CN113592605A - Product recommendation method, device, equipment and storage medium based on similar products - Google Patents
Product recommendation method, device, equipment and storage medium based on similar products Download PDFInfo
- Publication number
- CN113592605A CN113592605A CN202110913134.2A CN202110913134A CN113592605A CN 113592605 A CN113592605 A CN 113592605A CN 202110913134 A CN202110913134 A CN 202110913134A CN 113592605 A CN113592605 A CN 113592605A
- Authority
- CN
- China
- Prior art keywords
- product
- user
- products
- preset
- target
- 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
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0631—Item recommendations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/289—Phrasal analysis, e.g. finite state techniques or chunking
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/30—Semantic analysis
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Accounting & Taxation (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Finance (AREA)
- Data Mining & Analysis (AREA)
- General Health & Medical Sciences (AREA)
- Computational Linguistics (AREA)
- Health & Medical Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- General Business, Economics & Management (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to a data analysis technology, and discloses a product recommendation method based on similar products, which comprises the following steps: the method comprises the steps of obtaining a product portrait of a product to be recommended and user portraits of a candidate user group of the product to be recommended, selecting a first target user group from the candidate user group, and recommending the product to be recommended to the first target user group; feedback data of a to-be-recommended product of a first target user group is obtained, and the user portrait of the first target user group is subjected to portrait improvement according to the feedback data to obtain the portrait of the target user; selecting a second target user group from the candidate user groups according to the target user image; and recommending the product to be recommended to a second target user group. In addition, the invention also relates to a block chain technology, and the product portrait and the user portrait can be stored in the nodes of the block chain. The invention also provides a product recommendation device, electronic equipment and a storage medium based on similar products. The invention can improve the product recommendation accuracy.
Description
Technical Field
The invention relates to the technical field of data analysis, in particular to a product recommendation method and device based on similar products, electronic equipment and a computer-readable storage medium.
Background
The product recommendation is an important branch of AI artificial intelligence, most of the existing product recommendations are based on various data of users, various machine learning and deep learning algorithms are applied to generate user portraits for describing the personalized characteristics of the users in detail, and then the similarity between the user portraits and pre-generated product portraits is utilized to recommend the products to the users.
However, when data is processed, especially when facing a large amount of user data and a large amount of product data, refined data analysis of the artificial intelligence model occupies a large amount of computing resources, which results in reduction of recommendation efficiency, and in order to improve model robustness, the artificial intelligence model has a fixed data processing flow during data processing, that is, different data uses the same processing flow, so that data needs to be converted into a fixed form before data processing, which further prolongs the data processing flow and reduces the product recommendation efficiency.
Disclosure of Invention
The invention provides a product recommendation method and device based on similar products and a computer readable storage medium, and mainly aims to solve the problem of low efficiency in product recommendation.
In order to achieve the above object, the present invention provides a product recommendation method based on similar products, including:
the method comprises the steps of obtaining product description of each product in a preset product set, and extracting keywords of each product from the product description;
converting the keywords into label vectors, and calculating the similarity between every two products in the product set according to the label vectors;
the method comprises the steps of obtaining product recommendation requirements, selecting a preset user screening algorithm according to the product recommendation requirements, and screening a plurality of users from a preset user group by utilizing the user screening algorithm;
acquiring behavior data of each user in the plurality of users on the product set, and extracting preference products corresponding to each user from the behavior data;
randomly selecting two users from the plurality of users as a first user and a second user, and counting that products which exist in the preferred products of the first user but do not exist in the preferred products of the second user are different products;
and selecting the products with the similarity between the product set and the different products larger than a preset similarity threshold as similar products, and recommending the different products and the similar products to the second user.
Optionally, the extracting the keyword of each product from the product description includes:
performing word segmentation processing on the product description to obtain a product word segmentation set;
selecting one of the products from the product set one by one as a target product, and selecting one of the participles from the product participles of the target product one by one as a target participle;
counting a first frequency of the target word segmentation in the product word segmentation of the target product and counting a second frequency of the target word segmentation in the product word segmentation set;
calculating a key value of the target word segmentation according to the first frequency and the second frequency, and selecting the target word segmentation with the key value larger than a preset key threshold value as a keyword of the target product.
Optionally, the calculating a key value of the target segmented word according to the first frequency and the second frequency includes:
calculating a key value of the target participle by using the following key value algorithm:
wherein K is the key value, W is a preset coefficient, f1Is said first frequency, f2Is the second frequency.
Optionally, the converting the keyword into a tag vector includes:
selecting one product from the product set one by one as a target product, and converting the keywords of the selected target product into word vectors by using a preset word vector model;
counting the vector length of each vector in the word vectors, and selecting the word vector with the longest vector length in the word vectors as a target vector;
and extending the vector length of all vectors in the word vectors to the vector length of the target vector by using preset parameters, and splicing all word vectors as row vectors to obtain the label vectors.
Optionally, the selecting a preset user screening algorithm according to the product recommendation requirement includes:
performing convolution and pooling on the product recommendation requirement by using a preset semantic analysis model to obtain a requirement characteristic of the product recommendation requirement;
calculating a characteristic value of each characteristic in the demand characteristics by using a preset activation function, and selecting the characteristic with the characteristic value larger than a preset characteristic threshold value as a core semantic meaning of the product recommendation requirement;
acquiring an algorithm tag corresponding to each of a plurality of preset user screening algorithms, converting the core semantics into semantic vectors, respectively calculating a matching value of the semantic vectors and each algorithm tag, and selecting the user screening algorithm of which the matching value is smaller than a preset distance threshold.
Optionally, the extracting, from the behavior data, a preferred product corresponding to each user includes:
identifying a data type of the behavioral data;
compiling a preset character into a regular expression by utilizing a compiler corresponding to the data type;
selecting one user from the multiple users one by one as a target user, and extracting browsing times, browsing duration and purchasing conditions of the target user on each product in the product set from behavior data of the target user by using the regular expression;
and calculating the preference degree of the target user to each product in the product set according to the browsing times, the browsing duration and the purchasing condition by using a preset weight algorithm, and selecting the product with the preference degree larger than a preset preference threshold value as the preferred product of the target user.
Optionally, the calculating, by using a preset weight algorithm, a preference degree of the target user for each product in the product set according to the browsing times, the browsing duration and the purchase condition includes:
calculating the preference degree of the target user for each product in the product set by using the following weight algorithm:
wherein P is the preference degree, M is the browsing frequency of the nth product in the product set by the target user, and T isnA browsing duration S for the target user to browse the nth product in the product set for the mth timenA numerical representation, ω, of the purchase of the nth product of the set of products for the target user1、ω2、ω3Is a preset weight coefficient.
In order to solve the above problem, the present invention further provides a product recommendation device based on similar products, the device comprising:
the keyword extraction module is used for acquiring product descriptions of each product in a preset product set and extracting keywords of each product from the product descriptions;
the similarity calculation module is used for converting the keywords into label vectors and calculating the similarity between every two products in the product set according to the label vectors;
the system comprises a user screening module, a product recommendation module and a user selection module, wherein the user screening module is used for acquiring product recommendation requirements, selecting a preset user screening algorithm according to the product recommendation requirements, and screening a plurality of users from a preset user group by using the user screening algorithm;
the preference product screening module is used for acquiring behavior data of each user in the plurality of users on the product set and extracting preference products corresponding to each user from the behavior data;
a difference product screening module, configured to randomly select two users from the multiple users as a first user and a second user, and count that products that exist in the preferred products of the first user but do not exist in the preferred products of the second user are difference products;
and the product recommending module is used for selecting the products with the similarity between the products in the product set and the difference products larger than a preset similarity threshold as similar products and recommending the difference products and the similar products to the second user.
In order to solve the above problem, the present invention also provides an electronic device, including:
a memory storing at least one instruction; and
and the processor executes the instructions stored in the memory to realize the similar product-based product recommendation method.
In order to solve the above problem, the present invention further provides a computer-readable storage medium, which stores at least one instruction, and the at least one instruction is executed by a processor in an electronic device to implement the similar product based product recommendation method described above.
According to the embodiment of the invention, the corresponding user screening algorithm can be selected according to the product recommendation requirement to screen out a plurality of users belonging to the same category, products are pushed to the users according to the difference products among the products preferred by the users, the products similar to the difference products are selected according to the similarity among the products, and the products are pushed to the users at the same time, so that the refined analysis of user data and product data by utilizing complex modeling is avoided, the corresponding algorithm is selected according to the product recommendation requirement, the different data are prevented from being processed by utilizing a uniform process, and the product recommendation efficiency is improved. Therefore, the product recommendation method, the product recommendation device, the electronic equipment and the computer-readable storage medium based on similar products, which are provided by the invention, can solve the problem of low efficiency in product recommendation.
Drawings
Fig. 1 is a schematic flowchart of a product recommendation method based on similar products according to an embodiment of the present invention;
FIG. 2 is a schematic flow chart of vector stitching according to an embodiment of the present invention;
fig. 3 is a schematic flow chart of a user selection screening algorithm according to an embodiment of the present invention;
FIG. 4 is a functional block diagram of a similar product based product recommendation apparatus according to an embodiment of the present invention;
fig. 5 is a schematic structural diagram of an electronic device implementing the similar product based product recommendation method according to an embodiment of the present invention.
The implementation, functional features and advantages of the objects of the present invention will be further explained with reference to the accompanying drawings.
Detailed Description
It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
The embodiment of the application provides a product recommendation method based on similar products. The execution subject of the product recommendation method based on similar products includes, but is not limited to, at least one of electronic devices that can be configured to execute the method provided by the embodiments of the present application, such as a server, a terminal, and the like. In other words, the product recommendation method based on similar products may be executed by software or hardware installed in a terminal device or a server device, and the software may be a block chain platform. The server includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, and the like.
Referring to fig. 1, a flowchart of a product recommendation method based on similar products according to an embodiment of the present invention is shown. In this embodiment, the product recommendation method based on similar products includes:
s1, obtaining product description of each product in a preset product set, and extracting keywords of each product from the product description.
In the embodiment of the present invention, the product set is a set including a plurality of different products provided by a supplier, an agent, and the like of a product in advance, and the product description includes data such as a product name, a product function, and a product price of each product in the product set.
In the embodiment of the invention, a computer sentence (java sentence, python sentence, etc.) with a data crawling function can be used for crawling the product description which is authorized to be retrievable from a pre-constructed storage area, wherein the storage area comprises but is not limited to a database, a block chain node and a network cache.
In an embodiment of the present invention, the extracting keywords of each product from the product description includes:
performing word segmentation processing on the product description to obtain a product word segmentation set;
selecting one of the products from the product set one by one as a target product, and selecting one of the participles from the product participles of the target product one by one as a target participle;
counting a first frequency of the target word segmentation in the product word segmentation of the target product and counting a second frequency of the target word segmentation in the product word segmentation set;
calculating a key value of the target word segmentation according to the first frequency and the second frequency, and selecting the target word segmentation with the key value larger than a preset key threshold value as a keyword of the target product.
In detail, the product description may be subjected to word segmentation Processing by using a pre-constructed word segmentation model, so as to obtain a product word segmentation set including product words of the product description of each product in the product set, where the word segmentation model includes, but is not limited to, a Natural Language Processing (NLP) model and a Latent Dirichlet Allocation (LDA) model.
Specifically, the first frequency refers to the number of times that the target word occurs in the product word of the target product, and the second frequency refers to the number of times that the target word occurs in the product word set.
Further, the embodiment of the present invention may calculate a key value of the target segmented word by using the first frequency and the second frequency, where the key value is used to identify the generalization degree of the target segmented word to the product description of the target product, and when the key value is larger (i.e. the generalization degree of the target segmented word to the product description of the target product is larger), it indicates that the product description of the target segmented word to the target product is more critical.
In an embodiment of the present invention, the calculating a key value of the target word segmentation according to the first frequency and the second frequency includes:
calculating a key value of the target participle by using the following key value algorithm:
wherein K is the key value, W is a preset coefficient, f1Is said first frequency, f2Is the second frequency.
In the embodiment of the invention, after the key value is obtained through calculation, the target participle with the key value larger than a preset key threshold value is selected as the key word of the target product.
For example, there are product A and product B in the product set, where the product participle of product A includes a1And a2The product word of the product B comprises B1And b2And a is a1Has a critical value of 80, a2Has a critical value of 20, b1Has a critical value of 30, b2The critical value of (a) is 70, and when the preset critical threshold value is 60, a is selected1Selecting b as the key word of the product A2Is the keyword of the product B.
S2, converting the keywords into label vectors, and calculating the similarity between every two products in the product set according to the label vectors.
In the embodiment of the invention, in order to improve the accuracy of subsequent product recommendation, the keywords can be converted into a vector form, and because a plurality of keywords can be described in each product, the word vectors obtained by converting the keywords can be spliced to obtain the label vector.
In the embodiment of the present invention, as shown in fig. 2, the converting the keyword into a tag vector includes:
s21, selecting one product from the product set one by one as a target product, and converting the keywords of the selected target product into word vectors by using a preset word vector model;
s22, counting the vector length of each vector in the word vectors, and selecting the word vector with the longest vector length in the word vectors as a target vector;
and S23, extending the vector lengths of all the vectors in the word vectors to the vector length of the target vector by using preset parameters, and splicing all the word vectors as row vectors to obtain the label vectors.
In detail, the word vector model includes, but is not limited to, a word2vec model and a BERT model, but since lengths of word vectors obtained after keyword conversion may be inconsistent, in order to improve subsequent calculation efficiency and reduce occupation of calculation resources, the vector lengths of the word vectors may be unified.
Specifically, the vector length of each vector in the word vectors may be counted, the word vector with the longest vector length is selected as the target vector, and the vector lengths of the vectors in the word vectors except the target vector are extended to the vector length of the target vector by using preset parameters, so as to unify the vector lengths of all the vectors in the word vectors, where the preset parameters may be preset.
For example, the word vector of the target product includes vector a, vector B, and vector C, where vector a is: (1,2), vector B is: (1,3,6,7), vector C is: (4,2,5), it is known that, if the vector length of the vector a is 2, the vector length of the vector B is 4, and the vector length of the vector C is 3, the vector B is selected as the target vector, and when the preset parameter is 0, the vector a is extended to: (1,2,0,0), extending vector C to: (4,2,5,0).
In the embodiment of the invention, because the target product comprises a plurality of keywords, and each keyword corresponds to one word vector, the word vectors can be spliced as row vectors to obtain the label vector of the target product name.
For example, there is a vector A: (1,2,0,0), vector B: (1,3,6,7) vector C: (4,2,5,0), the vector a, the vector B and the vector C are taken as row vectors and are spliced into a label vector of the following form:
further, the similarity between every two products in the product set may be calculated according to the label vector, for example, if the product set includes product a, product B, and product C, the similarity between product a and product B may be calculated according to the label vector of product a and the label vector of product B, the similarity between product a and product C may be calculated according to the label vector of product a and the label vector of product C, and the similarity between product B and product C may be calculated according to the label vector of product B and the label vector of product C.
In this embodiment of the present invention, the calculating a similarity between every two products in the product set according to the tag vector includes:
calculating the similarity between every two products in the product set by using a similarity algorithm as follows:
wherein, Sim(k,j)Is the similarity between the kth product and the jth product in the product set, XkA label vector, Y, for a kth product in the set of productsjIs the jth of the product setA label vector for the product.
S3, obtaining product recommendation requirements, selecting a preset user screening algorithm according to the product recommendation requirements, and screening a plurality of users from a preset user group by using the user screening algorithm.
In the embodiment of the invention, the product recommendation condition is provided by suppliers, agents and the like of the product recommendation demand product in advance.
For example, users whose professions belong to the financial industry are recommended, users whose age group belongs to 20 to 35 years old are recommended, users whose residence belongs to xx provinces are recommended, users whose liveness on a preset platform is greater than a preset liveness threshold value are recommended, and the like.
In one practical application scenario of the invention, because the analysis data needed to be analyzed is inconsistent when users are screened under different recommendation conditions, the user screening algorithm corresponding to the user requirements is selected according to the product recommendation requirements, so that the accuracy of user screening is improved, and the accuracy and the efficiency of product recommendation for the users are improved.
In detail, the user filtering algorithm includes, but is not limited to, a collaborative filtering algorithm, an ALS recall algorithm, a category recall algorithm, and the like.
In the embodiment of the present invention, referring to fig. 3, the selecting a preset user screening algorithm according to the product recommendation requirement includes:
s31, performing convolution and pooling on the product recommendation requirement by using a preset semantic analysis model to obtain requirement characteristics of the product recommendation requirement;
s32, calculating a feature value of each feature in the demand features by using a preset activation function, and selecting the feature with the feature value larger than a preset feature threshold as a core semantic meaning of the product recommendation demand;
s33, obtaining an algorithm label corresponding to each algorithm in a plurality of preset user screening algorithms, converting the core semantics into semantic vectors, respectively calculating a matching value of the semantic vectors and each algorithm label, and selecting the user screening algorithm of which the matching value is smaller than a preset distance threshold.
In detail, the semantic analysis Model includes, but is not limited to, an HMM (Hidden Markov Model), and the semantic analysis Model is used to perform operations such as convolution and pooling on the product recommendation demand, so as to extract demand features of the product recommendation demand, but the extracted demand features may have features extracted by mistake, so that a preset activation function may be used to calculate a feature value of each feature in the demand features, and then the features in the demand features are screened according to the feature values, so as to obtain core semantics of the product recommendation demand, where the activation functions include, but are not limited to, a sigmoid activation function, a softmax activation function, and a relu activation function.
Specifically, the algorithm tag is a tag in a vector form generated in advance according to features of different user screening algorithms, and the algorithm tag can be used for identifying the features of the user screening algorithms.
In the embodiment of the present invention, the step of converting the core semantics into the semantic vector is the same as the step of converting the keyword into the tag vector in S2, and is not described herein again.
In this embodiment of the present invention, the calculating the matching value between the semantic vector and each algorithm label includes:
calculating a matching value of the semantic vector and each algorithm label by using the following distance algorithm:
wherein D is the matching value, A is the semantic vector, BiAnd screening algorithm labels of the ith user in the preset multiple user screening algorithms.
The embodiment of the invention selects the user screening algorithm with the distance smaller than the preset distance threshold value from the preset plurality of user screening algorithms, and calculates the preset user group by using the selected user screening algorithm so as to screen out a plurality of users which accord with the user screening algorithm in the user group.
In detail, the selected user screening algorithm may be multiple, and when the selected user screening algorithm is multiple, the selected user screening algorithm is used to calculate the preset user group one by one, and the users screened by each user screening algorithm are collected, so as to realize screening of multiple users from the preset user group.
S4, acquiring the behavior data of each user in the plurality of users to the product set, and extracting the preference product corresponding to each user from the behavior data.
In this embodiment of the present invention, the behavior data includes, but is not limited to, data of browsing times, browsing duration, purchasing records, and the like of each of the plurality of users for the products in the product set.
In detail, the step of obtaining the behavior data of each user in the plurality of users on the product set is consistent with the step of obtaining the product description of each product in the preset product set in S1, and is not repeated here.
The embodiment of the invention can analyze the behavior data to obtain the preference of each of the plurality of users and the user for the products in the product set.
For example, a product a and a product B exist in the product set, and a user a and a user B exist, wherein the browsing times of the product a by the user a is 20 times, the browsing time per time is 2 minutes, the user purchases the product a, the browsing times of the product B by the user a is 3 times, the browsing time per time is 1 minute, and the user does not purchase the product B; the browsing times of the user B on the product a are 7, the browsing time is 3 minutes each time, the user does not purchase the product a, the browsing times of the user B on the product a are 15, the browsing time is 5 minutes each time, and the user purchases the product B, so that the user a can be judged to be better than the product a, and the user B is better than the product B.
In this embodiment of the present invention, the extracting the preference product corresponding to each user from the behavior data includes:
identifying a data type of the behavioral data;
compiling a preset character into a regular expression by utilizing a compiler corresponding to the data type;
selecting one user from the multiple users one by one as a target user, and extracting browsing times, browsing duration and purchasing conditions of the target user on each product in the product set from behavior data of the target user by using the regular expression;
and calculating the preference degree of the target user to each product in the product set according to the browsing times, the browsing duration and the purchasing condition by using a preset weight algorithm, and selecting the product with the preference degree larger than a preset preference threshold value as the preferred product of the target user.
In detail, since the behavior data may include a plurality of data types (e.g., numeric data or non-numeric data), in order to improve the accuracy of analyzing the behavior data, a java statement having a data type identification function may be used to identify the data type of the behavior data, and a compiler corresponding to the data type may be used to compile a preset character into a regular expression.
Specifically, the regular expression may be used to extract a field in a specific format, so as to extract data such as browsing times, browsing duration, and purchase condition of each product in the product set from the behavior data.
For example, the browsing duration is often expressed in xx hours xx minutes xx seconds, so that all the data expressed in the format of xx hours xx minutes xx seconds in the behavior data of the target user can be extracted by using the regular expression of "xx hours xx minutes xx seconds", so as to obtain the browsing duration of each product in the product numbers of the target user.
Further, a preset weighting algorithm can be used for calculating the preference degree of the target user for each product in the product set according to the browsing times, the browsing duration and the purchasing condition.
In an embodiment of the present invention, the calculating, by using a preset weighting algorithm, a preference degree of the target user for each product in the product set according to the browsing times, the browsing duration, and the purchase condition includes:
calculating the preference degree of the target user for each product in the product set by using the following weight algorithm:
wherein P is the preference degree, M is the browsing frequency of the nth product in the product set by the target user, and T isnA browsing duration S for the target user to browse the nth product in the product set for the mth timenPurchase of the nth product of the product set for the target user (when the target user has purchased the nth product, S)nWhen the target user does not purchase the nth product, Sn＝0)，ω1、ω2、ω3Is a preset weight coefficient.
In the embodiment of the invention, the product with the preference degree larger than the preset preference threshold value is selected as the preferred product of the target user.
For example, a product a, a product b, a product c, and a product d exist in the product set, where the preference degree of the target user for the product a is 78, the preference degree of the target user for the product b is 34, the preference degree of the target user for the product c is 87, and the preference degree of the target user for the product d is 43, and when a preset preference threshold is 50, the product a and the product c are selected as preferred products of the target user.
And S5, randomly selecting two users from the plurality of users as a first user and a second user, and counting that products which exist in the preferred products of the first user but do not exist in the preferred products of the second user are different products.
In the embodiment of the present invention, two users may be arbitrarily selected from the plurality of users as a first user and a second user, and the preferred products of the first user and the second user are counted, and the products existing in the preferred products of the first user but not existing in the preferred products of the second user are collected, and the collected products are determined as the difference products of the first user to the second user.
For example, if a product a, a product B, and data C exist in a preferred product of a first user, and a product a, a product D, and a product E exist in a preferred product of a second user, statistics shows that the product B and the product C exist in the preferred product of the first user, but the product C and the product D do not exist in the preferred product of the second user, and the product C and the product D can be collected as the difference product.
In one practical application scenario of the present invention, because the first user and the second user are users selected from a preset user group by the same product recommendation requirement, the first user and the second user may be considered to belong to the same user group, and further, the preferred product of the first user may also be a potential preferred product of the second user.
According to the embodiment of the invention, the products which exist in the preferred products of the first user but do not exist in the preferred products of the second user are obtained by counting the preferred products of the first user and the second user, so that the products are different products, the products can be pushed to the second user by using the different products, the user data does not need to be analyzed in detail, and the product recommendation efficiency is improved.
S6, selecting the products in the product set, of which the similarity with the difference products is larger than a preset similarity threshold value, as similar products, and recommending the difference products and the similar products to the second user.
In one practical application scenario of the present invention, if the product recommendation is performed on the second user only according to the different product, the coverage of the recommended product on the product set is small, and the diversity of the product is small, so that in the embodiment of the present invention, a product with a similarity greater than a preset similarity threshold between the product set and the different product is selected as a similar product, and the different product and the similar product are recommended to the second user.
For example, the difference product includes a product a and a product B, the product set includes a product C, a product D, a product E, and a product F, and according to the similarity between every two products in the product set calculated in step S2, the similarity between the product a and the product C is 20, the similarity between the product a and the product D is 40, the similarity between the product a and the product E is 60, the similarity between the product a and the product F is 80, the similarity between the product B and the product C is 88, the similarity between the product B and the product D is 44, the similarity between the product B and the product E is 33, the similarity between the product B and the product F is 22, and when the preset similarity threshold is 50, it can be determined that the product E and the product F are similar products of the product a, and the product C is similar product of the product B, so that the product a, the product D, C, and C, are similar products Recommending the product B, the product C, the product E and the product F to the second user.
Further, in this embodiment of the present invention, after recommending the difference product and the similar product to the second user, the method further includes returning to step S5, reselecting two users as the first user and the second user, and recommending the product to the second user according to steps S5 and S6 until the product recommendation to all users in the multiple users is completed.
According to the embodiment of the invention, the corresponding user screening algorithm can be selected according to the product recommendation requirement to screen out a plurality of users belonging to the same category, products are pushed to the users according to the difference products among the products preferred by the users, the products similar to the difference products are selected according to the similarity among the products, and the products are pushed to the users at the same time, so that the refined analysis of user data and product data by utilizing complex modeling is avoided, the corresponding algorithm is selected according to the product recommendation requirement, the different data are prevented from being processed by utilizing a uniform process, and the product recommendation efficiency is improved. Therefore, the product recommendation method based on similar products can solve the problem of low efficiency in product recommendation.
Fig. 4 is a functional block diagram of a product recommendation device based on similar products according to an embodiment of the present invention.
The product recommendation device 100 based on similar products can be installed in an electronic device. According to the implemented functions, the similar product based product recommendation device 100 may include a keyword extraction module 101, a similarity calculation module 102, a user filtering module 103, a preferred product filtering module 104, a difference product filtering module 105 and a product recommendation module 106. The module of the present invention, which may also be referred to as a unit, refers to a series of computer program segments that can be executed by a processor of an electronic device and that can perform a fixed function, and that are stored in a memory of the electronic device.
In the present embodiment, the functions regarding the respective modules/units are as follows:
the keyword extraction module 101 is configured to obtain a product description of each product in a preset product set, and extract a keyword of each product from the product description;
the similarity calculation module 102 is configured to convert the keyword into a tag vector, and calculate a similarity between every two products in the product set according to the tag vector;
the user screening module 103 is configured to obtain product recommendation requirements, select a preset user screening algorithm according to the product recommendation requirements, and screen a plurality of users from a preset user group by using the user screening algorithm;
the preferred product screening module 104 is configured to obtain behavior data of each user in the plurality of users on the product set, and extract a preferred product corresponding to each user from the behavior data;
the difference product screening module 105 is configured to randomly select two users from the multiple users as a first user and a second user, and count that products that exist in the preferred products of the first user but do not exist in the preferred products of the second user are difference products;
the product recommending module 106 is configured to select a product in the product set, which has a similarity greater than a preset similarity threshold with the difference product, as a similar product, and recommend the difference product and the similar product to the second user.
In detail, when the modules in the product recommendation device 100 based on similar products according to the embodiment of the present invention are used, the same technical means as the product recommendation method based on similar products described in fig. 1 to 3 are adopted, and the same technical effects can be produced, which is not described herein again.
Fig. 5 is a schematic structural diagram of an electronic device implementing a product recommendation method based on similar products according to an embodiment of the present invention.
The electronic device 1 may comprise a processor 10, a memory 11, a communication bus 12 and a communication interface 13, and may further comprise a computer program, such as a similar product based product recommendation program, stored in the memory 11 and executable on the processor 10.
In some embodiments, the processor 10 may be composed of an integrated circuit, for example, a single packaged integrated circuit, or may be composed of a plurality of integrated circuits packaged with the same function or different functions, and includes one or more Central Processing Units (CPUs), a microprocessor, a digital Processing chip, a graphics processor, a combination of various control chips, and the like. The processor 10 is a Control Unit (Control Unit) of the electronic device, connects various components of the electronic device by using various interfaces and lines, and executes various functions and processes data of the electronic device by running or executing programs or modules stored in the memory 11 (for example, executing a product recommendation program based on similar products, etc.), and calling data stored in the memory 11.
The memory 11 includes at least one type of readable storage medium including flash memory, removable hard disks, multimedia cards, card-type memory (e.g., SD or DX memory, etc.), magnetic memory, magnetic disks, optical disks, etc. The memory 11 may in some embodiments be an internal storage unit of the electronic device, for example a removable hard disk of the electronic device. The memory 11 may also be an external storage device of the electronic device in other embodiments, such as a plug-in mobile hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), and the like, which are provided on the electronic device. Further, the memory 11 may also include both an internal storage unit and an external storage device of the electronic device. The memory 11 may be used not only to store application software installed in the electronic device and various types of data, such as codes of product recommendation programs based on similar products, etc., but also to temporarily store data that has been output or is to be output.
The communication bus 12 may be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus. The bus may be divided into an address bus, a data bus, a control bus, etc. The bus is arranged to enable connection communication between the memory 11 and at least one processor 10 or the like.
The communication interface 13 is used for communication between the electronic device and other devices, and includes a network interface and a user interface. Optionally, the network interface may include a wired interface and/or a wireless interface (e.g., WI-FI interface, bluetooth interface, etc.), which are typically used to establish a communication connection between the electronic device and other electronic devices. The user interface may be a Display (Display), an input unit such as a Keyboard (Keyboard), and optionally a standard wired interface, a wireless interface. Alternatively, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, an OLED (Organic Light-Emitting Diode) touch device, or the like. The display, which may also be referred to as a display screen or display unit, is suitable, among other things, for displaying information processed in the electronic device and for displaying a visualized user interface.
Fig. 5 only shows an electronic device with components, and it will be understood by a person skilled in the art that the structure shown in fig. 5 does not constitute a limitation of the electronic device 1, and may comprise fewer or more components than shown, or a combination of certain components, or a different arrangement of components.
For example, although not shown, the electronic device may further include a power supply (such as a battery) for supplying power to each component, and preferably, the power supply may be logically connected to the at least one processor 10 through a power management device, so that functions of charge management, discharge management, power consumption management and the like are realized through the power management device. The power supply may also include any component of one or more dc or ac power sources, recharging devices, power failure detection circuitry, power converters or inverters, power status indicators, and the like. The electronic device may further include various sensors, a bluetooth module, a Wi-Fi module, and the like, which are not described herein again.
It is to be understood that the described embodiments are for purposes of illustration only and that the scope of the appended claims is not limited to such structures.
The memory 11 in the electronic device 1 stores a similar product based product recommendation program that is a combination of instructions that, when executed in the processor 10, may implement:
the method comprises the steps of obtaining product description of each product in a preset product set, and extracting keywords of each product from the product description;
converting the keywords into label vectors, and calculating the similarity between every two products in the product set according to the label vectors;
the method comprises the steps of obtaining product recommendation requirements, selecting a preset user screening algorithm according to the product recommendation requirements, and screening a plurality of users from a preset user group by utilizing the user screening algorithm;
acquiring behavior data of each user in the plurality of users on the product set, and extracting preference products corresponding to each user from the behavior data;
randomly selecting two users from the plurality of users as a first user and a second user, and counting that products which exist in the preferred products of the first user but do not exist in the preferred products of the second user are different products;
and selecting the products with the similarity between the product set and the different products larger than a preset similarity threshold as similar products, and recommending the different products and the similar products to the second user.
Specifically, the specific implementation method of the processor 10 for the instruction may refer to the description of the relevant steps in the embodiment corresponding to fig. 1, which is not described herein again.
Further, the integrated modules/units of the electronic device 1, if implemented in the form of software functional units and sold or used as separate products, may be stored in a computer readable storage medium. The computer readable storage medium may be volatile or non-volatile. For example, the computer-readable medium may include: any entity or device capable of carrying said computer program code, recording medium, U-disk, removable hard disk, magnetic disk, optical disk, computer Memory, Read-Only Memory (ROM).
The present invention also provides a computer-readable storage medium, storing a computer program which, when executed by a processor of an electronic device, may implement:
the method comprises the steps of obtaining product description of each product in a preset product set, and extracting keywords of each product from the product description;
converting the keywords into label vectors, and calculating the similarity between every two products in the product set according to the label vectors;
the method comprises the steps of obtaining product recommendation requirements, selecting a preset user screening algorithm according to the product recommendation requirements, and screening a plurality of users from a preset user group by utilizing the user screening algorithm;
acquiring behavior data of each user in the plurality of users on the product set, and extracting preference products corresponding to each user from the behavior data;
randomly selecting two users from the plurality of users as a first user and a second user, and counting that products which exist in the preferred products of the first user but do not exist in the preferred products of the second user are different products;
and selecting the products with the similarity between the product set and the different products larger than a preset similarity threshold as similar products, and recommending the different products and the similar products to the second user.
In the embodiments provided in the present invention, it should be understood that the disclosed apparatus, device and method can be implemented in other ways. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the modules is only one logical functional division, and other divisions may be realized in practice.
The modules described as separate parts may or may not be physically separate, and parts displayed as modules may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment.
In addition, functional modules in the embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, or in a form of hardware plus a software functional module.
It will be evident to those skilled in the art that the invention is not limited to the details of the foregoing illustrative embodiments, and that the present invention may be embodied in other specific forms without departing from the spirit or essential attributes thereof.
The present embodiments are therefore to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. Any reference signs in the claims shall not be construed as limiting the claim concerned.
The block chain is a novel application mode of computer technologies such as distributed data storage, point-to-point transmission, a consensus mechanism, an encryption algorithm and the like. A block chain (Blockchain), which is essentially a decentralized database, is a series of data blocks associated by using a cryptographic method, and each data block contains information of a batch of network transactions, so as to verify the validity (anti-counterfeiting) of the information and generate a next block. The blockchain may include a blockchain underlying platform, a platform product service layer, an application service layer, and the like.
Furthermore, it is obvious that the word "comprising" does not exclude other elements or steps, and the singular does not exclude the plural. A plurality of units or means recited in the system claims may also be implemented by one unit or means in software or hardware. The terms first, second, etc. are used to denote names, but not any particular order.
Finally, it should be noted that the above embodiments are only for illustrating the technical solutions of the present invention and not for limiting, and although the present invention is described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that modifications or equivalent substitutions may be made on the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.
Claims (10)
1. A method for recommending similar products, the method comprising:
the method comprises the steps of obtaining product description of each product in a preset product set, and extracting keywords of each product from the product description;
converting the keywords into label vectors, and calculating the similarity between every two products in the product set according to the label vectors;
the method comprises the steps of obtaining product recommendation requirements, selecting a preset user screening algorithm according to the product recommendation requirements, and screening a plurality of users from a preset user group by utilizing the user screening algorithm;
acquiring behavior data of each user in the plurality of users on the product set, and extracting preference products corresponding to each user from the behavior data;
randomly selecting two users from the plurality of users as a first user and a second user, and counting that products which exist in the preferred products of the first user but do not exist in the preferred products of the second user are different products;
and selecting the products with the similarity between the product set and the different products larger than a preset similarity threshold as similar products, and recommending the different products and the similar products to the second user.
2. The similar product based product recommendation method of claim 1, wherein the extracting keywords of each product from the product description comprises:
performing word segmentation processing on the product description to obtain a product word segmentation set;
selecting one of the products from the product set one by one as a target product, and selecting one of the participles from the product participles of the target product one by one as a target participle;
counting a first frequency of the target word segmentation in the product word segmentation of the target product and counting a second frequency of the target word segmentation in the product word segmentation set;
calculating a key value of the target word segmentation according to the first frequency and the second frequency, and selecting the target word segmentation with the key value larger than a preset key threshold value as a keyword of the target product.
3. The similar product based product recommendation method of claim 2, wherein said calculating key values of the target participles according to the first frequency and the second frequency comprises:
calculating a key value of the target participle by using the following key value algorithm:
wherein K is the key value, W is a preset coefficient, f1Is said first frequency, f2Is the second frequency.
4. The similar product based product recommendation method of claim 1, wherein the converting the keyword into a tag vector comprises:
selecting one product from the product set one by one as a target product, and converting the keywords of the selected target product into word vectors by using a preset word vector model;
counting the vector length of each vector in the word vectors, and selecting the word vector with the longest vector length in the word vectors as a target vector;
and extending the vector length of all vectors in the word vectors to the vector length of the target vector by using preset parameters, and splicing all word vectors as row vectors to obtain the label vectors.
5. The product recommendation method based on similar products as claimed in claim 1, wherein said selecting a preset user filtering algorithm according to said product recommendation requirement comprises:
performing convolution and pooling on the product recommendation requirement by using a preset semantic analysis model to obtain a requirement characteristic of the product recommendation requirement;
calculating a characteristic value of each characteristic in the demand characteristics by using a preset activation function, and selecting the characteristic with the characteristic value larger than a preset characteristic threshold value as a core semantic meaning of the product recommendation requirement;
acquiring an algorithm tag corresponding to each of a plurality of preset user screening algorithms, converting the core semantics into semantic vectors, respectively calculating a matching value of the semantic vectors and each algorithm tag, and selecting the user screening algorithm of which the matching value is smaller than a preset distance threshold.
6. The similar product-based product recommendation method of any one of claims 1-5, wherein the extracting the preference product corresponding to each user from the behavior data comprises:
identifying a data type of the behavioral data;
compiling a preset character into a regular expression by utilizing a compiler corresponding to the data type;
selecting one user from the multiple users one by one as a target user, and extracting browsing times, browsing duration and purchasing conditions of the target user on each product in the product set from behavior data of the target user by using the regular expression;
and calculating the preference degree of the target user to each product in the product set according to the browsing times, the browsing duration and the purchasing condition by using a preset weight algorithm, and selecting the product with the preference degree larger than a preset preference threshold value as the preferred product of the target user.
7. The method for recommending products based on similar products according to claim 6, wherein said calculating the preference degree of said target user for each product in said product set according to said browsing times, said browsing duration and said purchase condition by using a preset weight algorithm comprises:
calculating the preference degree of the target user for each product in the product set by using the following weight algorithm:
wherein P is the preference degree, M is the browsing frequency of the nth product in the product set by the target user, and T isnA browsing duration S for the target user to browse the nth product in the product set for the mth timenA numerical representation, ω, of the purchase of the nth product of the set of products for the target user1、ω2、ω3Is a preset weight coefficient.
8. A similar product based product recommendation device, the device comprising:
the keyword extraction module is used for acquiring product descriptions of each product in a preset product set and extracting keywords of each product from the product descriptions;
the similarity calculation module is used for converting the keywords into label vectors and calculating the similarity between every two products in the product set according to the label vectors;
the system comprises a user screening module, a product recommendation module and a user selection module, wherein the user screening module is used for acquiring product recommendation requirements, selecting a preset user screening algorithm according to the product recommendation requirements, and screening a plurality of users from a preset user group by using the user screening algorithm;
the preference product screening module is used for acquiring behavior data of each user in the plurality of users on the product set and extracting preference products corresponding to each user from the behavior data;
a difference product screening module, configured to randomly select two users from the multiple users as a first user and a second user, and count that products that exist in the preferred products of the first user but do not exist in the preferred products of the second user are difference products;
and the product recommending module is used for selecting the products with the similarity between the products in the product set and the difference products larger than a preset similarity threshold as similar products and recommending the difference products and the similar products to the second user.
9. An electronic device, characterized in that the electronic device comprises:
at least one processor; and the number of the first and second groups,
a memory communicatively coupled to the at least one processor; wherein the content of the first and second substances,
the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the similar product based product recommendation method of any one of claims 1-7.
10. A computer-readable storage medium storing a computer program, wherein the computer program, when executed by a processor, implements the similar product based product recommendation method of any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110913134.2A CN113592605B (en) | 2021-08-10 | 2021-08-10 | Product recommendation method, device, equipment and storage medium based on similar products |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110913134.2A CN113592605B (en) | 2021-08-10 | 2021-08-10 | Product recommendation method, device, equipment and storage medium based on similar products |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113592605A true CN113592605A (en) | 2021-11-02 |
| CN113592605B CN113592605B (en) | 2023-08-22 |
Family
ID=78256760
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110913134.2A Active CN113592605B (en) | 2021-08-10 | 2021-08-10 | Product recommendation method, device, equipment and storage medium based on similar products |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113592605B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114723523A (en) * | 2022-04-06 | 2022-07-08 | 平安科技（深圳）有限公司 | Product recommendation method, device, equipment and medium based on user capability portrait |
| CN114912948A (en) * | 2022-04-24 | 2022-08-16 | 深圳船奇科技有限公司 | Cloud service-based cross-border e-commerce big data intelligent processing method, device and equipment |
| CN115204971A (en) * | 2022-06-23 | 2022-10-18 | 平安科技（深圳）有限公司 | Product recommendation method and device, electronic equipment and computer-readable storage medium |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103246672A (en) * | 2012-02-09 | 2013-08-14 | 中国科学技术大学 | Method and device for performing personalized recommendation on users |
| CN103942712A (en) * | 2014-05-09 | 2014-07-23 | 北京联时空网络通信设备有限公司 | Product similarity based e-commerce recommendation system and method thereof |
| WO2015032353A1 (en) * | 2013-09-06 | 2015-03-12 | 北京奇虎科技有限公司 | Video recommendation method and device |
| CN106997358A (en) * | 2016-01-22 | 2017-08-01 | 中移（杭州）信息技术有限公司 | Information recommendation method and device |
| CN110348920A (en) * | 2018-04-02 | 2019-10-18 | 中移（杭州）信息技术有限公司 | A kind of method and device of recommended products |
| CN110400202A (en) * | 2019-07-31 | 2019-11-01 | 中国工商银行股份有限公司 | Bank product recommended method and system |
| CN112488863A (en) * | 2020-12-01 | 2021-03-12 | 中国人寿保险股份有限公司 | Dangerous seed recommendation method and related equipment in user cold start scene |
-
2021
- 2021-08-10 CN CN202110913134.2A patent/CN113592605B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103246672A (en) * | 2012-02-09 | 2013-08-14 | 中国科学技术大学 | Method and device for performing personalized recommendation on users |
| WO2015032353A1 (en) * | 2013-09-06 | 2015-03-12 | 北京奇虎科技有限公司 | Video recommendation method and device |
| CN103942712A (en) * | 2014-05-09 | 2014-07-23 | 北京联时空网络通信设备有限公司 | Product similarity based e-commerce recommendation system and method thereof |
| CN106997358A (en) * | 2016-01-22 | 2017-08-01 | 中移（杭州）信息技术有限公司 | Information recommendation method and device |
| CN110348920A (en) * | 2018-04-02 | 2019-10-18 | 中移（杭州）信息技术有限公司 | A kind of method and device of recommended products |
| CN110400202A (en) * | 2019-07-31 | 2019-11-01 | 中国工商银行股份有限公司 | Bank product recommended method and system |
| CN112488863A (en) * | 2020-12-01 | 2021-03-12 | 中国人寿保险股份有限公司 | Dangerous seed recommendation method and related equipment in user cold start scene |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114723523A (en) * | 2022-04-06 | 2022-07-08 | 平安科技（深圳）有限公司 | Product recommendation method, device, equipment and medium based on user capability portrait |
| CN114723523B (en) * | 2022-04-06 | 2023-05-30 | 平安科技（深圳）有限公司 | Product recommendation method, device, equipment and medium based on user capability image |
| CN114912948A (en) * | 2022-04-24 | 2022-08-16 | 深圳船奇科技有限公司 | Cloud service-based cross-border e-commerce big data intelligent processing method, device and equipment |
| CN115204971A (en) * | 2022-06-23 | 2022-10-18 | 平安科技（深圳）有限公司 | Product recommendation method and device, electronic equipment and computer-readable storage medium |
| CN115204971B (en) * | 2022-06-23 | 2024-02-02 | 平安科技（深圳）有限公司 | Product recommendation method, device, electronic equipment and computer readable storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113592605B (en) | 2023-08-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113592605B (en) | Product recommendation method, device, equipment and storage medium based on similar products | |
| CN113449187B (en) | Product recommendation method, device, equipment and storage medium based on double images | |
| CN115002200B (en) | Message pushing method, device, equipment and storage medium based on user portrait | |
| CN114663198A (en) | Product recommendation method, device and equipment based on user portrait and storage medium | |
| CN113704614A (en) | Page generation method, device, equipment and medium based on user portrait | |
| CN115391669B (en) | Intelligent recommendation method and device and electronic equipment | |
| CN113435202A (en) | Product recommendation method and device based on user portrait, electronic equipment and medium | |
| CN114648392B (en) | Product recommendation method and device based on user portrait, electronic equipment and medium | |
| CN113886691A (en) | Intelligent recommendation method and device based on historical data, electronic equipment and medium | |
| CN115018588A (en) | Product recommendation method and device, electronic equipment and readable storage medium | |
| CN113886708A (en) | Product recommendation method, device, equipment and storage medium based on user information | |
| CN113656690A (en) | Product recommendation method and device, electronic equipment and readable storage medium | |
| CN113362162A (en) | Wind control identification method and device based on network behavior data, electronic equipment and medium | |
| CN115204971B (en) | Product recommendation method, device, electronic equipment and computer readable storage medium | |
| CN114780688A (en) | Text quality inspection method, device and equipment based on rule matching and storage medium | |
| CN115186188A (en) | Product recommendation method, device and equipment based on behavior analysis and storage medium | |
| CN114219367A (en) | User scoring method, device, equipment and storage medium | |
| CN113344674A (en) | Product recommendation method, device, equipment and storage medium based on user purchasing power | |
| CN114708073A (en) | Intelligent detection method and device for surrounding mark and serial mark, electronic equipment and storage medium | |
| CN113707302A (en) | Service recommendation method, device, equipment and storage medium based on associated information | |
| CN113486238A (en) | Information pushing method, device and equipment based on user portrait and storage medium | |
| CN113723114A (en) | Semantic analysis method, device and equipment based on multi-intent recognition and storage medium | |
| CN113592606B (en) | Product recommendation method, device, equipment and storage medium based on multiple decisions | |
| CN114970552B (en) | User return visit information analysis method, device, equipment and medium based on micro-service | |
| CN116188049B (en) | Potential user mining method and device based on chain analysis |
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 |