CN111080398A - Commodity recommendation method and device, computer equipment and storage medium - Google Patents

Commodity recommendation method and device, computer equipment and storage medium Download PDF

Info

Publication number
CN111080398A
CN111080398A CN201911135202.6A CN201911135202A CN111080398A CN 111080398 A CN111080398 A CN 111080398A CN 201911135202 A CN201911135202 A CN 201911135202A CN 111080398 A CN111080398 A CN 111080398A
Authority
CN
China
Prior art keywords
commodity
historical
frequency
low
unviewed
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.)
Granted
Application number
CN201911135202.6A
Other languages
Chinese (zh)
Other versions
CN111080398B (en
Inventor
邱子轩
钱文杰
林方舟
王佳丽
俞冰
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.)
Zhejiang Dasou Vehicle Software Technology Co Ltd
Original Assignee
Zhejiang Dasou Vehicle Software Technology Co Ltd
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 Zhejiang Dasou Vehicle Software Technology Co Ltd filed Critical Zhejiang Dasou Vehicle Software Technology Co Ltd
Priority to CN201911135202.6A priority Critical patent/CN111080398B/en
Publication of CN111080398A publication Critical patent/CN111080398A/en
Application granted granted Critical
Publication of CN111080398B publication Critical patent/CN111080398B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

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
    • 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
    • G06Q30/0201—Market modelling; Market analysis; Collecting market data
    • 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
    • G06Q30/0201—Market modelling; Market analysis; Collecting market data
    • G06Q30/0203—Market surveys; Market polls
    • 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

Abstract

The application relates to a commodity recommendation method, a commodity recommendation device, computer equipment and a storage medium. The method comprises the following steps: receiving a recommendation instruction sent by a terminal, wherein the recommendation instruction carries a low-frequency commodity identifier and a current user identifier; acquiring a historical browsing record corresponding to the low-frequency commodity identification, wherein the historical browsing record comprises a historical user identification and historical low-frequency commodity information; calculating a weight coefficient of the unviewed commodity information and the historical low-frequency commodity information in the database according to the historical browsing frequency of the historical browsing record, and calculating the similarity of the unviewed commodity information and the historical low-frequency commodity information according to the weight coefficient; on the basis of the similarity, counting the interest degree value of the unviewed commodity information corresponding to the current user identification; and sequencing the unviewed commodities of the unviewed commodity information according to the interest value, and recommending the unviewed commodities to the terminal according to a sequencing result. By adopting the method, the low-frequency commodity can be accurately recommended for the user.

Description

Commodity recommendation method and device, computer equipment and storage medium
Technical Field
The present application relates to the field of big data technologies, and in particular, to a method and an apparatus for recommending a commodity, a computer device, and a storage medium.
Background
Personalized recommendation systems are widely used to provide accurate suggestions to users, helping them to make product purchasing decisions efficiently. The most common recommendation method is to recommend products to the user that are likely to be purchased in the future based on the user's historical transaction information. However, in fact, in the field of automobile sales or house sales, such historical information is very limited, and the server cannot collect sufficiently effective information, and thus cannot provide a sufficiently valuable recommendation to the user.
Disclosure of Invention
In view of the above, it is necessary to provide a product recommendation method, device, computer device and storage medium capable of accurately recommending low-frequency products for users in view of the above technical problems.
A method of merchandise recommendation, the method comprising:
receiving a recommendation instruction sent by a terminal, wherein the recommendation instruction carries a low-frequency commodity identification and a current user identification;
acquiring a historical browsing record corresponding to the low-frequency commodity identification, wherein the historical browsing record comprises a historical user identification and historical low-frequency commodity information;
calculating a weight coefficient of unviewed commodity information and historical low-frequency commodity information in a database according to the historical browsing frequency of the historical browsing record, and calculating the similarity of the unviewed commodity information and the historical low-frequency commodity information according to the weight coefficient;
counting the interest degree value of the unviewed commodity information corresponding to the current user identification based on the similarity;
and sequencing the unviewed commodities of the unviewed commodity information according to the interest degree value, and recommending the unviewed commodities to the terminal according to a sequencing result.
In one embodiment, the calculating a weight coefficient between unviewed commodity information in a database and the historical low-frequency commodity information according to the historical browsing frequency of the historical browsing record, and calculating a similarity between the unviewed commodity information and the historical low-frequency commodity information according to the weight coefficient includes:
counting the number of labels of the commodity content labels overlapped with the historical low-frequency commodity information in the unviewed commodity information;
calculating a commodity weight coefficient corresponding to the historical low-frequency commodity information according to the number of the tags;
and calculating the similarity between the unviewed commodity information and the historical low-frequency commodity information according to the commodity weight coefficient, the historical user identification and the historical browsing frequency.
In one embodiment, the method for generating the item content tag includes:
acquiring historical low-frequency commodity information, wherein the historical low-frequency commodity information carries historical commodity evaluation;
analyzing the historical low-frequency commodity information through a semantic analysis text, and extracting emotion sentences carrying emotion information;
and aggregating the emotion sentences, and setting the aggregated emotion sentences as commodity content labels.
In one embodiment, the calculating the similarity between the unviewed commodity information and the historical low-frequency commodity information according to the commodity weight coefficient, the historical user identification and the historical browsing frequency includes:
according to the formula
Figure BDA0002279397460000021
Calculating the similarity between the unviewed commodity information and the historical low-frequency commodity information, wherein wijFor the similarity between the unviewed item i and the historical low-frequency item j, | N (i) | is the number of labels of the unviewed item i, and I (u) is the labelThe number of tags appearing in the full sample of merchandise, | n (i) | n (j) | is the number of tags existing in both the unviewed merchandise i and the historical low-frequency merchandise j.
In one embodiment, the obtaining of the historical browsing record corresponding to the low-frequency product identifier includes:
generating commodity browsing page information according to a commodity browsing request of a user;
extracting low-frequency commodity identification and commodity evaluation sentences corresponding to historical low-frequency commodities from the commodity browsing page information;
and aggregating commodity evaluation sentences according to the low-frequency commodity identification and the historical user identification to generate a historical browsing record.
In one embodiment, after recommending the unviewed goods to the terminal according to the sorting result, the method includes:
receiving a request for selecting browsed commodities fed back by a terminal;
generating a commodity selection record according to the commodity selection browsing request;
and calculating the similarity between the unviewed commodity information in the database and the historical low-frequency commodity information according to the commodity selection record and the historical browsing record.
An article recommendation device, the device comprising:
the instruction receiving module is used for receiving a recommendation instruction sent by the terminal, wherein the recommendation instruction carries a low-frequency commodity identifier and a current user identifier;
the record acquisition module is used for acquiring a historical browsing record corresponding to the low-frequency commodity identification, wherein the historical browsing record comprises a historical user identification and historical low-frequency commodity information;
the similarity calculation module is used for calculating the weight coefficient of the unviewed commodity information and the historical low-frequency commodity information in the database according to the historical browsing frequency of the historical browsing record and calculating the similarity of the unviewed commodity information and the historical low-frequency commodity information according to the weight coefficient;
the interest degree value calculation module is used for counting the interest degree value of the unviewed commodity information corresponding to the current user identification based on the similarity;
and the recommending module is used for sequencing the unviewed commodities of the unviewed commodity information according to the interest degree value and recommending the unviewed commodities to the terminal according to a sequencing result.
In one embodiment, the similarity calculation module includes:
the label counting unit is used for counting the number of labels of the commodity content labels overlapped with the history low-frequency commodity information in the unviewed commodity information;
the weight coefficient generating unit is used for calculating a commodity weight coefficient corresponding to the historical low-frequency commodity information according to the number of the labels;
and the similarity calculation unit is used for calculating the similarity between the unviewed commodity information and the historical low-frequency commodity information according to the commodity weight coefficient, the historical user identification and the historical browsing frequency.
A computer device comprising a memory storing a computer program and a processor implementing the steps of the above method when executing the computer program.
A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the above-mentioned method.
According to the commodity recommendation method, the commodity recommendation device, the computer equipment and the storage medium, the historical browsing record corresponding to the low-frequency commodity identification is obtained; calculating a weight coefficient of the unviewed commodity information and the historical low-frequency commodity information in the database according to the historical browsing frequency of the historical browsing record, and calculating the similarity of the unviewed commodity information and the historical low-frequency commodity information according to the weight coefficient; generating an interest degree value of the unviewed commodity information corresponding to the current user identification; according to the interestingness value, ordering the unviewed commodities of the unviewed commodity information, recommending the unviewed commodities to the terminal according to an ordering result, determining the distribution condition of the content tags through historical browsing frequency, and adopting the historical browsing frequency to replace the historical transaction quantity as the basis of an overall recommendation algorithm; and the weight coefficient among the commodities is calculated according to the historical browsing frequency, so that the individual weight of each label in the space of each object and the whole sample is realized, all limited information is used to the maximum extent, and the accuracy of information recommendation is improved.
Drawings
FIG. 1 is a diagram illustrating an exemplary application of a method for recommending merchandise;
FIG. 2 is a flowchart illustrating a method for recommending merchandise according to an embodiment;
FIG. 3 is a flowchart illustrating the similarity calculation step in one embodiment;
fig. 4 is a flowchart illustrating a method for generating a tag of contents of an article according to another embodiment;
FIG. 5 is a flowchart illustrating a method for recommending merchandise according to another embodiment;
FIG. 6 is a block diagram showing the structure of an article recommending apparatus according to an embodiment;
FIG. 7 is a diagram illustrating an internal structure of a computer device according to an embodiment.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.
The commodity recommendation method provided by the application can be applied to the application environment shown in fig. 1. Wherein the terminal 102 communicates with the server 104 via a network. The server 104 receives a recommendation instruction sent by the terminal 102, wherein the recommendation instruction carries a low-frequency commodity identifier and a current user identifier; the server 104 acquires a historical browsing record corresponding to the low-frequency commodity identification, wherein the historical browsing record comprises a historical user identification and historical low-frequency commodity information; the server 104 calculates the weight coefficient of the unviewed commodity information and the historical low-frequency commodity information in the database according to the historical browsing frequency of the historical browsing record, and calculates the similarity between the unviewed commodity information and the historical low-frequency commodity information according to the weight coefficient; the server 104 counts the interest degree value of the unviewed commodity information corresponding to the current user identification; the server 104 sorts the unviewed commodities of the unviewed commodity information according to the interest degree value, and recommends the unviewed commodities to the terminal 102 according to the sorting result. The terminal 102 may be, but not limited to, various personal computers, notebook computers, smart phones, tablet computers, and portable smart devices, and the server 104 may be implemented by an independent server or a server cluster formed by a plurality of servers.
In one embodiment, as shown in fig. 2, a method for recommending goods is provided, which is described by taking the method as an example applied to the server in fig. 1, and includes the following steps:
step 202, receiving a recommendation instruction sent by the terminal, wherein the recommendation instruction carries a low-frequency commodity identifier and a current user identifier.
And the recommending instruction is used for indicating that part of the commodities to be recommended, which correspond to the interest preference of the current user, are selected from the commodities to be recommended. The commodity to be recommended is a low-frequency commodity, and the low-frequency commodity refers to an automobile or a house property which is low in purchase frequency and has few digitalized transactions. The recommendation instruction carries a low-frequency commodity identification and a current user identification. And the server receives a recommendation instruction sent by the terminal.
And 204, acquiring a historical browsing record corresponding to the low-frequency commodity identification, wherein the historical browsing record comprises a historical user identification and historical low-frequency commodity information.
The historical browsing records are the browsing records of historical low-frequency commodities by all historical users. The historical browsing records comprise historical user identifications and historical low-frequency commodity information. The historical low-frequency commodity information is various types of sales information corresponding to the historical low-frequency commodities and can include historical low-frequency commodity identification, historical commodity content and historical commodity evaluation. And the server acquires the historical browsing records corresponding to the low-frequency commodity identification.
And step 206, calculating the weight coefficient of the unviewed commodity information and the historical low-frequency commodity information in the database according to the historical browsing frequency of the historical browsing record, and calculating the similarity of the unviewed commodity information and the historical low-frequency commodity information according to the weight coefficient.
The historical browsing frequency is the number of occurrences of the historical low-frequency commodities in the historical browsing records. The unviewed commodities refer to low-frequency commodities which are unviewed by the current user and correspond to the current user identifier. The weight coefficient corresponds to the current user identifier and can be obtained by calculation according to the historical browsing frequency of the unviewed commodities and the historical browsing frequency of the historical low-frequency commodity information. For example, the weight coefficient is 1 — the historical browsing frequency of the unviewed goods/the historical browsing frequency of the Σ historical low-frequency goods information; the weighting coefficient of the i commodities is the historical browsing frequency of the commodities i/the occurrence frequency of all the commodities in the historical browsing record.
The server can construct a commodity browsing frequency-user identification matrix according to the user identification and the commodity identification, wherein the number of users is M, and the number of commodities is N. In the following article browsing frequency-user identification matrix, M is 3, and N is 4. The element in the matrix that is empty represents that the corresponding user has no browsing behavior on the commodity, and the score of the commodity scored by the user can also be considered as 0. Therefore, with the user U1Corresponding to I1And I2The weight coefficient of the product is 3/(3+0+2+0) +0/(3+0+2+0) ═ 0.6; and user U2Corresponding to I1And I2The weight coefficient of the product is 0/(0+3+3+4) +3/(0+3+3+4) is 0.3; and user U3Corresponding to I1And I2The weight coefficient of the product is 4/(4+2+0+2) +2/(4+2+0+2) is 0.75. The server calculates I according to the commodity weight coefficient1And I2The commercial similarity was 0.6+0.3+0.75 ═ 1.65.
U1 U2 U3
I1 3 4
I2 3
I3 2 3 2
I4 4 2
The server calculates the weight coefficient of the unviewed commodity information and the historical low-frequency commodity information in the database according to the historical browsing frequency of the historical browsing record, and calculates the similarity of the unviewed commodity information and the historical low-frequency commodity information according to the weight coefficient. The server can count the historical browsing frequency of each historical low-frequency commodity in the historical browsing record, and calculate the weight coefficient of each historical low-frequency commodity information according to the counted historical browsing frequency. The server calculates a commodity approximate value between the unviewed commodity information and the historical low-frequency commodity information, and then calculates the similarity between the unviewed commodity information and the historical low-frequency commodity information according to the weight coefficient and the commodity approximate value.
And step 208, counting the interest degree value of the unviewed commodity information corresponding to the current user identification based on the similarity.
And the server counts the interest degree value of the unviewed commodity information corresponding to the current user identification. The server can construct a similarity matrix between the unviewed commodities and the historical low-frequency commodities according to the current user identification, and then calculate the interest value of each unviewed commodity according to the similarity matrix. The server can extract the similarity between the historical low-frequency commodities corresponding to the current user identification and the unviewed commodities, and then sum the extracted similarity to obtain the interest degree value.
And step 210, sequencing the unviewed commodities of the unviewed commodity information according to the interest degree value, and recommending the unviewed commodities to the terminal according to a sequencing result.
And the server sorts the unviewed commodities of the unviewed commodity information according to the interest degree value and recommends the unviewed commodities to the terminal according to a sorting result. The server can recommend the unviewed commodities with the maximum interest value to the terminal, and can also recommend the unviewed commodities with the interest values larger than the preset threshold value to the terminal.
In the commodity recommendation method, the historical browsing record corresponding to the low-frequency commodity identification is obtained; calculating a weight coefficient of the unviewed commodity information and the historical low-frequency commodity information in the database according to the historical browsing frequency of the historical browsing record, and calculating the similarity of the unviewed commodity information and the historical low-frequency commodity information according to the weight coefficient; generating an interest degree value of the unviewed commodity information corresponding to the current user identification; according to the interestingness value, ordering the unviewed commodities of the unviewed commodity information, recommending the unviewed commodities to the terminal according to an ordering result, determining the distribution condition of the content tags through historical browsing frequency, and adopting the historical browsing frequency to replace the historical transaction quantity as the basis of an overall recommendation algorithm; and the weight coefficient among the commodities is calculated according to the historical browsing frequency, so that the individual weight of each label in the space of each object and the whole sample is realized, all limited information is used to the maximum extent, and the accuracy of information recommendation is improved.
In one embodiment, as shown in fig. 3, calculating a weight coefficient of the unviewed commodity information and the historical low-frequency commodity information in the database according to the historical browsing frequency of the historical browsing record, and calculating the similarity between the unviewed commodity information and the historical low-frequency commodity information according to the weight coefficient, includes the following steps:
step 302, counting the number of labels of the commodity content labels overlapped by the unviewed commodity information and the historical low-frequency commodity information.
The commodity content tag is a keyword extracted from the commodity information based on evaluation, public praise or the like, and may carry emotional information, for example, the commodity information is "2018 type 2.0T two-drive luxury version 7 seat country VI, the rear seat space is large, the suspension system is good", and the commodity content tag may be "rear seat space is large", "suspension system is good", or the like. And the server counts the label quantity of the commodity content labels overlapped by the unviewed commodity information and the historical low-frequency commodity information.
And step 304, calculating a commodity weight coefficient corresponding to the historical low-frequency commodity information according to the number of the labels.
And the server calculates a commodity weight coefficient corresponding to the historical low-frequency commodity information according to the number of the tags. The server can count the total number of the labels in the historical browsing records, and calculate the weight coefficient of the commodity according to the number of the labels overlapped by the information of the unviewed commodity and the information of the historical low-frequency commodity and the total number of the labels.
And step 306, calculating the similarity between the unviewed commodity information and the historical low-frequency commodity information according to the commodity weight coefficient, the historical user identification and the historical browsing frequency.
And the server calculates the similarity between the unviewed commodity information and the historical low-frequency commodity information according to the commodity weight coefficient, the historical user identification and the historical browsing frequency. The server can calculate an approximate value between the unviewed commodity information and the historical low-frequency commodity information according to the historical user identification and the historical browsing frequency, and then adjust the approximate value according to the commodity weight coefficient to obtain the similarity. In one embodiment, the server is based on a formula
Figure BDA0002279397460000081
Calculating the similarity between the unviewed commodity information and the historical low-frequency commodity information, wherein wijSimilarity between the unviewed commodity i and the historical low-frequency commodity j is obtained; | n (i) | is the number of tags of unviewed item i; i (u) is the number of the tags u appearing in the full sample of the product, and is used for representing the weight of different importance of each tag, wherein the weight of the tag with less appearance I (u) is higher; ln is the punishment effect for smoothing long tail, and adding 1 is to avoid that the denominator is zero when the ln value is calculated, and | n (i) | n (j) | is the number of tags existing in the unviewed goods i and the historical low-frequency goods j at the same time.
In the commodity recommendation method, the distribution condition of the content tags can be determined through the historical browsing frequency, and the historical browsing frequency is used as the basis of the overall recommendation algorithm instead of the historical transaction quantity; and the weight coefficient among the commodities is calculated according to the historical browsing frequency, so that the individual weight of each label in the space of each object and the whole sample is realized, all limited information is used to the maximum extent, and the accuracy of information recommendation is improved.
In one embodiment, as shown in fig. 4, a method for generating a label of contents of an article includes the following steps:
step 402, historical low-frequency commodity information is obtained, and the historical low-frequency commodity information carries historical commodity evaluation.
The historical low-frequency commodity information is various sales information corresponding to the historical low-frequency commodity and can carry historical commodity evaluation. The historical low frequency merchandise information may be stored in a local database or database of another server. The server acquires historical low-frequency commodity information.
And step 404, analyzing the historical low-frequency commodity information through the semantic analysis text, and extracting emotional sentences carrying emotional information.
And analyzing the historical low-frequency commodity information by the server through the semantic analysis text, and extracting emotional sentences carrying emotional information. The server can analyze the historical low-frequency commodity information by adopting semantic analysis models such as a Class-based ngram model, a topoc-based ngram model, a cache-based ngram model, a skiping ngram model and an RNN (remote navigation network), and extracts emotion sentences with emotion words from the historical low-frequency commodity information according to preset emotion words. For example, the preset emotional words may be "good, excellent, and saving", and the server may extract emotional sentences with the emotional words "good, excellent, and saving" from the historical low-frequency commodity information. The server can also analyze the historical low-frequency commodity information, and according to the preset emotional words and the preset parts of speech, the emotional sentences which are provided with the emotional words and have parts of speech corresponding to the preset parts of speech are extracted from the historical low-frequency commodity information.
And 406, aggregating the emotion sentences, and setting the aggregated emotion sentences as commodity content labels.
And the server aggregates the emotion sentences, and sets the aggregated emotion sentences as commodity content labels.
In the commodity recommendation method, the commodity content tags are evaluation tags generated according to the commodity contents in the historical browsing records, and commodities can be rapidly classified through the commodity content tags, so that recommendation of the commodities is facilitated.
In one embodiment, obtaining the historical browsing records corresponding to the low-frequency commodity identifications comprises the following steps: generating commodity browsing page information according to a commodity browsing request of a user; extracting low-frequency commodity identification and commodity evaluation sentences corresponding to historical low-frequency commodities from commodity browsing page information; and aggregating the commodity evaluation sentences according to the low-frequency commodity identification and the historical user identification to generate a historical browsing record.
The commodity browsing page information is webpage information for displaying low-frequency commodities, and may carry a low-frequency commodity identification, a detailed introduction of the low-frequency commodities, evaluation of the commodities by a historical user, browsing times corresponding to a current user identification, and the like. The commodity browsing request is used for acquiring a browsing request of the low-frequency commodity information, and may carry a low-frequency commodity name or a low-frequency commodity identifier. The server generates commodity browsing page information according to a commodity browsing request of a user, can extract a low-frequency commodity name (and/or a low-frequency commodity identifier) from the commodity browsing information, and then generates the commodity browsing page information according to the low-frequency commodity name (and/or the low-frequency commodity identifier), the commodity evaluation of a historical user, the browsing times corresponding to the current user identifier and the like. And the server extracts the low-frequency commodity identification and the commodity evaluation sentence corresponding to the historical low-frequency commodity from the commodity browsing page information. And the server aggregates the commodity evaluation sentences according to the low-frequency commodity identification and the historical user identification to generate a historical browsing record.
In one embodiment, as shown in fig. 5, after recommending the unviewed goods to the terminal according to the sorting result, the method includes the following steps:
and 502, receiving a request for selecting the browsed commodities fed back by the terminal.
The request for selecting the browsed commodities is generated according to the unviewed commodities displayed by the terminal selected by the user and carries the current user identification and the selected low-frequency commodity identification. And the server receives a request for selecting the browsed commodities fed back by the terminal. When a user selects a commodity which is not browsed, the terminal generates a commodity selection browsing request, and the terminal sends the commodity selection browsing request to the server.
And 504, generating a commodity selection record according to the browsing commodity selection request.
And the server generates a commodity selection record according to the commodity selection browsing request. And the server acquires corresponding commodity browsing page information according to the commodity browsing selection request and sends the commodity browsing page information to the terminal. And the server correspondingly stores the current user identification and the selected low-frequency commodity identification in the request for selecting the browsed commodities and generates a commodity selection record.
And step 506, calculating the similarity between the unviewed commodity information and the historical low-frequency commodity information in the database according to the commodity selection record and the historical browsing record.
And the server calculates the similarity between the unviewed commodity information and the historical low-frequency commodity information in the database according to the commodity selection record and the historical browsing record. And the server recalculates the weight coefficient of the unviewed commodity information and the historical low-frequency commodity information in the database according to the commodity selection record and the historical browsing record, and recalculates the similarity of the unviewed commodity information and the historical low-frequency commodity information according to the weight coefficient. The server can sort the unviewed commodities of the unviewed commodity information again based on the similarity, and recommend the unviewed commodities to the terminal in real time according to a sorting result.
According to the commodity recommendation method, the low-frequency commodities are recommended in real time according to the commodity selection browsing request fed back by the terminal, and the accuracy of information recommendation is further improved.
It should be understood that although the various steps in the flow charts of fig. 2-5 are shown in order as indicated by the arrows, the steps are not necessarily performed in order as indicated by the arrows. The steps are not performed in the exact order shown and described, and may be performed in other orders, unless explicitly stated otherwise. Moreover, at least some of the steps in fig. 2-5 may include multiple sub-steps or multiple stages that are not necessarily performed at the same time, but may be performed at different times, and the order of performance of the sub-steps or stages is not necessarily sequential, but may be performed in turn or alternating with other steps or at least some of the sub-steps or stages of other steps.
In one embodiment, as shown in fig. 6, there is provided an article recommendation device including: an instruction receiving module 602, a record obtaining module 604, a similarity calculating module 606, an interestingness value calculating module 608, and a recommending module 610, wherein:
the instruction receiving module 602 is configured to receive a recommendation instruction sent by a terminal, where the recommendation instruction carries a low-frequency product identifier and a current user identifier.
The record obtaining module 604 is configured to obtain a historical browsing record corresponding to the low-frequency product identifier, where the historical browsing record includes a historical user identifier and historical low-frequency product information.
The similarity calculation module 606 is configured to calculate a weight coefficient between the unviewed commodity information and the historical low-frequency commodity information in the database according to the historical browsing frequency of the historical browsing record, and calculate a similarity between the unviewed commodity information and the historical low-frequency commodity information according to the weight coefficient.
And the interest degree value calculating module 608 is configured to count interest degree values of the unviewed commodity information corresponding to the current user identifier based on the similarity.
And the recommending module 610 is configured to sort the unviewed commodities of the unviewed commodity information according to the interest level value, and recommend the unviewed commodities to the terminal according to a sorting result.
In one embodiment, the similarity calculation module includes a tag statistics unit, a weight coefficient generation unit, and a similarity calculation unit, wherein:
and the tag counting unit is used for counting the number of tags of the commodity content tags overlapped by the unviewed commodity information and the historical low-frequency commodity information.
And the weight coefficient generating unit is used for calculating the commodity weight coefficient corresponding to the historical low-frequency commodity information according to the number of the labels.
And the similarity calculation unit is used for calculating the similarity between the unviewed commodity information and the historical low-frequency commodity information according to the commodity weight coefficient, the historical user identification and the historical browsing frequency.
In one embodiment, the similarity calculation module includes a product information acquisition unit, a sentence analysis unit, and a product content tag generation unit, wherein:
and the commodity information acquisition unit is used for acquiring historical low-frequency commodity information, and the historical low-frequency commodity information carries historical commodity evaluation.
And the sentence analysis unit is used for analyzing the historical low-frequency commodity information through the semantic analysis text and extracting the emotional sentences carrying the emotional information.
And a product content tag generation unit configured to aggregate the emotion sentences and set the aggregated emotion sentences as product content tags.
In some embodiments, the similarity calculation module comprises a similarity calculation unit, wherein:
a similarity calculation unit for calculating a similarity according to a formula
Figure BDA0002279397460000121
Calculating unviewed commodity information and historical low-frequency commoditiesSimilarity between information, wherein wijFor the similarity between the unviewed item i and the historical low-frequency item j, | n (i) | is the number of tags of the unviewed item i, i (u) is the number of tags of the full-sample item where the tags u appear, and | n (i) | n (j) | is the number of tags existing in both the unviewed item i and the historical low-frequency item j.
In one embodiment, the record obtaining module includes a browsing page information generating unit, an information extracting unit, and a browsing record generating unit, wherein:
and the browsing page information generating unit is used for generating commodity browsing page information according to the commodity browsing request of the user.
And the information extraction unit is used for extracting the low-frequency commodity identification and the commodity evaluation sentence corresponding to the historical low-frequency commodity from the commodity browsing page information.
And the browsing record generating unit is used for aggregating the commodity evaluation sentences according to the low-frequency commodity identification and the historical user identification to generate a historical browsing record.
In another embodiment, the apparatus further includes a selection request receiving unit, a selection record generating unit, and a similarity secondary calculating unit, wherein:
and the selection request receiving unit is used for receiving the commodity selection browsing request fed back by the terminal.
And the selected record generating unit is used for generating a commodity selected record according to the browsing commodity selection request.
And the similarity secondary calculation unit is used for calculating the similarity between the unviewed commodity information and the historical low-frequency commodity information in the database according to the commodity selection record and the historical browsing record.
For specific limitations of the product recommendation device, reference may be made to the above limitations of the product recommendation method, which are not described herein again. The modules in the commodity recommending device can be wholly or partially realized by software, hardware and a combination thereof. The modules can be embedded in a hardware form or independent from a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.
In one embodiment, a computer device is provided, which may be a server, the internal structure of which may be as shown in fig. 7. The computer device includes a processor, a memory, a network interface, and a database connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The database of the computer device is used for storing commodity recommendation data. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement a merchandise recommendation method.
Those skilled in the art will appreciate that the architecture shown in fig. 7 is merely a block diagram of some of the structures associated with the disclosed aspects and is not intended to limit the computing devices to which the disclosed aspects apply, as particular computing devices may include more or less components than those shown, or may combine certain components, or have a different arrangement of components.
In one embodiment, there is provided a computer device comprising a memory storing a computer program and a processor implementing the following steps when the processor executes the computer program: receiving a recommendation instruction sent by a terminal, wherein the recommendation instruction carries a low-frequency commodity identifier and a current user identifier; acquiring a historical browsing record corresponding to the low-frequency commodity identification, wherein the historical browsing record comprises a historical user identification and historical low-frequency commodity information; calculating a weight coefficient of the unviewed commodity information and the historical low-frequency commodity information in the database according to the historical browsing frequency of the historical browsing record, and calculating the similarity of the unviewed commodity information and the historical low-frequency commodity information according to the weight coefficient; on the basis of the similarity, counting the interest degree value of the unviewed commodity information corresponding to the current user identification; and sequencing the unviewed commodities of the unviewed commodity information according to the interest value, and recommending the unviewed commodities to the terminal according to a sequencing result.
In one embodiment, the calculating, by the processor when executing the computer program, a weight coefficient of the unviewed commodity information and the historical low-frequency commodity information in the database according to the historical browsing frequency of the historical browsing record, and calculating a similarity of the unviewed commodity information and the historical low-frequency commodity information according to the weight coefficient includes: counting the number of labels of the commodity content labels overlapped by the unviewed commodity information and the historical low-frequency commodity information; calculating a commodity weight coefficient corresponding to the historical low-frequency commodity information according to the number of the labels; and calculating the similarity between the unviewed commodity information and the historical low-frequency commodity information according to the commodity weight coefficient, the historical user identification and the historical browsing frequency.
In one embodiment, a method for generating a tag for content of an article, implemented by a processor executing a computer program, comprises: acquiring historical low-frequency commodity information, wherein the historical low-frequency commodity information carries historical commodity evaluation; analyzing historical low-frequency commodity information through a semantic analysis text, and extracting emotion sentences carrying emotion information; and aggregating the emotion sentences, and setting the aggregated emotion sentences as commodity content labels.
In one embodiment, the calculating of the similarity between the unviewed commodity information and the historical low-frequency commodity information according to the commodity weight coefficient, the historical user identification and the historical browsing frequency, which is realized when the processor executes the computer program, comprises: according to the formula
Figure BDA0002279397460000141
Calculating the similarity between the unviewed commodity information and the historical low-frequency commodity information, wherein wijFor the similarity between the unviewed item i and the historical low-frequency item j, | n (i) | is the number of tags of the unviewed item i, i (u) is the number of tags of the full-sample item where the tags u appear, and | n (i) | n (j) | is the number of tags existing in both the unviewed item i and the historical low-frequency item j.
In one embodiment, the obtaining of the historical browsing history corresponding to the low-frequency product identification, which is implemented when the processor executes the computer program, includes: generating commodity browsing page information according to a commodity browsing request of a user; extracting low-frequency commodity identification and commodity evaluation sentences corresponding to historical low-frequency commodities from commodity browsing page information; and aggregating the commodity evaluation sentences according to the low-frequency commodity identification and the historical user identification to generate a historical browsing record.
In one embodiment, after recommending, by the processor, the unviewed goods to the terminal according to the sorting result when executing the computer program, the method includes: receiving a request for selecting browsed commodities fed back by a terminal; generating a commodity selection record according to the commodity selection browsing request; and calculating the similarity between the unviewed commodity information in the database and the historical low-frequency commodity information according to the commodity selection record and the historical browsing record.
In one embodiment, a computer-readable storage medium is provided, having a computer program stored thereon, which when executed by a processor, performs the steps of: receiving a recommendation instruction sent by a terminal, wherein the recommendation instruction carries a low-frequency commodity identifier and a current user identifier; acquiring a historical browsing record corresponding to the low-frequency commodity identification, wherein the historical browsing record comprises a historical user identification and historical low-frequency commodity information; calculating a weight coefficient of the unviewed commodity information and the historical low-frequency commodity information in the database according to the historical browsing frequency of the historical browsing record, and calculating the similarity of the unviewed commodity information and the historical low-frequency commodity information according to the weight coefficient; on the basis of the similarity, counting the interest degree value of the unviewed commodity information corresponding to the current user identification; and sequencing the unviewed commodities of the unviewed commodity information according to the interest value, and recommending the unviewed commodities to the terminal according to a sequencing result.
In one embodiment, the computer program, when executed by a processor, for calculating a weight coefficient of unviewed commodity information and historical low-frequency commodity information in a database according to historical browsing frequency of historical browsing records, and calculating similarity of the unviewed commodity information and the historical low-frequency commodity information according to the weight coefficient, includes: counting the number of labels of the commodity content labels overlapped by the unviewed commodity information and the historical low-frequency commodity information; calculating a commodity weight coefficient corresponding to the historical low-frequency commodity information according to the number of the labels; and calculating the similarity between the unviewed commodity information and the historical low-frequency commodity information according to the commodity weight coefficient, the historical user identification and the historical browsing frequency.
In one embodiment, a method for generating a tag for content of an article, implemented by a computer program when executed by a processor, comprises: acquiring historical low-frequency commodity information, wherein the historical low-frequency commodity information carries historical commodity evaluation; analyzing historical low-frequency commodity information through a semantic analysis text, and extracting emotion sentences carrying emotion information; and aggregating the emotion sentences, and setting the aggregated emotion sentences as commodity content labels.
In one embodiment, the computer program, when executed by a processor, implements calculating a similarity between unviewed merchandise information and historical low frequency merchandise information based on the merchandise weight coefficient, the historical user identification, and the historical frequency of browsing, including: according to the formula
Figure BDA0002279397460000151
Calculating the similarity between the unviewed commodity information and the historical low-frequency commodity information, wherein wijFor the similarity between the unviewed item i and the historical low-frequency item j, | n (i) | is the number of tags of the unviewed item i, i (u) is the number of tags of the full-sample item where the tags u appear, and | n (i) | n (j) | is the number of tags existing in both the unviewed item i and the historical low-frequency item j.
In one embodiment, the computer program when executed by a processor implements obtaining historical browsing records corresponding to low-frequency commodity identifications, including: generating commodity browsing page information according to a commodity browsing request of a user; extracting low-frequency commodity identification and commodity evaluation sentences corresponding to historical low-frequency commodities from commodity browsing page information; and aggregating the commodity evaluation sentences according to the low-frequency commodity identification and the historical user identification to generate a historical browsing record.
In one embodiment, the computer program, when executed by a processor, after recommending to a terminal unviewed goods according to the ranking result, comprises: receiving a request for selecting browsed commodities fed back by a terminal; generating a commodity selection record according to the commodity selection browsing request; and calculating the similarity between the unviewed commodity information in the database and the historical low-frequency commodity information according to the commodity selection record and the historical browsing record.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in the embodiments provided herein may include non-volatile and/or volatile memory, among others. Non-volatile memory can include read-only memory (ROM), Programmable ROM (PROM), Electrically Programmable ROM (EPROM), Electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDRSDRAM), Enhanced SDRAM (ESDRAM), Synchronous Link DRAM (SLDRAM), Rambus Direct RAM (RDRAM), direct bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).
The technical features of the above embodiments can be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the above embodiments are not described, but should be considered as the scope of the present specification as long as there is no contradiction between the combinations of the technical features.
The above-mentioned embodiments only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present patent shall be subject to the appended claims.

Claims (10)

1. A method of merchandise recommendation, the method comprising:
receiving a recommendation instruction sent by a terminal, wherein the recommendation instruction carries a low-frequency commodity identification and a current user identification;
acquiring a historical browsing record corresponding to the low-frequency commodity identification, wherein the historical browsing record comprises a historical user identification and historical low-frequency commodity information;
calculating a weight coefficient of unviewed commodity information and historical low-frequency commodity information in a database according to the historical browsing frequency of the historical browsing record, and calculating the similarity of the unviewed commodity information and the historical low-frequency commodity information according to the weight coefficient;
counting the interest degree value of the unviewed commodity information corresponding to the current user identification based on the similarity;
and sequencing the unviewed commodities of the unviewed commodity information according to the interest degree value, and recommending the unviewed commodities to the terminal according to a sequencing result.
2. The method according to claim 1, wherein the calculating a weight coefficient of the unviewed commodity information and the historical low-frequency commodity information in the database according to the historical browsing frequency of the historical browsing record, and calculating the similarity of the unviewed commodity information and the historical low-frequency commodity information according to the weight coefficient comprises:
counting the number of labels of the commodity content labels overlapped with the historical low-frequency commodity information in the unviewed commodity information;
calculating a commodity weight coefficient corresponding to the historical low-frequency commodity information according to the number of the tags;
and calculating the similarity between the unviewed commodity information and the historical low-frequency commodity information according to the commodity weight coefficient, the historical user identification and the historical browsing frequency.
3. The method of claim 2, wherein the method for generating the merchandise content tag comprises:
acquiring historical low-frequency commodity information, wherein the historical low-frequency commodity information carries historical commodity evaluation;
analyzing the historical low-frequency commodity information through a semantic analysis text, and extracting emotion sentences carrying emotion information;
and aggregating the emotion sentences, and setting the aggregated emotion sentences as commodity content labels.
4. The method of claim 2, wherein the calculating the similarity between the unviewed commodity information and the historical low-frequency commodity information according to the commodity weight coefficient, the historical user identification and the historical browsing frequency comprises:
according to the formula
Figure FDA0002279397450000021
Calculating the similarity between the unviewed commodity information and the historical low-frequency commodity information, wherein wijFor the similarity between the unviewed item i and the historical low-frequency item j, | n (i) | is the number of tags of the unviewed item i, i (u) is the number of tags of the full-sample item where the tags u appear, and | n (i) | n (j) | is the number of tags existing in both the unviewed item i and the historical low-frequency item j.
5. The method of claim 1, wherein the obtaining of the historical browsing history corresponding to the low-frequency product identifier comprises:
generating commodity browsing page information according to a commodity browsing request of a user;
extracting low-frequency commodity identification and commodity evaluation sentences corresponding to historical low-frequency commodities from the commodity browsing page information;
and aggregating commodity evaluation sentences according to the low-frequency commodity identification and the historical user identification to generate a historical browsing record.
6. The method according to claim 1, wherein after recommending the unviewed goods to the terminal according to the sorting result, the method comprises:
receiving a request for selecting browsed commodities fed back by a terminal;
generating a commodity selection record according to the commodity selection browsing request;
and calculating the similarity between the unviewed commodity information in the database and the historical low-frequency commodity information according to the commodity selection record and the historical browsing record.
7. An article recommendation device, the device comprising:
the instruction receiving module is used for receiving a recommendation instruction sent by the terminal, wherein the recommendation instruction carries a low-frequency commodity identifier and a current user identifier;
the record acquisition module is used for acquiring a historical browsing record corresponding to the low-frequency commodity identification, wherein the historical browsing record comprises a historical user identification and historical low-frequency commodity information;
the similarity calculation module is used for calculating the weight coefficient of the unviewed commodity information and the historical low-frequency commodity information in the database according to the historical browsing frequency of the historical browsing record and calculating the similarity of the unviewed commodity information and the historical low-frequency commodity information according to the weight coefficient;
the interest degree value calculation module is used for counting the interest degree value of the unviewed commodity information corresponding to the current user identification based on the similarity;
and the recommending module is used for sequencing the unviewed commodities of the unviewed commodity information according to the interest degree value and recommending the unviewed commodities to the terminal according to a sequencing result.
8. The apparatus of claim 7, wherein the similarity calculation module comprises:
the label counting unit is used for counting the number of labels of the commodity content labels overlapped with the history low-frequency commodity information in the unviewed commodity information;
the weight coefficient generating unit is used for calculating a commodity weight coefficient corresponding to the historical low-frequency commodity information according to the number of the labels;
and the similarity calculation unit is used for calculating the similarity between the unviewed commodity information and the historical low-frequency commodity information according to the commodity weight coefficient, the historical user identification and the historical browsing frequency.
9. A computer device comprising a memory and a processor, the memory storing a computer program, wherein the processor implements the steps of the method of any one of claims 1 to 6 when executing the computer program.
10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method of any one of claims 1 to 6.
CN201911135202.6A 2019-11-19 2019-11-19 Commodity recommendation method, commodity recommendation device, computer equipment and storage medium Active CN111080398B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201911135202.6A CN111080398B (en) 2019-11-19 2019-11-19 Commodity recommendation method, commodity recommendation device, computer equipment and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201911135202.6A CN111080398B (en) 2019-11-19 2019-11-19 Commodity recommendation method, commodity recommendation device, computer equipment and storage medium

Publications (2)

Publication Number Publication Date
CN111080398A true CN111080398A (en) 2020-04-28
CN111080398B CN111080398B (en) 2024-04-05

Family

ID=70311185

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201911135202.6A Active CN111080398B (en) 2019-11-19 2019-11-19 Commodity recommendation method, commodity recommendation device, computer equipment and storage medium

Country Status (1)

Country Link
CN (1) CN111080398B (en)

Cited By (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111563206A (en) * 2020-07-03 2020-08-21 北京每日优鲜电子商务有限公司 Article information sequence display method and device, electronic equipment and medium
CN111899068A (en) * 2020-07-07 2020-11-06 拉扎斯网络科技（上海）有限公司 Commodity shopping guide method and device, storage medium and computer equipment
CN112446709A (en) * 2020-12-15 2021-03-05 安徽银通物联有限公司 Face recognition payment authentication method and system
CN112581230A (en) * 2020-12-24 2021-03-30 安徽航天信息科技有限公司 Commodity recommendation method and device
CN113222697A (en) * 2021-05-11 2021-08-06 湖北三赫智能科技有限公司 Commodity information pushing method, commodity information pushing device, computer equipment and readable storage medium
WO2022001349A1 (en) * 2020-07-01 2022-01-06 北京京东振世信息技术有限公司 Method and device for information analysis
CN116681490A (en) * 2023-06-13 2023-09-01 广东七洲科技股份有限公司 Information recommendation method based on identity authentication
CN116894713A (en) * 2023-09-07 2023-10-17 酒仙网络科技股份有限公司 Wine sales management system based on e-commerce platform
CN117033803A (en) * 2023-10-10 2023-11-10 山东爱书人家庭教育科技有限公司 Information recommendation method, system, device and medium

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2010224873A (en) * 2009-03-24 2010-10-07 Nomura Research Institute Ltd Commodity retrieval server, commodity retrieval method, program, and recording medium
CN103106259A (en) * 2013-01-25 2013-05-15 西北工业大学 Mobile webpage content recommending method based on situation
CN105469263A (en) * 2014-09-24 2016-04-06 阿里巴巴集团控股有限公司 Commodity recommendation method and device
WO2017101317A1 (en) * 2015-12-14 2017-06-22 乐视控股（北京）有限公司 Method and apparatus for displaying intelligent recommendations on different terminals
CN108182621A (en) * 2017-12-07 2018-06-19 合肥美的智能科技有限公司 The Method of Commodity Recommendation and device for recommending the commodity, equipment and storage medium
WO2018218403A1 (en) * 2017-05-27 2018-12-06 深圳大学 Content pushing method and device
CN110442784A (en) * 2019-07-08 2019-11-12 中国平安人寿保险股份有限公司 Information recommendation method, device, computer equipment and storage medium

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2010224873A (en) * 2009-03-24 2010-10-07 Nomura Research Institute Ltd Commodity retrieval server, commodity retrieval method, program, and recording medium
CN103106259A (en) * 2013-01-25 2013-05-15 西北工业大学 Mobile webpage content recommending method based on situation
CN105469263A (en) * 2014-09-24 2016-04-06 阿里巴巴集团控股有限公司 Commodity recommendation method and device
WO2017101317A1 (en) * 2015-12-14 2017-06-22 乐视控股（北京）有限公司 Method and apparatus for displaying intelligent recommendations on different terminals
WO2018218403A1 (en) * 2017-05-27 2018-12-06 深圳大学 Content pushing method and device
CN108182621A (en) * 2017-12-07 2018-06-19 合肥美的智能科技有限公司 The Method of Commodity Recommendation and device for recommending the commodity, equipment and storage medium
CN110442784A (en) * 2019-07-08 2019-11-12 中国平安人寿保险股份有限公司 Information recommendation method, device, computer equipment and storage medium

Cited By (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2022001349A1 (en) * 2020-07-01 2022-01-06 北京京东振世信息技术有限公司 Method and device for information analysis
CN111563206A (en) * 2020-07-03 2020-08-21 北京每日优鲜电子商务有限公司 Article information sequence display method and device, electronic equipment and medium
CN111563206B (en) * 2020-07-03 2024-01-12 秦皇岛肆拾贰宇数字科技有限公司 Method, device, electronic equipment and medium for displaying object information sequence
CN111899068A (en) * 2020-07-07 2020-11-06 拉扎斯网络科技（上海）有限公司 Commodity shopping guide method and device, storage medium and computer equipment
CN112446709A (en) * 2020-12-15 2021-03-05 安徽银通物联有限公司 Face recognition payment authentication method and system
CN112581230A (en) * 2020-12-24 2021-03-30 安徽航天信息科技有限公司 Commodity recommendation method and device
CN113222697A (en) * 2021-05-11 2021-08-06 湖北三赫智能科技有限公司 Commodity information pushing method, commodity information pushing device, computer equipment and readable storage medium
CN116681490A (en) * 2023-06-13 2023-09-01 广东七洲科技股份有限公司 Information recommendation method based on identity authentication
CN116681490B (en) * 2023-06-13 2024-02-23 广东七洲科技股份有限公司 Information recommendation method based on identity authentication
CN116894713A (en) * 2023-09-07 2023-10-17 酒仙网络科技股份有限公司 Wine sales management system based on e-commerce platform
CN117033803A (en) * 2023-10-10 2023-11-10 山东爱书人家庭教育科技有限公司 Information recommendation method, system, device and medium
CN117033803B (en) * 2023-10-10 2024-01-09 山东爱书人家庭教育科技有限公司 Information recommendation method, system, device and medium

Also Published As

Publication number Publication date
CN111080398B (en) 2024-04-05

Similar Documents

Publication Publication Date Title
CN111080398B (en) Commodity recommendation method, commodity recommendation device, computer equipment and storage medium
CN109582876B (en) Tourist industry user portrait construction method and device and computer equipment
CN110866181B (en) Resource recommendation method, device and storage medium
CN105989004B (en) Information delivery preprocessing method and device
CN111199428B (en) Commodity recommendation method and device, storage medium and computer equipment
CN108205768B (en) Database establishing method, data recommending device, equipment and storage medium
US20170091805A1 (en) Advertisement Recommendation Method and Advertisement Recommendation Server
CN109189904A (en) Individuation search method and system
CN111178949B (en) Service resource matching reference data determining method, device, equipment and storage medium
JP2016511906A (en) Ranking product search results
CN110827112B (en) Deep learning commodity recommendation method and device, computer equipment and storage medium
CN109816134B (en) Method and device for predicting delivery address and storage medium
CN111325609A (en) Commodity recommendation list determining method and device, electronic equipment and storage medium
CN110647696B (en) Business object sorting method and device
CN112001754A (en) User portrait generation method, device, equipment and computer readable medium
CN112148973B (en) Data processing method and device for information push
CN112598472B (en) Product recommendation method, device, system, medium and program product
CN111400613A (en) Article recommendation method, device, medium and computer equipment
CN111967914A (en) User portrait based recommendation method and device, computer equipment and storage medium
CN113032668A (en) Product recommendation method, device and equipment based on user portrait and storage medium
CN112487283A (en) Method and device for training model, electronic equipment and readable storage medium
CN111680213A (en) Information recommendation method, data processing method and device
CN113919882A (en) Intelligent design method of personalized discount coupon, electronic device and storage medium
CN113763035A (en) Advertisement delivery effect prediction method and device, computer equipment and storage medium
CN112288279A (en) Business risk assessment method and device based on natural language processing and linear regression

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
GR01 Patent grant
GR01 Patent grant