counts the commodity conditions purchased by each user in a preset time period and the corresponding commodity types one by one, determines whether the commodities purchased by the user are used by one according to the number of purchased commodities and the commodity use time, and indicates that the user is likely to be wholesale when the purchased number exceeds a preset value, so that the basis of analyzing the use time of each commodity by all the data is unreliable, rejects the commodities and finally screens to obtain the individual user.
The period analysis module 300 is configured to collect commodity purchase data corresponding to all individual users, analyze each commodity based on the neural network, and determine a service period of each commodity.
In the system, the period analysis module 300 collects commodity purchase data corresponding to all individual users, specifically counts all commodities purchased by the individual users in a preset time period, the corresponding commodity purchase data comprises purchase time, and accordingly the neural network is trained by using a preset data set, so that the neural network can analyze the service time of all the commodities to determine the service period of each commodity, specifically counts the average service time of each commodity, takes the average service time as the service period of the commodity, takes the mode in the service time as the service period of the commodity, sets labels for the individual users according to the condition that the individual users purchase the commodity, determines the service periods of the same commodity corresponding to different groups according to label distribution conditions, if the A user and the B user have corresponding labels, the service periods of the same product are regarded as the same, generates specific labels according to the types of the purchased commodities, generates corresponding labels according to the attributes of the commodities, such as mother and infant clothes, and the like, and classifies the users according to the label distribution condition.
The commodity recommendation module 400 is configured to obtain historical order information of a user to be recommended, generate a commodity pushing frequency scheme, and execute the commodity pushing frequency scheme.
In the system, the commodity recommendation module 400 acquires historical order information of a user to be recommended, acquires order information of the user needing commodity pushing, and can determine the time when the user receives the commodity according to the historical order information, so that the time when the user purchases the commodity next time is estimated according to the time when the commodity is received and the service cycle of the commodity, the pushing frequency is gradually increased before the time when the commodity is purchased next time, such as 5 times on the first last day, 4 times on the second last day, 3 times on the third last day, and the like, and pushing is not performed when the time from the next time when the commodity is purchased is more than 5 days, so that invalid pushing is avoided, and pushing effectiveness is influenced.
As shown in fig. 6, as a preferred embodiment of the present application, the user screening module 200 includes:
the order extraction unit 201 is configured to call the user information one by one, and preset merchandise order data in a time period according to the user information.
In this module, the order extraction unit 201 invokes the user information one by one, when determining the user information, considers the user corresponding to the same mobile phone number or the same real name information as one user, merges all the corresponding order data, for example, a certain user generates a plurality of orders through two mobile phone numbers respectively, and considers the plurality of orders as the data of the user, thereby screening the orders according to time to obtain commodity order data in a corresponding time period.
The first screening unit 202 is configured to reject users with the number of commodities purchased in a single order exceeding a first preset value, and obtain primary screening users.
In this module, the first filtering unit 202 eliminates users who purchase goods with a single order with a quantity exceeding the first preset value, and for users who purchase goods with a single order with a quantity exceeding the preset value, the purpose of purchasing the products is not to use the products themselves, so that the users cannot be used for analyzing the time of use of the goods.
And the second screening unit 203 is configured to count the usage frequency of the commodities purchased by the primary screening user, and reject the primary screening user whose usage frequency exceeds a second preset value, so as to obtain the individual user.
In this module, the second screening unit 203 counts the frequency of use of the commodities purchased by the primary screening user to determine the frequency of use of a certain commodity by the user, for example, the single use of a certain commodity is 5ml, the total capacity is 100ml, the single use per day is described in the specification as not more than 10ml, if the user finishes using and re-purchases within three days, the user is described as not using according to the specification, and therefore the user cannot be used as a basis for analyzing the use time of the commodity, and the remaining user can be regarded as an individual user through two screening.
As shown in fig. 7, as a preferred embodiment of the present application, the period analysis module 300 includes:
the neural network training unit 301 is configured to train the neural network through a preset order data set.
In this module, the neural network training unit 301 performs training using a preset order data set, in which input data and result data of training are included, and inputs the input data into the neural network model to complete the training.
The data extraction unit 302 is configured to collect commodity purchase data corresponding to all individual users, and extract commodity models and order times contained therein.
In this module, the data extraction unit 302 collects commodity purchase data corresponding to all individual users, and in order to determine the use time of the commodity, the commodity model and the order time contained in the commodity purchase data need to be extracted, and the time of using the commodity by each user can be estimated according to the order time.
And the commodity analysis unit 303 is used for taking the commodity model and the order time as input, analyzing by using the trained neural network, and determining the service cycle of each commodity.
In this module, the commodity analysis unit 303 takes the commodity model and the order time as inputs, and performs analysis by using the trained neural network to determine the usage time distribution condition corresponding to each commodity model, thereby determining a usage time as the usage period of the commodity, and determining the usage periods of all commodities according to the usage time.
As shown in fig. 8, as a preferred embodiment of the present application, the commodity recommendation module 400 includes:
the information extraction unit 401 is configured to obtain historical order information of a user to be recommended, and extract commodity information.
In this module, the information extraction unit 401 acquires the user history order information to be recommended, in which the model number of the commodity purchased by the user and the corresponding purchase time are recorded, and determines the use time of the commodity according to the time signed by the user.
The remaining amount calculating unit 402 is configured to query a corresponding usage period according to the commodity information, and determine a period remaining amount.
And a commodity pushing unit 403, configured to generate a commodity pushing frequency scheme according to the period allowance, and execute the commodity pushing frequency scheme.
In the module, corresponding service periods are inquired according to commodity information, and a period allowance is determined, if the service period of a certain commodity is 10 days, and if a user signs for 7 days, the period allowance is 7 days, so that the number of pushing times per day in the rest 7 days is determined, and pushing is carried out according to the number of pushing times; the commodities are divided according to the use time, and pushed according to the use time.
It should be understood that, although the steps in the flowcharts of the embodiments of the present application are shown in order as indicated by the arrows, these steps are not necessarily performed in order as indicated by the arrows. The steps are not strictly limited to the order of execution unless explicitly recited herein, and the steps may be executed in other orders. Moreover, at least some of the steps in various embodiments may include multiple sub-steps or stages that are not necessarily performed at the same time, but may be performed at different times, nor do the order in which the sub-steps or stages are performed necessarily performed in sequence, but may be performed alternately or alternately with at least a portion of the sub-steps or stages of other steps or other steps.
Those skilled in the art will appreciate that all or part of the processes in the methods of the above embodiments may be implemented by a computer program for instructing relevant hardware, where the program may be stored in a non-volatile computer readable storage medium, and where the program, when executed, may include processes in the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in embodiments provided herein may include non-volatile and/or volatile memory. The nonvolatile 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), memory bus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), among others.
The technical features of the above-described embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above-described embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.
The foregoing examples illustrate only a few embodiments of the application and are described in detail herein without thereby limiting the scope of the application. It should be noted that it will be apparent to those skilled in the art that several variations and modifications can be made without departing from the spirit of the application, which are all within the scope of the application. Accordingly, the scope of protection of the present application is to be determined by the appended claims.
The foregoing description of the preferred embodiments of the application is not intended to be limiting, but rather is intended to cover all modifications, equivalents, and alternatives falling within the spirit and principles of the application.

Claims (10)

1. A neural network-based commodity recommendation method, the method comprising:
acquiring commodity purchase data, wherein the commodity purchase data comprises purchase user information and purchase time information;
counting commodities purchased by each user and commodity types according to the user information, and screening to obtain individual users;
acquiring commodity purchase data corresponding to all individual users, analyzing each commodity based on a neural network, and determining the service cycle of each commodity;
and acquiring historical order information of the user to be recommended, generating a commodity pushing frequency scheme, and executing.
2. The neural network-based commodity recommendation method according to claim 1, wherein said step of counting commodities purchased by each user and commodity types according to user information, and screening to obtain individual users specifically comprises:
user information is called one by one, and commodity order data in a preset time period are obtained according to the user information;
removing users with the quantity of the single order purchased commodities exceeding a first preset value to obtain primary screening users;
counting the use frequency of commodities purchased by the primary screening users, and eliminating the primary screening users with the use frequency exceeding a second preset value to obtain individual users.
3. The neural network-based commodity recommendation method according to claim 1, wherein the step of collecting commodity purchase data corresponding to all individual users, analyzing each commodity based on the neural network, and determining a service cycle of each commodity specifically comprises:
training the neural network through a preset order data set;
collecting commodity purchase data corresponding to all individual users, and extracting commodity models and order time contained in the commodity purchase data;
and taking the commodity model and the order time as input, and analyzing by using the trained neural network to determine the service cycle of each commodity.
4. The method for recommending commodities based on a neural network according to claim 1, wherein the steps of obtaining historical order information of a user to be recommended, generating a commodity pushing frequency scheme, and executing specifically include:
acquiring historical order information of a user to be recommended, and extracting commodity information;
inquiring the corresponding use period according to commodity information, and determining a period allowance;
and generating a commodity pushing frequency scheme according to the period allowance, and executing.
5. The neural network-based commodity recommendation method according to claim 4, wherein commodity pushing is not performed when the period margin is greater than a preset value.
6. The neural network-based commodity recommendation method according to claim 4, wherein commodities are divided according to usage time and pushed according to the usage time.
7. A neural network-based commodity recommendation system, the system comprising:
the data acquisition module is used for acquiring commodity purchase data, wherein the commodity purchase data comprises purchase user information and purchase time information;
the user screening module is used for counting commodities purchased by each user and commodity types according to the user information, and screening to obtain individual users;
the period analysis module is used for collecting commodity purchase data corresponding to all individual users, analyzing all commodities based on the neural network and determining the service period of each commodity;
and the commodity recommending module is used for acquiring historical order information of the user to be recommended, generating a commodity pushing frequency scheme and executing the commodity pushing frequency scheme.
8. The neural network-based commodity recommendation system according to claim 7, wherein said user screening module comprises:
the order extraction unit is used for calling the user information one by one and presetting commodity order data in a time period according to the user information;
the first screening unit is used for eliminating users with the quantity of the single order purchased commodities exceeding a first preset value to obtain primary screening users;
and the second screening unit is used for counting the use frequency of the commodities purchased by the primary screening users, eliminating the primary screening users with the use frequency exceeding a second preset value, and obtaining the individual users.
9. The neural network-based commodity recommendation system according to claim 7, wherein said period analysis module comprises:
the neural network training unit is used for training the neural network through a preset order data set;
the data extraction unit is used for collecting commodity purchase data corresponding to all individual users and extracting commodity models and order time contained in the commodity purchase data;
and the commodity analysis unit is used for taking commodity model and order time as input, analyzing by utilizing the trained neural network and determining the service cycle of each commodity.
10. The neural network-based commodity recommendation system according to claim 7, wherein said commodity recommendation module comprises:
the information extraction unit is used for acquiring historical order information of the user to be recommended and extracting commodity information;
the residual calculating unit is used for inquiring the corresponding use period according to the commodity information and determining the period residual;
and the commodity pushing unit is used for generating a commodity pushing frequency scheme according to the period allowance and executing the commodity pushing frequency scheme.
CN202310647342.1A 2023-06-02 2023-06-02 Commodity recommendation method and system based on neural network Pending CN116629973A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202310647342.1A CN116629973A (en) 2023-06-02 2023-06-02 Commodity recommendation method and system based on neural network

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202310647342.1A CN116629973A (en) 2023-06-02 2023-06-02 Commodity recommendation method and system based on neural network

Publications (1)

Publication Number Publication Date
CN116629973A true CN116629973A (en) 2023-08-22

Family

ID=87636455

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202310647342.1A Pending CN116629973A (en) 2023-06-02 2023-06-02 Commodity recommendation method and system based on neural network

Country Status (1)

Country Link
CN (1) CN116629973A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117557318A (en) * 2023-12-29 2024-02-13 青岛巨商汇网络科技有限公司 Management intelligent analysis method and system based on virtual shopping

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117557318A (en) * 2023-12-29 2024-02-13 青岛巨商汇网络科技有限公司 Management intelligent analysis method and system based on virtual shopping
CN117557318B (en) * 2023-12-29 2024-06-11 青岛巨商汇网络科技有限公司 Management intelligent analysis method and system based on virtual shopping

Similar Documents

Publication Publication Date Title
CN109598095B (en) Method and device for establishing scoring card model, computer equipment and storage medium
CN107016026B (en) User tag determination method, information push method, user tag determination device, information push device
CN109118296A (en) Movable method for pushing, device and electronic equipment
CN102629360B (en) A kind of effective dynamic commodity recommend method and commercial product recommending system
CN107689008A (en) A kind of user insures the method and device of behavior prediction
CN109300039A (en) The method and system of intellectual product recommendation are carried out based on artificial intelligence and big data
JP7391190B2 (en) Generating training data for machine learning models
CN104850567A (en) Method and device for identifying association between network users
CN110097394A (en) The latent objective recommended method of product and device
CN111597348A (en) User image drawing method, device, computer equipment and storage medium
CN116629973A (en) Commodity recommendation method and system based on neural network
CN108197795A (en) The account recognition methods of malice group, device, terminal and storage medium
CN111986027A (en) Abnormal transaction processing method and device based on artificial intelligence
CN112861980A (en) Calendar task table mining method based on big data and computer equipment
CN108776857A (en) NPS short messages method of investigation and study, system, computer equipment and storage medium
CN110704706A (en) Training method and classification method of classification model, related equipment and classification system
CN112991079A (en) Multi-card co-occurrence hospitalizing fraud detection method, system, cloud and medium
CN113656692A (en) Product recommendation method, device, equipment and medium based on knowledge migration algorithm
CN104484329B (en) Consumption hot spot method for tracing and device based on comment centre word timing variations analysis
CN113065892A (en) Information pushing method, device, equipment and storage medium
CN113536131A (en) Data processing method and device, storage medium and electronic equipment
CN108241989A (en) Otherness data capture method and device
CN113138977A (en) Transaction conversion analysis method, device, equipment and storage medium
US20200211037A1 (en) System and method for product demand transfer estimation through machine learning
CN114756731A (en) Advertisement channel data processing method and device, storage medium 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