CN112001776A - Service information pushing method and device, storage medium and electronic equipment - Google Patents
Service information pushing method and device, storage medium and electronic equipment Download PDFInfo
- Publication number
- CN112001776A CN112001776A CN202010854690.2A CN202010854690A CN112001776A CN 112001776 A CN112001776 A CN 112001776A CN 202010854690 A CN202010854690 A CN 202010854690A CN 112001776 A CN112001776 A CN 112001776A
- Authority
- CN
- China
- Prior art keywords
- commodity
- commodities
- predicted
- pushing
- weight
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 47
- 230000006399 behavior Effects 0.000 claims abstract description 66
- 238000003062 neural network model Methods 0.000 claims abstract description 18
- 238000012545 processing Methods 0.000 claims description 17
- 230000006870 function Effects 0.000 claims description 13
- 238000012549 training Methods 0.000 claims description 6
- 238000004590 computer program Methods 0.000 claims description 5
- 230000000875 corresponding effect Effects 0.000 description 16
- 238000010586 diagram Methods 0.000 description 9
- 239000000126 substance Substances 0.000 description 5
- 238000004891 communication Methods 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- 230000009471 action Effects 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 238000012512 characterization method Methods 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 230000003111 delayed effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000005484 gravity Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 239000002184 metal Substances 0.000 description 1
- 230000009022 nonlinear effect Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 238000012216 screening Methods 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
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/047—Probabilistic or stochastic networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0623—Item investigation
- G06Q30/0625—Directed, with specific intent or strategy
- G06Q30/0629—Directed, with specific intent or strategy for generating comparisons
Abstract
The application provides a service information pushing method and device, a storage medium and electronic equipment. Generating a characteristic sequence according to known commodities and time stamps in the historical behaviors of the user; taking the characteristic sequence as the input of a neural network model to obtain the weight of the predicted commodity and each known commodity in the characteristic sequence relative to the predicted commodity; determining related commodities according to the weight and the similarity of each known commodity; pushing the predicted commodities to a client, and transmitting a pushing explanation to the client, wherein the pushing explanation comprises related commodities; when the consumer refers to the predicted commodity through the client, the consumer can refer to the related commodity in the pushing explanation at the same time, so that the reason for receiving the pushing of the predicted commodity is clear. Therefore, the rejection of the user to the recommended subject matter is reduced, and the conditions of distrust and unwilling to click to know are reduced. The doubts of the consumers on the recommended objects are effectively eliminated, so that the consumers trust the recommendation system, and the frequency of the recommendation system used by the users is improved.
Description
Technical Field
The application relates to the field of internet, in particular to a service information pushing method and device, a storage medium and electronic equipment.
Background
With the development of the internet, various websites flood the lives of people. The website type is five-flower eight, and the commodities in the website are rich and diverse. The consumers need to spend a long time searching for interested commodities in the shopping websites, the time of the consumers is delayed, and the shopping experience of the consumers is influenced.
The prior art provides a recommendation system, which aims to recommend a target object which may be interested to a user, so as to save the time of the user and improve the satisfaction of the user. However, most users at present are relatively resistant to recommended subject matters, do not trust and do not wish to click to know. How to effectively eliminate the doubt of the consumer about the recommended subject matter is a problem to be solved urgently by those skilled in the art.
Disclosure of Invention
The present application aims to provide a service information pushing method, a service information pushing device, a storage medium and an electronic device, so as to solve the above problems.
In order to achieve the above purpose, the embodiments of the present application employ the following technical solutions:
in a first aspect, an embodiment of the present application provides a service information pushing method, where the method includes:
generating a characteristic sequence according to known commodities and time stamps in the historical behaviors of the user;
the historical behaviors comprise a commodity clicking behavior, a commodity collecting behavior, a commodity purchasing behavior and a behavior of adding commodities into a shopping cart, the characteristic sequence is a splicing sequence of known commodities and time intervals in the historical behaviors, and the time intervals are time stamps of the historical behaviors and intervals of predicted time;
taking the characteristic sequence as the input of a neural network model to obtain the weight of a predicted commodity and each known commodity in the characteristic sequence relative to the predicted commodity;
determining related commodities according to the weight and the similarity of each known commodity, wherein the similarity is the similarity between the known commodity and the predicted commodity, and the related commodities are commodities related to the predicted commodity;
and pushing the predicted commodity to a client, and transmitting a pushing explanation to the client, wherein the pushing explanation comprises the associated commodity.
In a second aspect, an embodiment of the present application provides a service information pushing method, where the service information pushing apparatus includes:
the processing unit is used for generating a characteristic sequence according to known commodities and time stamps in the historical behaviors of the user;
the historical behaviors comprise a commodity clicking behavior, a commodity collecting behavior, a commodity purchasing behavior and a behavior of adding commodities into a shopping cart, the characteristic sequence is a splicing sequence of known commodities and time intervals in the historical behaviors, and the time intervals are time stamps of the historical behaviors and intervals of predicted time;
the processing unit is further used for inputting the characteristic sequence into a neural network model so as to obtain a predicted commodity and the weight of each known commodity in the characteristic sequence relative to the predicted commodity; the system is also used for determining related commodities according to the weight and the similarity of each known commodity, wherein the similarity is the similarity between the known commodity and the predicted commodity, and the related commodities are commodities related to the predicted commodity;
and the pushing unit is used for pushing the predicted commodity to a client and transmitting a pushing explanation to the client, wherein the pushing explanation comprises the related commodity.
In a third aspect, the present application provides a storage medium, on which a computer program is stored, and the computer program, when executed by a processor, implements the method described above.
In a fourth aspect, an embodiment of the present application provides an electronic device, including: a processor and memory for storing one or more programs; the one or more programs, when executed by the processor, implement the methods described above.
Compared with the prior art, the service information pushing method, the service information pushing device, the storage medium and the electronic equipment provided by the embodiment of the application have the beneficial effects that: generating a characteristic sequence according to known commodities and time stamps in the historical behaviors of the user; taking the characteristic sequence as the input of a neural network model to obtain the weight of the predicted commodity and each known commodity in the characteristic sequence relative to the predicted commodity; determining related commodities according to the weight and the similarity of each known commodity; pushing the predicted commodities to a client, and transmitting a pushing explanation to the client, wherein the pushing explanation comprises related commodities; when the consumer refers to the predicted commodity through the client, the consumer can refer to the related commodity in the pushing explanation at the same time, so that the reason for receiving the pushing of the predicted commodity is clear. Therefore, the rejection of the user to the recommended subject matter is reduced, and the conditions of distrust and unwilling to click to know are reduced. The doubts of the consumers on the recommended objects are effectively eliminated, so that the consumers trust the recommendation system, and the frequency of the recommendation system used by the users is improved.
In order to make the aforementioned objects, features and advantages of the present application more comprehensible, preferred embodiments accompanied with figures are described in detail below.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are required to be used in the embodiments will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present application and therefore should not be considered as limiting the scope, and it will be apparent to those skilled in the art that other related drawings can be obtained from the drawings without inventive effort.
Fig. 1 is a schematic structural diagram of an electronic device according to an embodiment of the present disclosure;
fig. 2 is a schematic flowchart of a service information pushing method according to an embodiment of the present application;
fig. 3 is a schematic view of the substeps of S105 provided in the embodiment of the present application;
fig. 4 is another schematic flow chart of a service information pushing method according to an embodiment of the present application;
fig. 5 is a schematic unit diagram of a service information pushing apparatus according to an embodiment of the present application.
In the figure: 10-a processor; 11-a memory; 12-a bus; 13-a communication interface; 201-a processing unit; 202-push unit.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are some embodiments of the present application, but not all embodiments. The components of the embodiments of the present application, generally described and illustrated in the figures herein, can be arranged and designed in a wide variety of different configurations.
Thus, the following detailed description of the embodiments of the present application, presented in the accompanying drawings, is not intended to limit the scope of the claimed application, but is merely representative of selected embodiments of the application. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, it need not be further defined and explained in subsequent figures. Meanwhile, in the description of the present application, the terms "first", "second", and the like are used only for distinguishing the description, and are not to be construed as indicating or implying relative importance.
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.
In the description of the present application, it should be noted that the terms "upper", "lower", "inner", "outer", and the like indicate orientations or positional relationships based on orientations or positional relationships shown in the drawings or orientations or positional relationships conventionally found in use of products of the application, and are used only for convenience in describing the present application and for simplification of description, but do not indicate or imply that the referred devices or elements must have a specific orientation, be constructed in a specific orientation, and be operated, and thus should not be construed as limiting the present application.
In the description of the present application, it is also to be noted that, unless otherwise explicitly specified or limited, the terms "disposed" and "connected" are to be interpreted broadly, e.g., as being either fixedly connected, detachably connected, or integrally connected; can be mechanically or electrically connected; they may be connected directly or indirectly through intervening media, or they may be interconnected between two elements. The specific meaning of the above terms in the present application can be understood in a specific case by those of ordinary skill in the art.
Some embodiments of the present application will be described in detail below with reference to the accompanying drawings. The embodiments described below and the features of the embodiments can be combined with each other without conflict.
The prior art provides a recommendation system, which aims to recommend a target object which may be interested to a user, so as to save the time of the user and improve the satisfaction of the user. However, most users at present are relatively resistant to recommended subject matters, do not trust and do not wish to click to know. How to effectively eliminate the doubt of the consumer on the recommended subject matter, so that the consumer trusts the recommendation system, thereby improving the frequency of the recommendation system used by the user, is a problem to be solved by the technical staff in the field.
The embodiment of the application provides an electronic device which can be a server device. Please refer to fig. 1, a schematic structural diagram of an electronic device. The electronic device comprises a processor 10, a memory 11, a bus 12. The processor 10 and the memory 11 are connected by a bus 12, and the processor 10 is configured to execute an executable module, such as a computer program, stored in the memory 11.
The processor 10 may be an integrated circuit chip having signal processing capabilities. In implementation, the steps of the service information pushing method may be implemented by integrated logic circuits of hardware or instructions in the form of software in the processor 10. The Processor 10 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.
The Memory 11 may comprise a high-speed Random Access Memory (RAM) and may further comprise a non-volatile Memory (non-volatile Memory), such as at least one disk Memory.
The bus 12 may be an ISA (Industry Standard architecture) bus, a PCI (peripheral Component interconnect) bus, an EISA (extended Industry Standard architecture) bus, or the like. Only one bi-directional arrow is shown in fig. 1, but this does not indicate only one bus 12 or one type of bus 12.
The memory 11 is used for storing programs, such as programs corresponding to the service information pushing device. The service information pushing device includes at least one software function module which can be stored in the memory 11 in the form of software or firmware (firmware) or solidified in an Operating System (OS) of the electronic device. The processor 10 executes the program to implement the service information push method after receiving the execution instruction.
Possibly, the electronic device provided by the embodiment of the present application further includes a communication interface 13. The communication interface 13 is connected to the processor 10 via a bus. The electronic device may push a message to the client or receive request information of the client through the communication interface 13.
It should be understood that the structure shown in fig. 1 is merely a structural schematic diagram of a portion of an electronic device, which may also include more or fewer components than shown in fig. 1, or have a different configuration than shown in fig. 1. The components shown in fig. 1 may be implemented in hardware, software, or a combination thereof.
The service information pushing method provided in the embodiment of the present invention can be applied to, but is not limited to, the electronic device shown in fig. 1, and please refer to fig. 2:
and S101, generating a characteristic sequence according to the known commodities and the time stamp in the historical behaviors of the user.
The historical behaviors comprise behaviors of clicking commodities, collecting commodities, purchasing commodities and adding commodities into a shopping cart, the characteristic sequence is a splicing sequence of known commodities and time intervals in the historical behaviors, the time intervals are intervals of timestamps of the historical behaviors and predicted time, and the timestamps represent the occurrence time of the corresponding historical behaviors.
In particular, the consumer's interests may be consistent over a period of time. The goods in which the consumer is interested can be acquired through the known goods of the historical behaviors of the consumer and the time stamp, so that the appropriate goods can be recommended to the consumer.
And S104, taking the characteristic sequence as the input of the neural network model to obtain the weight of the predicted commodity and each known commodity in the characteristic sequence relative to the predicted commodity.
Specifically, the weights characterize the reference specific gravity of the corresponding known good at the time the predicted good was obtained. Possibly, the weight is related to the number of occurrences of a known good in the sequence of features, the number of occurrences of a good of the same type, and the corresponding order of timestamps.
And S105, determining related commodities according to the weight and the similarity of each known commodity, wherein the similarity is the similarity between the known commodity and the predicted commodity, and the related commodities are commodities related to the predicted commodity.
Possibly, according to the similarity between each known commodity and the predicted commodity and the weight of the known commodity to the preset commodity, the known commodity with the relevance degree greater than the relevance threshold value or the known commodity with the highest relevance degree is determined and serves as the relevant commodity.
S106, pushing the predicted commodities to the client, and transmitting a pushing explanation to the client, wherein the pushing explanation comprises the related commodities.
Specifically, the predicted commodity is pushed to the client, and meanwhile, the pushing explanation is transmitted to the client. Because the push explanation comprises the associated commodities, when the consumer refers to the predicted commodities through the client, the consumer can refer to the associated commodities in the push explanation at the same time, so that the reason for receiving the push of the predicted commodities is clear. Therefore, the rejection of the user to the recommended subject matter is reduced, and the conditions of distrust and unwilling to click to know are reduced. The doubts of the consumers on the recommended objects are effectively eliminated, so that the consumers trust the recommendation system, and the frequency of the recommendation system used by the users is improved.
To sum up, in the service information pushing method provided by the embodiment of the application, the feature sequence is generated according to the known goods and the time stamps in the historical behaviors of the user; taking the characteristic sequence as the input of a neural network model to obtain the weight of the predicted commodity and each known commodity in the characteristic sequence relative to the predicted commodity; determining related commodities according to the weight and the similarity of each known commodity; pushing the predicted commodities to a client, and transmitting a pushing explanation to the client, wherein the pushing explanation comprises related commodities; when the consumer refers to the predicted commodity through the client, the consumer can refer to the related commodity in the pushing explanation at the same time, so that the reason for receiving the pushing of the predicted commodity is clear. Therefore, the rejection of the user to the recommended subject matter is reduced, and the conditions of distrust and unwilling to click to know are reduced. The doubts of the consumers on the recommended objects are effectively eliminated, so that the consumers trust the recommendation system, and the frequency of the recommendation system used by the users is improved.
Regarding the feature sequence, the embodiments of the present application provide a possible implementation manner, please refer to the following.
Suppose that the space of the user is
Wherein the content of the first and second substances,
the goods space is
Wherein the content of the first and second substances,
the historical behavior sequence set of the user is C ═ S1，S2，…，SUAnd (c) the step of (c) in which,
wherein the content of the first and second substances,
a time stamp is represented which is a time stamp,
representing a user
The commodities corresponding to the behaviors performed at the time stamp are collected, and S is setuAre ordered chronologically, i.e.
Assume that the window length of the original input sequence is L, i.e. the goal is to record the historical behavior of the user:
in the case of (1), the prediction is at tL+1The recommended goods at the time. Here, the commodity and the time are processed separately. The processing of the time series is: extracting the time stamp in the input sequence, subtracting the predicted time to obtain a relative position (namely a time interval), and embedding to obtain the representation of the time sequence:
T＝[tL+1-t1，tL+1-t2，，tL+1-tL]·ET→[p1，p2，，pL]。
therefore, the temperature of the molten metal is controlled,
the processing of the user commodities comprises the following steps: the size of the commodity library is N, and each commodity corresponds to one-hot code (one-hot code or one-bit effective code). Converting the one-hot code corresponding to each commodity into a low-dimensional representation, wherein the one-hot code and the low-dimensional representation correspond to a mapping table: look up table
Extracting the commodities in the input sequence to obtain:
vi＝si·Einput。
wherein v isiA low dimensional representation, s, characterizing the ith good of the useriA one-hot code characterizing the ith good of the user.
Connecting the time series of characterizations with the low-dimensional characterization of the good to obtain:
wherein the content of the first and second substances,
characterizing an ith group of codes in a characteristic sequence corresponding to a user u;
representing a low-dimensional label corresponding to a commodity in the ith historical behavior of the user u;
and representing the time interval between the ith historical behavior of the user u and a preset time point.
It should be noted that the ith group code in the feature sequence corresponding to the user u is a splicing representation of the ith behavior of the user u and the corresponding time interval.
The obtained characteristic sequence is as follows: x ═ X1，x2，…，xL]。
Regarding the feature sequence, the embodiment of the present application further provides a possible implementation manner, where the length of the feature sequence is L, and the feature sequence is a concatenation sequence including known commodities and time intervals in L historical behaviors near the prediction time.
Alternatively, the signature sequence is a concatenated sequence comprising known goods and time intervals in historical behavior within a preset time range adjacent to the predicted time.
In particular, because historical behaviors that are far apart may not have reference to the current prediction, it is desirable to define the length of the signature sequence or to define the range of occurrences of historical behaviors in the signature sequence.
On the basis of fig. 2, regarding the content in S105, the embodiment of the present application further provides a possible implementation manner, please refer to fig. 3, where S105 includes:
s105-1, selecting the suspected reason products according to the weight, wherein the suspected reason products are the known products with the weight larger than the set threshold value and/or the known products with the weight of a set number of digits before the ranking.
Specifically, the commodities with the weight larger than a predetermined threshold value for the predicted commodity among the known commodities are screened out and are used as suspected cause commodities;
or screening out the commodities with the given digits before the weight ranking from the known commodities, and taking the commodities as suspected cause commodities;
alternatively, the product with the weight greater than a predetermined threshold value and a predetermined number of top-ranked weights among the known products is selected as the product of suspected cause.
Possibly, the higher the weight of the good, the higher its ranking, and the greater its impact on the predicted good may be.
And S105-2, respectively calculating the similarity between each suspected reason commodity and the predicted commodity.
Specifically, the similarity between each suspected-cause commodity and the predicted commodity can be determined according to the similarity.
And S105-3, determining the suspected-cause commodity with the maximum similarity as the related commodity.
Possibly, when the degree of similarity is higher, it indicates that the suspected-cause product is more highly correlated with the predicted product.
On the basis of fig. 2, regarding the neural network model, the embodiment of the present application further provides a possible implementation manner, please refer to the following.
In order to capture the dependency between historical behavior and predicted item of goods, self-attention mechanism is adopted. Input sequence X experience dlA self-attentive encoder block having dhHead, daAnd a hidden unit. Suppose the output of the ith head of the jth self-attentive encoder is:
wherein HjIs the input to the jth self-association encoder,
(parameters).
self-attention is defined as:
from the above, known ashAnd the head splices corresponding outputs to obtain the output of the jth self-attentive encoder:
wherein the content of the first and second substances,
(parameters).
Thereafter, the residual network is used to get the output of the jth attention block:
Aj＝LN(Hj+dropout(Fj(Zj)))。
wherein, FjRepresenting a feedforward layer, LN a layerormalization。
Point-wiseFeed-forwarddnetworks: to enhance the nonlinear properties of the model, FFN is used, defined as follows:
Fj＝FFN(Aj)；
thus:
Hj+1＝LN(Aj+dropout(GELU(AjW(1)+b(1))W(2)+b(2)；
wherein, W(1)、W(2)、b(1)And b(2)Are all parameters.
To compute a weight estimate between each entered known good and the last clicked good, the hidden layer state of the last layer is used
Embedding x with last item and time thereofLAnd then once again calculated by a similar mechanism:
the weight of each input known good is obtained through the calculation. It should be noted that, the intermediate value of α needs to be recorded in the process to prepare for subsequent calculation.
The results concat of the last layer are taken together, denoted by X ', and the final output is predicted using X' as input: p (v) ═ softmax (RELU (X ' W ' + b ') W ″ + b ").
Wherein p (v) represents the size of the probability of each recommended commodity, and W ', W', b ', and b' are parameters.
In the embodiment of the present application, Self-attention is a Self-attention model, encoder is a coding block, a corresponding decoder is a decoding block, and block represents a block.
On the basis of fig. 2, regarding the training method of the neural network model, a possible implementation manner is further provided in the embodiment of the present application, please refer to fig. 4, where the service information pushing method further includes:
and S102, training the neural network model according to the historical behaviors.
Possibly, the neural network model is trained according to the characteristic sequences corresponding to the known commodities and the time stamps in the historical behaviors.
And S103, when the loss function of the neural network model reaches the local optimal solution, the neural network model is considered to be converged, and the training is stopped.
Wherein, the expression of the loss function (objective function) is:
wherein y represents a real commodity; p (x) characterizing the predicted commodity; n represents the number of commodities; j represents an objective function; alpha represents the regularization weight; θ characterizes the given parameter.
In the above equation, the training result of the model is detected by the loss function, and overfitting is prevented by regularization.
On the basis of fig. 2, regarding to optimize the pushing result, the embodiment of the present application further provides a possible implementation manner, and the service information pushing method further includes:
when an ignoring instruction transmitted by the client is received, wherein the ignoring instruction comprises commodities which are not concerned by the user, the historical behaviors corresponding to the commodities in the ignoring instruction are deleted, and the interference on the user is avoided.
Referring to fig. 5, fig. 5 is a schematic diagram of a service information pushing apparatus according to an embodiment of the present application, where optionally, the service information pushing apparatus is applied to the electronic device described above.
Service information push method a service information push apparatus includes: a processing unit 201 and a pushing unit 202.
The processing unit 201 is configured to generate a feature sequence according to the known product and the timestamp in the historical behavior of the user.
The historical behaviors comprise behaviors of clicking commodities, collecting commodities, purchasing commodities and adding the commodities into a shopping cart, the characteristic sequence is a splicing sequence of known commodities and time intervals in the historical behaviors, and the time intervals are time stamps of the historical behaviors and the intervals of predicted time. Specifically, the processing unit 201 may execute S101 described above.
The processing unit 201 is further configured to use the feature sequence as an input of a neural network model to obtain a weight of the predicted commodity and each known commodity in the feature sequence relative to the predicted commodity; and the method is also used for determining the associated commodities according to the weight and the similarity of each known commodity, wherein the similarity is the similarity between the known commodity and the predicted commodity, and the associated commodities are commodities associated with the predicted commodity. Specifically, the processing unit 201 may execute S104 and S105 described above.
The pushing unit 202 is configured to push the predicted commodity to the client, and transmit a pushing interpretation to the client, where the pushing interpretation includes the associated commodity. Specifically, the pushing unit 202 may perform S106 described above.
Further, the processing unit 201 is further configured to select a suspected-cause product according to the weight, where the suspected-cause product is a known product whose weight is greater than a predetermined threshold and/or a known product whose weight is a predetermined number of digits before the weight is sorted; respectively calculating the similarity between each suspected reason commodity and the predicted commodity; and determining the suspected reason commodity with the maximum similarity as the related commodity. Specifically, the processing unit 201 may execute the above-described S105-1 to S105-3.
Possibly, the length of the feature sequence is L, and the feature sequence is a spliced sequence containing known commodities and time intervals in L historical behaviors near the predicted time.
It should be noted that, the service information pushing apparatus provided in this embodiment may execute the method flows shown in the above method flow embodiments to achieve the corresponding technical effects. For the sake of brevity, the corresponding contents in the above embodiments may be referred to where not mentioned in this embodiment.
The embodiment of the invention also provides a storage medium, wherein the storage medium stores computer instructions and programs, and the computer instructions and the programs execute the service information pushing method of the embodiment when being read and run. The storage medium may include memory, flash memory, registers, or a combination thereof, etc.
The following provides an electronic device, which may be a server device, and as shown in fig. 1, the electronic device may implement the service information pushing method described above; specifically, the electronic device includes: processor 10, memory 11, bus 12. The processor 10 may be a CPU. The memory 11 is used for storing one or more programs, and when the one or more programs are executed by the processor 10, the service information push method of the above-described embodiment is performed.
In the embodiments provided in the present application, it should be understood that the disclosed apparatus and method may be implemented in other ways. The apparatus embodiments described above are merely illustrative, and for example, the flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of apparatus, methods and computer program products according to various embodiments of the present application. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
In addition, functional modules in the embodiments of the present application may be integrated together to form an independent part, or each module may exist separately, or two or more modules may be integrated to form an independent part.
The functions, if implemented in the form of software functional modules and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present application or portions thereof that substantially contribute to the prior art may be embodied in the form of a software product stored in a storage medium and including instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present application. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
The above description is only a preferred embodiment of the present application and is not intended to limit the present application, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, improvement and the like made within the spirit and principle of the present application shall be included in the protection scope of the present application.
It will be evident to those skilled in the art that the present application is not limited to the details of the foregoing illustrative embodiments, and that the present application 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 application 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 sign in a claim should not be construed as limiting the claim concerned.
Claims (10)
1. A service information pushing method, characterized in that the method comprises:
generating a characteristic sequence according to known commodities and time stamps in the historical behaviors of the user;
the historical behaviors comprise a commodity clicking behavior, a commodity collecting behavior, a commodity purchasing behavior and a behavior of adding commodities into a shopping cart, the characteristic sequence is a splicing sequence of known commodities and time intervals in the historical behaviors, and the time intervals are intervals of the time stamps and the predicted time of the historical behaviors;
taking the characteristic sequence as the input of a neural network model to obtain the weight of a predicted commodity and each known commodity in the characteristic sequence relative to the predicted commodity;
determining related commodities according to the weight and the similarity of each known commodity, wherein the similarity is the similarity between the known commodity and the predicted commodity, and the related commodities are commodities related to the predicted commodity;
and pushing the predicted commodity to a client, and transmitting a pushing explanation to the client, wherein the pushing explanation comprises the associated commodity.
2. The service information pushing method according to claim 1, wherein the step of determining the associated goods according to the weight and the similarity of each known good comprises:
selecting suspected reason commodities according to the weight, wherein the suspected reason commodities are known commodities of which the weight is greater than a set threshold value and/or known commodities of which the weight is a set number of digits before the weight is sorted;
respectively calculating the similarity between each suspected reason commodity and the predicted commodity;
and determining the suspected reason commodity with the maximum similarity as the related commodity.
3. The service information pushing method according to claim 1, wherein the length of the feature sequence is L, and the feature sequence is a concatenated sequence including known goods and time intervals in the history behaviors L times near a predicted time.
4. The service information pushing method according to claim 1, wherein the characteristic sequence is a concatenated sequence including known goods and time intervals in the historical behaviors within a preset time range adjacent to a predicted time.
5. The service information pushing method of claim 1, wherein the method further comprises:
training the neural network model according to the historical behaviors;
when the loss function of the neural network model reaches the local optimal solution, the neural network model is considered to be converged, and the training is stopped;
the expression of the loss function is:
wherein y represents a real commodity; p (x) characterizing the predicted commodity; n represents the number of commodities; j represents an objective function; alpha represents the regularization weight; θ characterizes the given parameter.
6. A service information push method, a service information push apparatus, the apparatus comprising:
the processing unit is used for generating a characteristic sequence according to known commodities and time stamps in the historical behaviors of the user;
the historical behaviors comprise a commodity clicking behavior, a commodity collecting behavior, a commodity purchasing behavior and a behavior of adding commodities into a shopping cart, the characteristic sequence is a splicing sequence of known commodities and time intervals in the historical behaviors, and the time intervals are the time stamps and the predicted time intervals of the historical behaviors;
the processing unit is further used for inputting the characteristic sequence into a neural network model so as to obtain a predicted commodity and the weight of each known commodity in the characteristic sequence relative to the predicted commodity; the system is also used for determining related commodities according to the weight and the similarity of each known commodity, wherein the similarity is the similarity between the known commodity and the predicted commodity, and the related commodities are commodities related to the predicted commodity;
and the pushing unit is used for pushing the predicted commodity to a client and transmitting a pushing explanation to the client, wherein the pushing explanation comprises the related commodity.
7. The service information pushing device according to claim 6, wherein the processing unit is further configured to select a suspected-cause product according to the weight, wherein the suspected-cause product is a known product with the weight greater than a predetermined threshold and/or a known product with a predetermined number of digits before the weight ranking; respectively calculating the similarity between each suspected reason commodity and the predicted commodity; and determining the suspected reason commodity with the maximum similarity as the related commodity.
8. The service information pushing apparatus according to claim 6, wherein the length of the characteristic sequence is L, and the characteristic sequence is a concatenated sequence including known goods and time intervals in the history behaviors L times near a predicted time.
9. 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 1-5.
10. An electronic device, comprising: a processor and memory for storing one or more programs; the one or more programs, when executed by the processor, implement the method of any of claims 1-5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010854690.2A CN112001776A (en) | 2020-08-24 | 2020-08-24 | Service information pushing method and device, storage medium and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010854690.2A CN112001776A (en) | 2020-08-24 | 2020-08-24 | Service information pushing method and device, storage medium and electronic equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112001776A true CN112001776A (en) | 2020-11-27 |
Family
ID=73473092
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010854690.2A Pending CN112001776A (en) | 2020-08-24 | 2020-08-24 | Service information pushing method and device, storage medium and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112001776A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113065067A (en) * | 2021-03-31 | 2021-07-02 | 达而观信息科技（上海）有限公司 | Article recommendation method and device, computer equipment and storage medium |
| CN113360773A (en) * | 2021-07-07 | 2021-09-07 | 脸萌有限公司 | Recommendation method and device, storage medium and electronic equipment |
| CN115641191A (en) * | 2022-12-02 | 2023-01-24 | 菏泽牡丹区一涵网络科技有限责任公司 | Data pushing method based on data analysis and AI system |
| WO2023142927A1 (en) * | 2022-01-27 | 2023-08-03 | 北京有竹居网络技术有限公司 | Method for obtaining recommendation interpretation, device, and computer readable medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2008152606A (en) * | 2006-12-19 | 2008-07-03 | Dainippon Printing Co Ltd | Information recommendation system |
| CN104216883A (en) * | 2013-05-29 | 2014-12-17 | 酷盛(天津）科技有限公司 | Video recommendation reason generating system and method |
| CN105869024A (en) * | 2016-04-20 | 2016-08-17 | 北京小米移动软件有限公司 | Commodity recommending method and device |
| CN106228386A (en) * | 2016-07-12 | 2016-12-14 | 腾讯科技（深圳）有限公司 | A kind of information-pushing method and device |
| CN110569432A (en) * | 2019-08-14 | 2019-12-13 | 中国平安人寿保险股份有限公司 | commodity sequence calculation method and device, computer equipment and storage medium |
-
2020
- 2020-08-24 CN CN202010854690.2A patent/CN112001776A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2008152606A (en) * | 2006-12-19 | 2008-07-03 | Dainippon Printing Co Ltd | Information recommendation system |
| CN104216883A (en) * | 2013-05-29 | 2014-12-17 | 酷盛(天津）科技有限公司 | Video recommendation reason generating system and method |
| CN105869024A (en) * | 2016-04-20 | 2016-08-17 | 北京小米移动软件有限公司 | Commodity recommending method and device |
| CN106228386A (en) * | 2016-07-12 | 2016-12-14 | 腾讯科技（深圳）有限公司 | A kind of information-pushing method and device |
| CN110569432A (en) * | 2019-08-14 | 2019-12-13 | 中国平安人寿保险股份有限公司 | commodity sequence calculation method and device, computer equipment and storage medium |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113065067A (en) * | 2021-03-31 | 2021-07-02 | 达而观信息科技（上海）有限公司 | Article recommendation method and device, computer equipment and storage medium |
| CN113360773A (en) * | 2021-07-07 | 2021-09-07 | 脸萌有限公司 | Recommendation method and device, storage medium and electronic equipment |
| WO2023142927A1 (en) * | 2022-01-27 | 2023-08-03 | 北京有竹居网络技术有限公司 | Method for obtaining recommendation interpretation, device, and computer readable medium |
| CN115641191A (en) * | 2022-12-02 | 2023-01-24 | 菏泽牡丹区一涵网络科技有限责任公司 | Data pushing method based on data analysis and AI system |
| CN115641191B (en) * | 2022-12-02 | 2023-09-29 | 广州市原象信息科技有限公司 | Data pushing method and AI system based on data analysis |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112001776A (en) | Service information pushing method and device, storage medium and electronic equipment | |
| WO2018040944A1 (en) | System, method, and device for identifying malicious address/malicious purchase order | |
| CN106649681B (en) | Data processing method, device and equipment | |
| WO2016101777A1 (en) | Analysis and collection system for user interest data and method therefor | |
| US8355997B2 (en) | Method and system for developing a classification tool | |
| US20170032417A1 (en) | Detecting and generating online behavior from a clickstream | |
| WO2016035072A2 (en) | Sentiment rating system and method | |
| CN111753092B (en) | Data processing method, model training method, device and electronic equipment | |
| CN111783016B (en) | Website classification method, device and equipment | |
| Jiang et al. | Recommending new features from mobile app descriptions | |
| JP5729308B2 (en) | Reputation analysis apparatus, reputation analysis method, and reputation analysis program | |
| CN111738541A (en) | Method and device for acquiring store food quality information and electronic equipment | |
| Marwat et al. | Sentiment analysis of product reviews to identify deceptive rating information in social media: a SentiDeceptive approach | |
| CN111444424A (en) | Information recommendation method and information recommendation system | |
| JP2008146293A (en) | Evaluation system, method and program for browsing target information | |
| CN112989118B (en) | Video recall method and device | |
| CN112765966A (en) | Method and device for removing duplicate of associated word, computer readable storage medium and electronic equipment | |
| CN110837732B (en) | Method and device for identifying intimacy between target persons, electronic equipment and storage medium | |
| Rhohmawati et al. | Sentiment analysis using maximum entropy on application reviews (study case: Shopee on Google play) | |
| Sumathi et al. | Fake review detection of e-commerce electronic products using machine learning techniques | |
| CN114358879A (en) | Real-time price monitoring method and system based on big data | |
| CN111178987B (en) | Method and device for training user behavior prediction model | |
| CN110807646A (en) | Data analysis method, device and computer readable storage medium | |
| Lee et al. | Abnormal usage sequence detection for identification of user needs via recurrent neural network semantic variational autoencoder | |
| US20170076339A1 (en) | Method, Apparatus, Computer Program Product and System for Reputation Generation |
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 | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20240111 Address after: 200,030 Unit 5B03, Floor 5, Building 2, No. 277, Longlan Road, Xuhui District, Shanghai Applicant after: SHANGHAI SECOND PICKET NETWORK TECHNOLOGY CO.,LTD. Address before: Floors 4, 5 and 6, No. 3, Lane 1473, Zhenguang Road, Putuo District, Shanghai, 200333 Applicant before: Shanghai Fengzhi Technology Co.,Ltd. |
|
| TA01 | Transfer of patent application right |