CN111523976A - Commodity recommendation method and device, electronic equipment and storage medium - Google Patents
Commodity recommendation method and device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN111523976A CN111523976A CN202010328864.1A CN202010328864A CN111523976A CN 111523976 A CN111523976 A CN 111523976A CN 202010328864 A CN202010328864 A CN 202010328864A CN 111523976 A CN111523976 A CN 111523976A
- Authority
- CN
- China
- Prior art keywords
- commodity
- user
- recommended
- behavior
- historical
- 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
- 238000000034 method Methods 0.000 title claims abstract description 38
- 230000006399 behavior Effects 0.000 claims description 196
- 238000012549 training Methods 0.000 claims description 27
- 238000004891 communication Methods 0.000 claims description 19
- 230000015654 memory Effects 0.000 claims description 17
- 238000004590 computer program Methods 0.000 claims description 8
- 239000013598 vector Substances 0.000 description 16
- 238000013507 mapping Methods 0.000 description 12
- 230000011218 segmentation Effects 0.000 description 7
- 238000013528 artificial neural network Methods 0.000 description 6
- 238000000605 extraction Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 5
- 238000001914 filtration Methods 0.000 description 5
- 230000006870 function Effects 0.000 description 5
- 230000008569 process Effects 0.000 description 4
- 230000009471 action Effects 0.000 description 3
- 230000004913 activation Effects 0.000 description 3
- 230000008901 benefit Effects 0.000 description 3
- 230000003993 interaction Effects 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 238000004422 calculation algorithm Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 239000010985 leather Substances 0.000 description 2
- 238000011176 pooling Methods 0.000 description 2
- 230000009467 reduction Effects 0.000 description 2
- 230000006403 short-term memory Effects 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 239000000835 fiber Substances 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 230000000306 recurrent effect Effects 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 230000001960 triggered effect Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
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
- G06Q30/0202—Market predictions or forecasting for commercial activities
Landscapes
- Business, Economics & Management (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Strategic Management (AREA)
- Development Economics (AREA)
- Engineering & Computer Science (AREA)
- General Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Marketing (AREA)
- General Physics & Mathematics (AREA)
- Economics (AREA)
- Theoretical Computer Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Data Mining & Analysis (AREA)
- Game Theory and Decision Science (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The embodiment of the invention provides a commodity recommendation method, a commodity recommendation device, electronic equipment and a storage medium, wherein the method comprises the following steps: acquiring historical behavior characteristics, user attributes and commodity query texts corresponding to target users; determining a commodity to be recommended; splicing the commodity attribute and the commodity identification corresponding to any commodity to be recommended with the historical behavior characteristics, the user attribute and the commodity query text; inputting the splicing result into a commodity recommendation model, and outputting a user behavior predicted value corresponding to the splicing result; and recommending the commodity to be recommended to the target user based on the user behavior predicted value. Therefore, the user behavior predicted value is predicted through the historical behavior characteristic, the user attribute and the commodity query text corresponding to the target user, and the commodity attribute and the commodity identification corresponding to the commodity to be recommended, the commodity to be recommended is recommended to the target user based on the user behavior predicted value, and the click rate of commodity recommendation can be effectively improved.
Description
Technical Field
The present invention relates to the field of data processing technologies, and in particular, to a method and an apparatus for recommending a commodity, an electronic device, and a storage medium.
Background
In recent years, with the rapid development of information technology and the internet, the position of electronic commerce in the society and life is more and more remarkable, and an electronic commerce system provides more and more choices for users. Meanwhile, the rapid expansion of the e-commerce scale causes users to spend a lot of time browsing irrelevant commodities, and it is highly desirable for the sellers to recommend commodities to the matched user population.
In the related art, when a user carelessly browses goods in an electronic commerce system, the electronic commerce system may recommend goods to the user based on a user's history browsing behavior. Because the commodities are recommended to the user only based on the historical browsing behaviors of the user, the dimensionality is single, most recommended commodities are easy to be caused and are not commodities which are intended by the user, the user does not browse, collect, buy, place orders and the like, and the click rate of commodity recommendation is low.
Disclosure of Invention
The embodiment of the invention aims to provide a commodity recommendation method, a commodity recommendation device, electronic equipment and a storage medium, so as to achieve the beneficial effect of improving the click rate of commodity recommendation. The specific technical scheme is as follows:
in a first aspect of an embodiment of the present invention, a method for recommending a commodity is provided, where the method includes:
acquiring historical behavior characteristics, user attributes and commodity query texts corresponding to target users, wherein the historical behavior characteristics comprise historical browsing behavior characteristics, historical collection behavior characteristics, historical purchase adding behavior characteristics and historical order placing behavior characteristics;
determining a commodity to be recommended;
splicing the commodity attribute and the commodity identification corresponding to any commodity to be recommended with the historical behavior characteristics, the user attribute and the commodity query text;
inputting the splicing result into a commodity recommendation model, and outputting a user behavior predicted value corresponding to the splicing result;
and recommending the commodity to be recommended to the target user based on the user behavior predicted value.
In an optional embodiment, the obtaining of the historical behavior feature, the user attribute, and the commodity query text corresponding to the target user includes:
under the condition of receiving a commodity recommendation request, determining a target user triggering the commodity recommendation request;
and acquiring historical behavior characteristics, user attributes and commodity query texts corresponding to the target user.
In an optional embodiment, the determining the item to be recommended includes:
and performing multi-path recall on the commodities according to the historical behavior characteristics and the user attributes to determine the commodities to be recommended.
In an optional embodiment, the user behavior prediction value comprises:
the user order taking behavior prediction value is obtained by the user order taking behavior prediction value.
In an optional embodiment, the recommending the to-be-recommended commodity to the target user based on the user behavior prediction value includes:
calculating the weighted sum of the predicted value of the user browsing behavior, the predicted value of the user collecting behavior, the predicted value of the user purchasing behavior and the predicted value of the user ordering behavior;
and recommending the commodities to be recommended to the target user based on the weighted sum.
In an optional embodiment, the recommending the to-be-recommended item to the target user based on the weighted sum includes:
sorting the commodities to be recommended based on the weighted sum;
and selecting a preset number of the commodities to be recommended from the sorted commodities to be recommended to recommend to the target user.
In an optional embodiment, the commodity recommendation model is specifically obtained by:
obtaining training samples, wherein the training samples comprise: historical behavior characteristics, user attributes and commodity query texts corresponding to the user, and commodity attributes and commodity identifications corresponding to the commodities;
and training the commodity recommendation initial model based on the training sample to obtain the commodity recommendation model.
In a second aspect of the embodiments of the present invention, there is also provided a commodity recommending apparatus, including:
the system comprises an acquisition module, a display module and a display module, wherein the acquisition module is used for acquiring historical behavior characteristics, user attributes and commodity query texts corresponding to target users, wherein the historical behavior characteristics comprise historical browsing behavior characteristics, historical collection behavior characteristics, historical purchase behavior characteristics and historical ordering behavior characteristics;
the commodity determining module is used for determining commodities to be recommended;
the splicing module is used for splicing the commodity attribute and the commodity identification corresponding to any commodity to be recommended with the historical behavior characteristics, the user attribute and the commodity query text;
the predicted value output module is used for inputting the splicing result into the commodity recommendation model and outputting the user behavior predicted value corresponding to the splicing result;
and the commodity recommending module is used for recommending the commodity to be recommended to the target user based on the user behavior predicted value.
In a third aspect of the embodiments of the present invention, there is further provided an electronic device, including a processor, a communication interface, a memory, and a communication bus, where the processor, the communication interface, and the memory complete communication with each other through the communication bus;
a memory for storing a computer program;
a processor configured to implement the product recommendation method according to any one of the first aspect described above when executing a program stored in the memory.
In a fourth aspect of the embodiments of the present invention, there is also provided a storage medium having instructions stored therein, which when run on a computer, cause the computer to execute the product recommendation method according to any one of the first aspects.
In a fifth aspect of the embodiments of the present invention, there is also provided a computer program product containing instructions, which when run on a computer, causes the computer to execute the item recommendation method of any one of the above first aspects.
According to the technical scheme provided by the embodiment of the invention, the historical behavior characteristics, the user attributes and the commodity query text corresponding to the target user are acquired, the commodity attributes and the commodity identifications corresponding to any commodity to be recommended are spliced and input into the commodity recommendation model, and the corresponding user behavior predicted value can be output, so that the commodity to be recommended is recommended to the target user based on the user behavior predicted value, wherein the historical behavior characteristics comprise historical browsing behavior characteristics, historical collection behavior characteristics, historical purchase adding behavior characteristics and historical order placing behavior characteristics. Therefore, the user behavior predicted value is predicted through the historical browsing behavior characteristic, the historical collection behavior characteristic, the historical purchase behavior characteristic, the historical ordering behavior characteristic, the user attribute and the commodity query text corresponding to the target user, and the commodity attribute and the commodity identification corresponding to the commodity to be recommended, the commodity to be recommended is recommended to the target user based on the user behavior predicted value, and the click rate of commodity recommendation can be effectively improved.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and together with the description, serve to explain the principles of the invention.
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious for those skilled in the art that other drawings can be obtained according to the drawings without inventive exercise.
Fig. 1 is a schematic flow chart illustrating an implementation of a commodity recommendation method in an embodiment of the present invention;
fig. 2 is a schematic flow chart of an implementation of obtaining a commodity recommendation model according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of an initial model architecture for commodity recommendation shown in an embodiment of the present invention;
fig. 4 is a schematic structural diagram of a commodity recommending apparatus according to an embodiment of the present invention;
fig. 5 is a schematic structural diagram of an electronic device shown in the embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be obtained by a person skilled in the art without any inventive step based on the embodiments of the present invention, are within the scope of the present invention.
As shown in fig. 1, an implementation flow diagram of a commodity recommendation method provided in an embodiment of the present invention is shown, and the method may specifically include the following steps:
s101, acquiring historical behavior characteristics, user attributes and commodity query texts corresponding to target users;
in the embodiment of the invention, when a target user browses commodities in an electronic commerce system (such as APP in the Kyoto mall) without intention, a click behavior occurs, a commodity recommendation request is triggered, and the commodity recommendation request is sent to a server.
Under the condition that the server side receives the commodity recommendation request, the embodiment of the invention determines the target user triggering the commodity recommendation request, so as to obtain the historical behavior characteristics, the user attributes and the commodity query text corresponding to the target user.
Specifically, the commodity recommendation request carries a user ID, and a target user triggering the commodity recommendation request can be determined according to the user ID.
And acquiring historical behavior characteristics, user attributes and commodity query texts corresponding to the target user in a structured historical behavior characteristic index table according to the user ID. For example, for the structured historical behavior feature index table, it can be shown in table 1 below.
TABLE 1
In the embodiment of the present invention, the historical behavior characteristics may be a historical browsing behavior characteristic, a historical collection behavior characteristic, a historical purchase behavior characteristic, and a historical order placing behavior characteristic, which is not limited in the embodiment of the present invention.
For example, the target user browses 3 different commodities such as a mobile phone, a computer, a leather bag and the like in the recent period of time, collects 2 different commodities such as a mobile phone, a leather bag and the like, purchases the mobile phone additionally, and places an order on the mobile phone;
in order to enable the commodity recommendation model to identify the historical behavior characteristics, the invention performs 1-N discrete numerical mapping on all commodities, N represents all commodity types, and in addition, in order to ensure the dimension uniformity of the historical behavior characteristics, 0 is uniformly used for filling missing rows;
assuming that the cell phone mapping ID is 10, the computer mapping ID is 200, and the wallet mapping ID is 3000, four behavior sequences can be obtained for the historical behavior signature, as shown in table 2 below.
| Historical browsing behavior characteristics | [10,200,3000] |
| Historical collection behavior features | [10,3000,0] |
| Historical shopping behavior characteristics | [10,0,0] |
| Historical ordering behavior characteristics | [10,0,0] |
TABLE 2
In this embodiment of the present invention, the user attribute may be a place of daily use, a current location, a gender, a school calendar, an age, and the like, which is not limited in this embodiment of the present invention. The user attribute may be an ID sequence obtained by using chinese word segmentation and mapping.
In the embodiment of the present invention, the commodity query text (i.e. the commodity query text currently input by the target user) may be a query text ID sequence currently input by the user after using chinese word segmentation and mapping, which is not limited in the embodiment of the present invention.
S102, determining a commodity to be recommended;
in the embodiment of the invention, in order to recommend the commodity to the target user, the commodity to be recommended needs to be determined, so that the commodity to be recommended is recommended to the target user.
The commodities to be recommended may be full commodities, that is, all commodities in the electronic commerce system, or may be partial commodities, that is, partial commodities in the electronic commerce system, which is not limited in the embodiment of the present invention.
In order to reduce the calculation amount, reduce the pressure of the commodity recommendation model and improve the timeliness, dimension reduction can be performed on all commodities in the electronic commerce system, namely, part of commodities in the electronic commerce system are determined to be commodities to be recommended. For example, 100 items in the e-commerce system may be determined to be items to be recommended.
Specifically, the embodiment of the present invention may perform multi-way recall on the commodities according to the historical behavior characteristics and the user attributes, and determine the commodities to be recommended, thereby achieving the purpose of performing dimension reduction on all the commodities in the electronic commerce system.
In the embodiment of the present invention, multiple recalls are performed on the commodity according to the historical behavior features and the user attributes, and a specifically used commodity recall algorithm may be a currently existing algorithm, which is not described in detail herein.
S103, splicing the commodity attribute and the commodity identification corresponding to any commodity to be recommended with the historical behavior characteristics, the user attribute and the commodity query text;
for the determined commodities to be recommended, aiming at the commodity attributes and the commodity identifications corresponding to any commodity to be recommended, the commodity attributes and the commodity identifications corresponding to the commodity to be recommended are spliced with the historical behavior characteristics, the user attributes and the commodity query texts, so that input parameters of a commodity recommendation model can be formed.
For example, through the step S102, 100 kinds of commodities to be recommended are determined, and for a commodity attribute and a commodity identifier corresponding to any one of the commodities to be recommended, the commodity attribute and the commodity identifier corresponding to the commodity to be recommended are spliced with the historical behavior feature, the user attribute, and the commodity query text, and the splicing result may be as shown in table 3 below.
TABLE 3
In the embodiment of the present invention, the commodity attribute may be a commodity price, a commodity category, a commodity name, and the like, where non-digitized information such as the commodity name is subjected to word segmentation and mapped to a fixed-length ID sequence, which is not limited in the embodiment of the present invention.
In the embodiment of the present invention, the product identifier may be a product name, a product ID, and the like, where the product name and the like are subjected to non-digital word segmentation and mapped into a fixed-length ID sequence, which is not limited in the embodiment of the present invention.
S104, inputting the splicing result into a commodity recommendation model, and outputting a user behavior prediction value corresponding to the splicing result;
through the step S103, for the commodity attribute and the commodity identifier corresponding to any commodity to be recommended, the commodity attribute and the commodity identifier corresponding to the commodity to be recommended are spliced with the historical behavior feature, the user attribute and the commodity query text, and the splicing result can be input to the commodity recommendation model, so that the user behavior prediction value corresponding to the splicing result is output.
For the user behavior prediction value corresponding to the splicing result, the embodiment of the present invention may include: the user order taking behavior prediction value is obtained by the user order taking behavior prediction value.
In this way, for any item to be recommended, the possibility (or probability) that the target user browses, collects, purchases, and orders the item to be recommended may be represented by the user browsing behavior predicted value, the user collecting behavior predicted value, the user purchasing behavior predicted value, and the user ordering behavior predicted value, respectively.
For example, through the above step S103, 100 splicing results can be obtained and sequentially input to the commodity recommendation model, so as to output a user browsing behavior predicted value, a user collection behavior predicted value, a user purchase behavior predicted value, and a user order placing behavior predicted value corresponding to the splicing results, which can be shown in table 4 below.
TABLE 4
As can be seen from table 4, in the embodiment of the present invention, the product recommendation model may output the user browsing behavior prediction value, the user collection behavior prediction value, the user purchase adding behavior prediction value, and the user order placing behavior prediction value, which respectively represent the probability that the target user browses the product to be recommended, the probability that the target user collects the product to be recommended, the probability that the target user purchases the product to be recommended, and the probability that the target user orders the product to be recommended.
And S105, recommending the commodity to be recommended to the target user based on the user behavior predicted value.
For the user behavior predicted values output by the commodity recommendation model, namely the user browsing behavior predicted value, the user collection behavior predicted value, the user purchase adding behavior predicted value and the user ordering behavior predicted value, the embodiment of the invention can recommend commodities to be recommended to the target user according to the user behavior predicted values.
Specifically, the embodiment of the invention can calculate the weighted sum of the predicted values of the user browsing behaviors, the user collecting behaviors, the user purchase adding behaviors and the user order placing behaviors, and recommend the commodities to be recommended to the target user based on the weighted sum.
For example, for a vendor, a target user pays 0.2 yuan for one browsing behavior, pays 0.3 yuan for one collection behavior, pays 0.5 yuan for one purchase addition, and pays 1 yuan for one order placing behavior on a commodity a to be recommended, and a user browsing behavior predicted value, a user collection behavior predicted value, a user purchase addition behavior predicted value, and a user order placing behavior predicted value output by the commodity recommendation model according to the embodiment of the present invention may be as shown in table 4, and respectively represent a probability that the target user browses the commodity a to be recommended, a probability of collecting the commodity a to be recommended, a probability of purchasing the commodity a to be recommended, and a probability of placing the commodity a to be recommended, so that an expected benefit of recommending the commodity a to be recommended to the target user is:
0.2*0.3+0.3*0.1+0.5*0.05+1*0.01＝0.125；
and sequentially processing other commodities to be recommended according to the steps, so that a plurality of expected benefits, namely a plurality of weighted sums can be obtained, and the commodities to be recommended can be recommended to the target user according to the weighted sums.
The embodiment of the invention can sort the commodities to be recommended based on the weighted sum; and selecting a preset number of the commodities to be recommended from the sorted commodities to be recommended to recommend to the target user.
For example, 100 weighted sums can be calculated, and the 100 items to be recommended are sorted (from large to small) based on the 100 weighted sums, where the weighted sums correspond to the items to be recommended one by one, as shown in table 5 below.
| Serial number | Weighted sum | Merchandise to be recommended |
| 1 | 0.13 | A |
| 2 | 0.125 | B |
| …… | …… | …… |
TABLE 5
As can be seen from table 5, from the sorted commodities to be recommended, the commodities to be recommended (the commodity a to be recommended, the commodity B to be recommended, and … …) ranked in the top ten may be selected and recommended to the target user.
Through the above description of the technical scheme provided by the embodiment of the invention, the historical behavior characteristics, the user attributes and the commodity query text corresponding to the target user are obtained, the commodity attributes and the commodity identifications corresponding to any commodity to be recommended are spliced and input to the commodity recommendation model, and the corresponding user behavior predicted value can be output, so that the commodity to be recommended is recommended to the target user based on the user behavior predicted value, wherein the historical behavior characteristics comprise historical browsing behavior characteristics, historical collection behavior characteristics, historical purchase adding behavior characteristics and historical order placing behavior characteristics. Therefore, the user behavior predicted value is predicted through the historical browsing behavior characteristic, the historical collection behavior characteristic, the historical purchase behavior characteristic, the historical ordering behavior characteristic, the user attribute and the commodity query text corresponding to the target user, and the commodity attribute and the commodity identification corresponding to the commodity to be recommended, the commodity to be recommended is recommended to the target user based on the user behavior predicted value, and the click rate of commodity recommendation can be effectively improved.
As shown in fig. 2, an implementation flow diagram for obtaining a commodity recommendation model according to an embodiment of the present invention is provided, and the method specifically includes the following steps:
s201, obtaining a training sample, wherein the training sample comprises: historical behavior characteristics, user attributes and commodity query texts corresponding to the user, and commodity attributes and commodity identifications corresponding to the commodities;
in the embodiment of the invention, the display logs of the commodities and the behavior interaction logs of browsing, collecting, purchasing, ordering and the like of the user can be obtained.
And filtering the commodities in the commodity display logs based on the client exposure times or the display time, and filtering out commodities which are not displayed in the user visual area (namely the commodities with the display time of 0 second), or filtering out commodities with the exposure times not greater than a preset threshold value.
And performing anti-cheating filtering on behaviors of browsing, collecting, purchasing, ordering and the like of the user in the behavior interaction log, for example, filtering out behaviors that the browsing time of the user does not exceed 3 seconds.
Through the data cleaning operation, the commodity attribute and the commodity identification corresponding to the commodity can be extracted from the display log of the commodity, the historical behavior characteristic, the user attribute and the commodity query text corresponding to the user are extracted from the behavior interaction log, the historical behavior characteristic, the user attribute and the commodity query text are spliced, the user ID is used as a KEY, and the KEY is stored in the structured historical behavior characteristic index table, so that a plurality of training samples can be formed.
The non-digital information such as the commodity query text, the commodity name and the like can be subjected to word segmentation and mapped into a fixed-length ID sequence, which is not limited in the embodiment of the invention.
The training samples in the embodiment of the present invention include a positive sample and a negative sample, where the positive sample may be understood as a behavior that a user browses, collects, purchases, places an order, and the like, for a certain commodity, and the negative sample may be understood as a behavior that the user exposes to a certain commodity, and the user does not browse, collect, purchase, places an order, and the like.
For the training sample in the embodiment of the present invention, one bit of information may be used to indicate the task category of the current sample, for example, 1 indicates a browsing task, 2 indicates a collecting task, 3 indicates an ordering task, and 4 indicates an ordering task, which are respectively used for training different classifiers and respectively indicate four different tasks, such as browsing, collecting, ordering, and ordering.
In addition, for the training sample of the embodiment of the present invention, a part of the training sample may be divided for testing the sample, and the training sample is used for testing the commodity recommendation model.
S202, training the commodity recommendation initial model based on the training samples to obtain the commodity recommendation model.
The commodity recommendation model is obtained by performing (supervised) training on the commodity recommendation initial model based on the training samples and continuously updating parameters in the commodity recommendation initial model in an iterative manner until preset requirements are met.
In the embodiment of the present invention, when the number of iterations reaches the threshold value or the value of the evaluation function such as Logloss is no longer reduced, it is considered that the preset requirement is satisfied, and the model training may be stopped, which is not limited in the embodiment of the present invention.
In the embodiment of the present invention, the internal structure of the initial model for recommending commodities may be roughly divided into an input feature layer, a feature embedding layer, a feature extraction layer, and a multitask recommendation layer, as shown in fig. 3:
1. an input feature layer comprising: similar to the above, the user attribute, the commodity attribute, the historical behavior feature, the commodity identifier, and the commodity query text are not described in detail herein;
2. the characteristic embedding layer is used for carrying out low-dimensional embedding mapping on the high-dimensional discrete input characteristics provided by the characteristic input layer, wherein the discrete input provided by the characteristic input layer comprises two types of forms:
a single discrete ID such as a current product identifier (e.g., a product name) is subjected to embedding (an embedded layer for mapping a sparse high-dimensional discrete vector to a low-dimensional dense vector) mapping to obtain a one-dimensional dense vector of 1 × D, where D is the length of the embedding mapping and is usually tens or hundreds;
in another category, such as a commodity query text and historical behavior features, a one-dimensional embedding vector is obtained for each term after the commodity query text is subjected to word segmentation, so that finally the query text can obtain a two-dimensional vector of L1 × D, wherein L1 represents the number of terms after the query text is subjected to word segmentation, and a two-dimensional vector of L2 × D is also obtained for the historical behavior features, wherein L2 represents the length of a behavior sequence.
3. The feature extraction layer is used for performing high-order feature extraction on feature vectors mapped by the feature embedding layer and continuous value features such as user ages, commodity prices and the like input by the feature input layer;
the method comprises the following steps that (1) non-time sequence behavior information is extracted by using average pooling (MeanPooling) on vectors obtained by mapping historical behavior features of a feature embedding layer, and LSTM (Long Short-Term Memory, a Long Short-Term Memory network, a time cycle neural network, which is specially designed for solving the Long-Term dependence problem of a general RNN (recurrent neural network), and all RNNs have a chain type form of a repeated neural network module) are used for extracting time sequence behavior information of front and back dependence;
for the vectors obtained by mapping the commodity query text by the feature embedding layer, firstly, the vectors are reduced into two one-dimensional embedding vectors by using LSTM and average pooling respectively to represent the original text vectors, and then the semantic similarity between the vectors and the commodity identification (such as commodity name) is calculated on a model structure respectively, except that the semantic similarity between the query text and the commodity name is calculated by commonly adopted vector multiplication, the semantic similarity between the query text and the commodity name is fully mined by combining the square of the vector difference between the query text and the commodity name and the sum of the squares of the vector differences on the basis of the semantic similarity;
in order to further extract high-order information of input features of the model, all features are spliced and then connected with two layers of fully-connected neural networks, wherein a relu (Rectified Linear Unit, also called a corrected Linear Unit, is a commonly used activation function in the artificial neural network) is used as an activation function, and a dropout layer with a dropout rate of 0.5 is connected in front of each layer of fully-connected neural network input layer.
4. And the multi-task recommendation layer is used for realizing a multi-task recommendation model shared by the feature extraction layer, 4 independent classifiers are respectively connected behind the high-order feature layer obtained by the feature extraction layer to respectively represent four different tasks such as browsing, collecting, purchasing, ordering and the like, each classification layer uses sigmoid as an activation function, and the four classification layers share the same feature extraction layer.
Corresponding to the above method embodiment, an embodiment of the present invention further provides a commodity recommending apparatus, as shown in fig. 4, the apparatus may include: the system comprises an acquisition module 410, a commodity determination module 420, a splicing module 430, a predicted value output module 440 and a commodity recommendation module 450.
The obtaining module 410 is configured to obtain historical behavior features, user attributes and a commodity query text corresponding to a target user, where the historical behavior features include historical browsing behavior features, historical collection behavior features, historical purchase behavior features and historical order placing behavior features;
a goods determining module 420, configured to determine goods to be recommended;
the splicing module 430 is configured to splice, for any commodity attribute and commodity identifier corresponding to a commodity to be recommended, the historical behavior feature, the user attribute, and the commodity query text;
the predicted value output module 440 is configured to input the splicing result to the commodity recommendation model, and output a user behavior predicted value corresponding to the splicing result;
and the commodity recommending module 450 is configured to recommend the commodity to be recommended to the target user based on the user behavior prediction value.
In a specific implementation manner of the embodiment of the present invention, the obtaining module 410 is specifically configured to:
under the condition of receiving a commodity recommendation request, determining a target user triggering the commodity recommendation request;
and acquiring historical behavior characteristics, user attributes and commodity query texts corresponding to the target user.
In a specific implementation manner of the embodiment of the present invention, the commodity determining module 420 is specifically configured to:
and performing multi-path recall on the commodities according to the historical behavior characteristics and the user attributes to determine the commodities to be recommended.
In a specific implementation manner of the embodiment of the present invention, the user behavior prediction value includes:
the user order taking behavior prediction value is obtained by the user order taking behavior prediction value.
In a specific implementation manner of the embodiment of the present invention, the commodity recommending module 450 includes:
the weighting and calculating sub-module 451 is used for calculating the weighted sum of the predicted values of the user browsing behaviors, the user collecting behaviors, the user purchasing behavior and the user ordering behavior;
and the commodity recommending submodule 452 is configured to recommend the commodity to be recommended to the target user based on the weighted sum.
In a specific implementation manner of the embodiment of the present invention, the commodity recommendation sub-module 452 is specifically configured to:
sorting the commodities to be recommended based on the weighted sum;
and selecting a preset number of the commodities to be recommended from the sorted commodities to be recommended to recommend to the target user.
In a specific implementation manner of the embodiment of the present invention, the apparatus further includes:
a model training module 460, configured to obtain training samples, where the training samples include: historical behavior characteristics, user attributes and commodity query texts corresponding to the user, and commodity attributes and commodity identifications corresponding to the commodities;
and training the commodity recommendation initial model based on the training sample to obtain the commodity recommendation model.
An embodiment of the present invention further provides an electronic device, as shown in fig. 5, including a processor 51, a communication interface 52, a memory 53 and a communication bus 54, where the processor 51, the communication interface 52, and the memory 53 complete mutual communication through the communication bus 54,
a memory 53 for storing a computer program;
the processor 51 is configured to implement the following steps when executing the program stored in the memory 53:
acquiring historical behavior characteristics, user attributes and commodity query texts corresponding to target users; determining commodities to be recommended, wherein the historical behavior characteristics comprise historical browsing behavior characteristics, historical collection behavior characteristics, historical purchase behavior characteristics and historical ordering behavior characteristics; splicing the commodity attribute and the commodity identification corresponding to any commodity to be recommended with the historical behavior characteristics, the user attribute and the commodity query text; inputting the splicing result into a commodity recommendation model, and outputting a user behavior predicted value corresponding to the splicing result; and recommending the commodity to be recommended to the target user based on the user behavior predicted value.
The communication bus mentioned in the electronic device may be a Peripheral Component Interconnect (PCI) bus, an Extended Industry Standard Architecture (EISA) bus, or the like. The communication bus may be divided into an address bus, a data bus, a control bus, etc. For ease of illustration, only one thick line is shown, but this does not mean that there is only one bus or one type of bus.
The communication interface is used for communication between the electronic equipment and other equipment.
The Memory may include a Random Access Memory (RAM) or a non-volatile Memory (non-volatile Memory), such as at least one disk Memory. Optionally, the memory may also be at least one memory device located remotely from the processor.
The Processor may be a general-purpose Processor, and includes a Central Processing Unit (CPU), a Network Processor (NP), and the like; the device can also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other Programmable logic device, a discrete Gate or transistor logic device, or a discrete hardware component.
In still another embodiment of the present invention, a storage medium is further provided, where instructions are stored, and when the instructions are executed on a computer, the instructions cause the computer to execute the product recommendation method in any one of the above embodiments.
In yet another embodiment of the present invention, there is also provided a computer program product containing instructions which, when run on a computer, cause the computer to perform the method of recommending items as described in any of the above embodiments.
In the above embodiments, the implementation may be wholly or partially realized by software, hardware, firmware, or any combination thereof. When implemented in software, may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When loaded and executed on a computer, cause the processes or functions described in accordance with the embodiments of the invention to occur, in whole or in part. The computer may be a general purpose computer, a special purpose computer, a network of computers, or other programmable device. The computer instructions may be stored on a storage medium or transmitted from one storage medium to another, for example, from one website, computer, server, or data center to another website, computer, server, or data center via wire (e.g., coaxial cable, fiber optic, Digital Subscriber Line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.). The storage medium may be any available medium that can be accessed by a computer or a data storage device including one or more available media integrated servers, data centers, and the like. The usable medium may be a magnetic medium (e.g., floppy Disk, hard Disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., Solid State Disk (SSD)), among others.
It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
All the embodiments in the present specification are described in a related manner, and the same and similar parts among the embodiments may be referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, for the system embodiment, since it is substantially similar to the method embodiment, the description is simple, and for the relevant points, reference may be made to the partial description of the method embodiment.
The above description is only for the preferred embodiment of the present invention, and is not intended to limit the scope of the present invention. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention shall fall within the protection scope of the present invention.
Claims (10)
1. A method for recommending an article, the method comprising:
acquiring historical behavior characteristics, user attributes and commodity query texts corresponding to target users, wherein the historical behavior characteristics comprise historical browsing behavior characteristics, historical collection behavior characteristics, historical purchase adding behavior characteristics and historical order placing behavior characteristics;
determining a commodity to be recommended;
splicing the commodity attribute and the commodity identification corresponding to any commodity to be recommended with the historical behavior characteristics, the user attribute and the commodity query text;
inputting the splicing result into a commodity recommendation model, and outputting a user behavior predicted value corresponding to the splicing result;
and recommending the commodity to be recommended to the target user based on the user behavior predicted value.
2. The method of claim 1, wherein the obtaining of the historical behavior features, the user attributes and the commodity query text corresponding to the target user comprises:
under the condition of receiving a commodity recommendation request, determining a target user triggering the commodity recommendation request;
and acquiring historical behavior characteristics, user attributes and commodity query texts corresponding to the target user.
3. The method of claim 1, wherein the determining the item to be recommended comprises:
and performing multi-path recall on the commodities according to the historical behavior characteristics and the user attributes to determine the commodities to be recommended.
4. The method of claim 1, wherein the user behavior prediction value comprises:
the user order taking behavior prediction value is obtained by the user order taking behavior prediction value.
5. The method according to claim 4, wherein recommending the commodity to be recommended to the target user based on the user behavior prediction value comprises:
calculating the weighted sum of the predicted value of the user browsing behavior, the predicted value of the user collecting behavior, the predicted value of the user purchasing behavior and the predicted value of the user ordering behavior;
and recommending the commodities to be recommended to the target user based on the weighted sum.
6. The method of claim 5, wherein the recommending the item to be recommended to the target user based on the weighted sum comprises:
sorting the commodities to be recommended based on the weighted sum;
and selecting a preset number of the commodities to be recommended from the sorted commodities to be recommended to recommend to the target user.
7. The method according to any one of claims 1 to 6, wherein the commodity recommendation model is derived by:
obtaining training samples, wherein the training samples comprise: historical behavior characteristics, user attributes and commodity query texts corresponding to the user, and commodity attributes and commodity identifications corresponding to the commodities;
and training the commodity recommendation initial model based on the training sample to obtain the commodity recommendation model.
8. An article recommendation device, the device comprising:
the system comprises an acquisition module, a display module and a display module, wherein the acquisition module is used for acquiring historical behavior characteristics, user attributes and commodity query texts corresponding to target users, wherein the historical behavior characteristics comprise historical browsing behavior characteristics, historical collection behavior characteristics, historical purchase behavior characteristics and historical ordering behavior characteristics;
the commodity determining module is used for determining commodities to be recommended;
the splicing module is used for splicing the commodity attribute and the commodity identification corresponding to any commodity to be recommended with the historical behavior characteristics, the user attribute and the commodity query text;
the predicted value output module is used for inputting the splicing result into the commodity recommendation model and outputting the user behavior predicted value corresponding to the splicing result;
and the commodity recommending module is used for recommending the commodity to be recommended to the target user based on the user behavior predicted value.
9. An electronic device is characterized by comprising a processor, a communication interface, a memory and a communication bus, wherein the processor and the communication interface are used for realizing mutual communication by the memory through the communication bus;
a memory for storing a computer program;
a processor for implementing the method steps of any one of claims 1 to 7 when executing a program stored in the memory.
10. A storage medium on which a computer program is stored which, when being executed by a processor, carries out the method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010328864.1A CN111523976B (en) | 2020-04-23 | 2020-04-23 | Commodity recommendation method and device, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010328864.1A CN111523976B (en) | 2020-04-23 | 2020-04-23 | Commodity recommendation method and device, electronic equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111523976A true CN111523976A (en) | 2020-08-11 |
| CN111523976B CN111523976B (en) | 2023-12-08 |
Family
ID=71904550
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010328864.1A Active CN111523976B (en) | 2020-04-23 | 2020-04-23 | Commodity recommendation method and device, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111523976B (en) |
Cited By (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112288390A (en) * | 2020-10-26 | 2021-01-29 | 北京爱奇艺科技有限公司 | Client use state detection method, client use state detection device, message pushing method, message pushing device and related equipment |
| CN112287232A (en) * | 2020-11-09 | 2021-01-29 | 北京京东拓先科技有限公司 | Method and device for generating recommendation information |
| CN112651778A (en) * | 2020-12-25 | 2021-04-13 | 平安科技（深圳）有限公司 | User behavior prediction method, device, equipment and medium |
| CN112785390A (en) * | 2021-02-02 | 2021-05-11 | 微民保险代理有限公司 | Recommendation processing method and device, terminal device and storage medium |
| CN112882955A (en) * | 2021-03-30 | 2021-06-01 | 中信银行股份有限公司 | Test case recommendation method and device and electronic equipment |
| CN112950261A (en) * | 2021-02-20 | 2021-06-11 | 北京读我网络技术有限公司 | Method and system for determining user value |
| CN113077321A (en) * | 2021-04-21 | 2021-07-06 | 北京沃东天骏信息技术有限公司 | Article recommendation method and device, electronic equipment and storage medium |
| CN113536138A (en) * | 2021-08-16 | 2021-10-22 | 北京爱奇艺科技有限公司 | Network resource recommendation method and device, electronic equipment and readable storage medium |
| CN113610580A (en) * | 2021-08-10 | 2021-11-05 | 平安科技（深圳）有限公司 | Product recommendation method and device, electronic equipment and readable storage medium |
| CN113744015A (en) * | 2020-10-20 | 2021-12-03 | 北京沃东天骏信息技术有限公司 | Sorting method, device, equipment and computer storage medium |
| CN113744019A (en) * | 2021-01-12 | 2021-12-03 | 北京沃东天骏信息技术有限公司 | Commodity recommendation method, commodity recommendation device, commodity recommendation equipment and storage medium |
| CN113761378A (en) * | 2021-09-14 | 2021-12-07 | 上海任意门科技有限公司 | Content ordering method, computing device and computer-readable storage medium |
| CN113781149A (en) * | 2021-01-22 | 2021-12-10 | 北京沃东天骏信息技术有限公司 | Information recommendation method and device, computer-readable storage medium and electronic equipment |
| CN113793180A (en) * | 2021-09-15 | 2021-12-14 | 北京沃东天骏信息技术有限公司 | User preference analysis method, device, equipment and computer storage medium |
| CN113869973A (en) * | 2021-09-15 | 2021-12-31 | 中信建投证券股份有限公司 | Product recommendation method, product recommendation system, and computer-readable storage medium |
| CN114493759A (en) * | 2021-12-30 | 2022-05-13 | 胜斗士(上海)科技技术发展有限公司 | Method for applying recommendation policy to target user |
| CN114741605A (en) * | 2022-04-26 | 2022-07-12 | 泰康保险集团股份有限公司 | Method and device for recommending annuity products, electronic equipment and readable medium |
| CN115248803A (en) * | 2022-09-22 | 2022-10-28 | 天津联想协同科技有限公司 | Collection method and device suitable for network disk file, network disk and storage medium |
| CN117273865A (en) * | 2023-11-14 | 2023-12-22 | 深圳市灵智数字科技有限公司 | Commodity recommendation method and device, electronic equipment and storage medium |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070033098A1 (en) * | 2005-08-05 | 2007-02-08 | International Business Machines Corporation | Method, system and storage medium for creating sales recommendations |
| US20130297381A1 (en) * | 2012-05-07 | 2013-11-07 | Logistimo India Private Limited | Method and system for managing a low-resource supply chain |
| 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 |
| US20170024724A1 (en) * | 2015-07-24 | 2017-01-26 | Sk Planet Co., Ltd. | System for recommending optimal card, apparatus for recommending optimal card and method for the same |
| US20190188771A1 (en) * | 2017-12-15 | 2019-06-20 | Capital One Services, Llc | Systems and Methods for Transaction-Based Real Time Pre-Intent Recommendations for a Sequential Purchase |
| CN110162706A (en) * | 2019-05-22 | 2019-08-23 | 南京邮电大学 | A kind of personalized recommendation method and system based on interaction data cluster |
| CN110555717A (en) * | 2019-07-29 | 2019-12-10 | 华南理工大学 | method for mining potential purchased goods and categories of users based on user behavior characteristics |
| CN110880127A (en) * | 2019-10-24 | 2020-03-13 | 北京三快在线科技有限公司 | Consumption level prediction method and device, electronic equipment and storage medium |
| CN111027714A (en) * | 2019-12-11 | 2020-04-17 | 腾讯科技（深圳）有限公司 | Artificial intelligence-based object recommendation model training method, recommendation method and device |
-
2020
- 2020-04-23 CN CN202010328864.1A patent/CN111523976B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070033098A1 (en) * | 2005-08-05 | 2007-02-08 | International Business Machines Corporation | Method, system and storage medium for creating sales recommendations |
| US20130297381A1 (en) * | 2012-05-07 | 2013-11-07 | Logistimo India Private Limited | Method and system for managing a low-resource supply chain |
| 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 |
| US20170024724A1 (en) * | 2015-07-24 | 2017-01-26 | Sk Planet Co., Ltd. | System for recommending optimal card, apparatus for recommending optimal card and method for the same |
| US20190188771A1 (en) * | 2017-12-15 | 2019-06-20 | Capital One Services, Llc | Systems and Methods for Transaction-Based Real Time Pre-Intent Recommendations for a Sequential Purchase |
| CN110162706A (en) * | 2019-05-22 | 2019-08-23 | 南京邮电大学 | A kind of personalized recommendation method and system based on interaction data cluster |
| CN110555717A (en) * | 2019-07-29 | 2019-12-10 | 华南理工大学 | method for mining potential purchased goods and categories of users based on user behavior characteristics |
| CN110880127A (en) * | 2019-10-24 | 2020-03-13 | 北京三快在线科技有限公司 | Consumption level prediction method and device, electronic equipment and storage medium |
| CN111027714A (en) * | 2019-12-11 | 2020-04-17 | 腾讯科技（深圳）有限公司 | Artificial intelligence-based object recommendation model training method, recommendation method and device |
Cited By (25)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113744015A (en) * | 2020-10-20 | 2021-12-03 | 北京沃东天骏信息技术有限公司 | Sorting method, device, equipment and computer storage medium |
| CN112288390A (en) * | 2020-10-26 | 2021-01-29 | 北京爱奇艺科技有限公司 | Client use state detection method, client use state detection device, message pushing method, message pushing device and related equipment |
| WO2022095893A1 (en) * | 2020-11-09 | 2022-05-12 | 北京京东拓先科技有限公司 | Method and apparatus for generating recommendation information |
| CN112287232B (en) * | 2020-11-09 | 2024-04-16 | 北京京东拓先科技有限公司 | Method and device for generating recommendation information |
| CN112287232A (en) * | 2020-11-09 | 2021-01-29 | 北京京东拓先科技有限公司 | Method and device for generating recommendation information |
| CN112651778A (en) * | 2020-12-25 | 2021-04-13 | 平安科技（深圳）有限公司 | User behavior prediction method, device, equipment and medium |
| CN113744019A (en) * | 2021-01-12 | 2021-12-03 | 北京沃东天骏信息技术有限公司 | Commodity recommendation method, commodity recommendation device, commodity recommendation equipment and storage medium |
| CN113781149B (en) * | 2021-01-22 | 2024-07-16 | 北京沃东天骏信息技术有限公司 | Information recommendation method and device, computer readable storage medium and electronic equipment |
| CN113781149A (en) * | 2021-01-22 | 2021-12-10 | 北京沃东天骏信息技术有限公司 | Information recommendation method and device, computer-readable storage medium and electronic equipment |
| CN112785390A (en) * | 2021-02-02 | 2021-05-11 | 微民保险代理有限公司 | Recommendation processing method and device, terminal device and storage medium |
| CN112785390B (en) * | 2021-02-02 | 2024-02-09 | 微民保险代理有限公司 | Recommendation processing method, device, terminal equipment and storage medium |
| CN112950261A (en) * | 2021-02-20 | 2021-06-11 | 北京读我网络技术有限公司 | Method and system for determining user value |
| CN112882955A (en) * | 2021-03-30 | 2021-06-01 | 中信银行股份有限公司 | Test case recommendation method and device and electronic equipment |
| CN113077321A (en) * | 2021-04-21 | 2021-07-06 | 北京沃东天骏信息技术有限公司 | Article recommendation method and device, electronic equipment and storage medium |
| CN113610580B (en) * | 2021-08-10 | 2023-09-19 | 平安科技（深圳）有限公司 | Product recommendation method and device, electronic equipment and readable storage medium |
| CN113610580A (en) * | 2021-08-10 | 2021-11-05 | 平安科技（深圳）有限公司 | Product recommendation method and device, electronic equipment and readable storage medium |
| CN113536138A (en) * | 2021-08-16 | 2021-10-22 | 北京爱奇艺科技有限公司 | Network resource recommendation method and device, electronic equipment and readable storage medium |
| CN113761378A (en) * | 2021-09-14 | 2021-12-07 | 上海任意门科技有限公司 | Content ordering method, computing device and computer-readable storage medium |
| CN113869973A (en) * | 2021-09-15 | 2021-12-31 | 中信建投证券股份有限公司 | Product recommendation method, product recommendation system, and computer-readable storage medium |
| CN113793180A (en) * | 2021-09-15 | 2021-12-14 | 北京沃东天骏信息技术有限公司 | User preference analysis method, device, equipment and computer storage medium |
| CN114493759A (en) * | 2021-12-30 | 2022-05-13 | 胜斗士(上海)科技技术发展有限公司 | Method for applying recommendation policy to target user |
| CN114741605A (en) * | 2022-04-26 | 2022-07-12 | 泰康保险集团股份有限公司 | Method and device for recommending annuity products, electronic equipment and readable medium |
| CN115248803A (en) * | 2022-09-22 | 2022-10-28 | 天津联想协同科技有限公司 | Collection method and device suitable for network disk file, network disk and storage medium |
| CN115248803B (en) * | 2022-09-22 | 2023-02-17 | 天津联想协同科技有限公司 | Collection method and device suitable for network disk file, network disk and storage medium |
| CN117273865A (en) * | 2023-11-14 | 2023-12-22 | 深圳市灵智数字科技有限公司 | Commodity recommendation method and device, electronic equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111523976B (en) | 2023-12-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111523976B (en) | Commodity recommendation method and device, electronic equipment and storage medium | |
| CN111784455B (en) | Article recommendation method and recommendation equipment | |
| CN108121737B (en) | Method, device and system for generating business object attribute identifier | |
| Malik et al. | EPR-ML: E-Commerce Product Recommendation Using NLP and Machine Learning Algorithm | |
| CN109492180A (en) | Resource recommendation method, device, computer equipment and computer readable storage medium | |
| CN113077317B (en) | Article recommendation method, device, equipment and storage medium based on user data | |
| CN107220852A (en) | Method, device and server for determining target recommended user | |
| CN110827112B (en) | Deep learning commodity recommendation method and device, computer equipment and storage medium | |
| CN110647696B (en) | Business object sorting method and device | |
| CN113763095B (en) | Information recommendation method and device and model training method and device | |
| CN111666275B (en) | Data processing method and device, electronic equipment and storage medium | |
| CN111626832A (en) | Product recommendation method and device and computer equipment | |
| CN111340566B (en) | Commodity classification method and device, electronic equipment and storage medium | |
| CN113837842A (en) | Commodity recommendation method and equipment based on user behavior data | |
| CN113781149B (en) | Information recommendation method and device, computer readable storage medium and electronic equipment | |
| CN111782927A (en) | Article recommendation method and device, computer storage medium | |
| CN110930223A (en) | Recommendation recall method, device and storage medium based on field-aware factorization machine | |
| CN112131491B (en) | Hierarchical ordering method, computing device and computer readable storage medium | |
| CN111680213A (en) | Information recommendation method, data processing method and device | |
| CN118193806A (en) | Target retrieval method, target retrieval device, electronic equipment and storage medium | |
| CN114429384B (en) | Intelligent product recommendation method and system based on e-commerce platform | |
| CN110827044A (en) | Method and device for extracting user interest mode | |
| CN111639989B (en) | Commodity recommendation method and readable storage medium | |
| CN111768262B (en) | Information recommendation method and device, electronic equipment and storage medium | |
| CN113761002A (en) | Information pushing method, device, equipment and computer readable storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| CB02 | Change of applicant information | ||
| CB02 | Change of applicant information |
Address after: Room 221, 2 / F, block C, 18 Kechuang 11th Street, Daxing District, Beijing, 100176 Applicant after: Jingdong Technology Holding Co.,Ltd. Address before: Room 221, 2 / F, block C, 18 Kechuang 11th Street, Daxing District, Beijing, 100176 Applicant before: Jingdong Digital Technology Holding Co.,Ltd. Address after: Room 221, 2 / F, block C, 18 Kechuang 11th Street, Daxing District, Beijing, 100176 Applicant after: Jingdong Digital Technology Holding Co.,Ltd. Address before: Room 221, 2 / F, block C, 18 Kechuang 11th Street, Daxing District, Beijing, 100176 Applicant before: JINGDONG DIGITAL TECHNOLOGY HOLDINGS Co.,Ltd. |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |