CN112765401B - Short video recommendation method based on non-local network and local network - Google Patents

Short video recommendation method based on non-local network and local network Download PDF

Info

Publication number
CN112765401B
CN112765401B CN202110034609.0A CN202110034609A CN112765401B CN 112765401 B CN112765401 B CN 112765401B CN 202110034609 A CN202110034609 A CN 202110034609A CN 112765401 B CN112765401 B CN 112765401B
Authority
CN
China
Prior art keywords
click
short video
sequence
user
behavior
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.)
Active
Application number
CN202110034609.0A
Other languages
Chinese (zh)
Other versions
CN112765401A (en
Inventor
顾盼
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beijing Xindong Internet Information Technology Co.,Ltd.
Guangzhou Runxin Intellectual Property Operation Co ltd
Original Assignee
China Jiliang University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by China Jiliang University filed Critical China Jiliang University
Priority to CN202110034609.0A priority Critical patent/CN112765401B/en
Publication of CN112765401A publication Critical patent/CN112765401A/en
Application granted granted Critical
Publication of CN112765401B publication Critical patent/CN112765401B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/70—Information retrieval; Database structures therefor; File system structures therefor of video data
    • G06F16/73—Querying
    • G06F16/735—Filtering based on additional data, e.g. user or group profiles
    • 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
    • G06N3/048—Activation functions
    • 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

Abstract

The invention discloses a short video recommendation method based on a non-local network and a local network. According to the method, user interest representation is obtained according to a multi-behavior interaction sequence of a user on the short video, and then the click rate of the user on the target short video is predicted. The original sequence recommendation method aims at single-line sequence and cannot be applied to multi-line interactive sequence characterization. While the short video interaction sequence of the user includes both "click" and "affirmative" behavior. The method therefore proposes a method combining a non-local network and a local network. Mainly consists of three parts: the first part is to adopt an attention mechanism method to obtain the influence of 'positive' behaviors in a short video interaction sequence of a user on each 'click' behavior; the second part is to adopt a recurrent neural network based on a non-local attention mechanism to generate a user interest representation; and the third part is that the click rate of the user on the target short video is predicted according to the user interest representation.

Description

Short video recommendation method based on non-local network and local network
Technical Field
The invention belongs to the technical field of internet service, and particularly relates to a short video recommendation method based on a non-local network and a local network.
Background
Short video is a new type of video with a short time. The shooting of the short video does not need to use professional equipment and professional skills. The user can conveniently shoot and upload to the short video platform directly through the mobile phone, so that the short video frequency quantity of the short video platform is increased very quickly. The requirement on the effective short video recommendation system is very urgent, and the effective short video recommendation system can improve the user experience and the user viscosity, so that huge commercial value is brought to the platform.
In recent years, many researchers have proposed personalized recommendation methods based on videos. These methods can be divided into three categories: collaborative filtering, content-based recommendations, and hybrid recommendation methods. But short video has different characteristics compared to video: the descriptive text is of low quality, short duration and the user has a long sequence of interactions over a period of time. Therefore, short video recommendations are a more challenging task. And there are many kinds of user's interactive behaviors including "click", "like" and "favorite" in the short video recommendation problem. Different interactive behaviors represent different likelihoods. "click" indicates that the user would like to watch the short video, but the emotion is not strong; the 'like' and 'favorite' belong to the strong and definite emotions of the user, the 'like' indicates that the user likes the short video and is willing to watch the video of the same type, and the 'favorite' indicates that the user not only likes the short video currently but also wants to watch the video later. Short videos that are "liked" and "favorite" by the user are also "clicked" by the user at the same time. "like" and "favorite" can be generalized to the same type of behavior, i.e., "positive" behavior. At this time, the interaction sequence of the user has two kinds of interaction behaviors, one is a "click" behavior, and the other is an "affirmative" behavior. Some methods have been proposed by researchers to address the short video recommendation problem. For example, Chen et al use a hierarchical attention mechanism to calculate the importance of both the item and category levels to obtain more accurate predictions. Li et al used a graph-based recurrent neural network to model and finally get the user's preferences.
The Chen et al method only uses the click behavior information of the user and does not consider other behavior information of the user. Li et al apply the sequence recommendation method to "click" and "affirmative" behavior sequences, respectively. Finally, experiments show that the user interest representation based on the 'positive' behavior sequence has no obvious effect on improving the recommendation effect. The reasons are two: the time interval of the 'positive' action sequence of the user is longer, and the sequence is not strong; the "positive" behavior sequence is modeled separately, ignoring the effect of the "positive" behavior on subsequent "click" behavior. The method creatively provides a multi-behavior interaction sequence modeling method, two behavior sequences of click and affirmation are put into one behavior sequence to be processed, and the user interest vector representation is generated. Where the "click" behavior is sequential, and the "affirmative" behavior is unordered because the behavior events are more spaced apart. The method combines a non-local network (non-local network) and a local network (local network), wherein the non-local network adopts an attention mechanism (attention mechanism) and learns the influence of 'positive' behaviors on 'click' behaviors in the past period of time; the local network adopts a gated recurrent neural network (GRU) to learn the sequentiality of the click behavior. The method is a recurrent neural network based on a non-local attention mechanism, and improves the structure of an original neural network, so that the network can simultaneously learn the influence of 'positive' behaviors on 'clicking' behaviors and the influence of 'clicking' behaviors on 'clicking' behaviors.
Disclosure of Invention
The technical problem to be solved by the invention is to predict the click rate of a user on a target short video according to a multi-behavior click sequence of the user on the short video. There are many kinds of user interaction behaviors including "click", "like", and "favorite". Different interactive behaviors represent different likelihoods. "click" indicates that the user would like to watch the short video, but the emotion is not strong; the 'like' and 'favorite' both belong to the strong and definite emotions of the user, the 'like' means that the user likes the short video and is willing to watch the video of the same type, and the 'favorite' means that the user not only likes the short video currently but also wants to watch the video later. Short videos that are "liked" and "favorite" by the user are also "clicked" by the user at the same time. "like" and "favorite" can be generalized to the same type of behavior, i.e., "positive" behavior. At this time, the interaction sequence of the user has two kinds of interaction behaviors, one is a "click" behavior, and the other is an "affirmative" behavior. However, the original sequence recommendation methods are all directed to a sequence of interactive actions. Therefore, the invention adopts the following technical scheme:
a short video recommendation method based on a non-local network and a local network comprises the following steps:
and (3) obtaining the influence of the 'positive' behavior on each 'click' behavior in the short video multi-behavior interaction sequence of the user by adopting an attention mechanism method. Sequence of interactive actions for a user
Figure BDA0002893654860000011
Can be represented as X ═ X1,…,xl]Wherein
Figure BDA0002893654860000012
Is the feature vector of the cover picture of the short video, and d is the feature vector length. Wherein a sequence of "positive" behaviors is represented as
Figure BDA0002893654860000013
And X*Is a subset of X. "click" behavior sequenceI.e. X ═ X1,…,xl]. The influence of the 'positive' behavior sequence on the 'click' behavior is obtained by using an attention mechanism method in a non-local network method. Typically, the last-click short video (last-click) in the sequence is used to represent the user's current click interest, so the attention mechanism is based on the last-click short video:
Figure BDA0002893654860000021
Figure BDA0002893654860000022
wherein the content of the first and second substances,
Figure BDA0002893654860000023
and
Figure BDA0002893654860000024
are parameters that the model needs to be trained. x is the number oftRepresenting the last short video vector representation in the click sequence,
Figure BDA0002893654860000025
the ith short video vector representation representing the "positive" sequence in the current "click" sequence. Sigma is sigmoid function.
Figure BDA0002893654860000026
Ith short video vector characterization representing a "positive" sequence in the current "click" sequence
Figure BDA0002893654860000027
The degree of importance of.
Figure BDA0002893654860000028
Is xtThe impact of "positive" behavior in the ending "click" behavior sequence on the current click interest.
A user interest characterization is generated using a recurrent neural network based on a non-local attention mechanism. The original gated recurrent neural network (GRU) can only handle single-action sequences, with the structure:
zt＝σ(Wxz·xt+Whz·ht-1)
rt＝σ(Wxr·xt+Whr·ht-1)
Figure BDA0002893654860000029
Figure BDA00028936548600000210
wherein r istIs a reset gate, ztTo update the gates (update gate), these two gating vectors determine which information can be used as the output of the gated loop unit.
Figure BDA00028936548600000211
Is the current memory content. x is the number oftIs the node input for the current layer.
Figure BDA00028936548600000212
Figure BDA00028936548600000213
And
Figure BDA00028936548600000214
respectively, control the update gate ztAnd a reset gate rtThe parameter (c) of (c).
Figure BDA00028936548600000215
And
Figure BDA00028936548600000216
is to control the pre-memory content
Figure BDA00028936548600000217
The parameter (c) of (c). AnMatrix multiplication at the element level, σ is the sigmoid function.
Gated recurrent neural networks, however, do not apply to multi-behavior sequences. In order to be suitable for a multi-behavior sequence, the method improves the original gated recurrent neural network, so that the gated recurrent neural network unit (unit) selects information not only considering the current short video in the sequence and the state of the last gated recurrent neural network unit, but also considering the influence of "positive" behavior, as follows:
Figure BDA00028936548600000218
Figure BDA00028936548600000219
Figure BDA00028936548600000220
Figure BDA00028936548600000221
wherein z istTo update the gate (update gate), rtIs a reset gate (reset gate) and these two gating vectors determine which information can be used as the output of the gated loop unit.
Figure BDA00028936548600000222
Is the current memory content. x is the number oftIs the node input at the current level,
Figure BDA00028936548600000223
is the effect of "positive" behavior.
Figure BDA00028936548600000224
Figure BDA00028936548600000225
And
Figure BDA00028936548600000226
respectively, control the update gate ztAnd a reset gate rtThe parameter (c) of (c).
Figure BDA00028936548600000227
And
Figure BDA00028936548600000228
is to control the pre-memory content
Figure BDA00028936548600000229
The parameter (c) of (c). As is the element-level matrix multiplication, σ is the sigmoid function. Hidden state h of last layer of gate control recurrent neural networktThe output of (a) is the user interest representation v.
Predicting the target short video x of the user according to the user interest representationnewClick rate of (2):
Figure BDA00028936548600000230
where v is a user interest representation, xnewIs the target short video.
Figure BDA00028936548600000231
And the predicted value of the click rate of the user on the target short video is shown.
Figure BDA00028936548600000232
Figure BDA00028936548600000233
And
Figure BDA00028936548600000234
is a matrix of transitions that is,
Figure BDA00028936548600000235
is an offset vector, b2Is a bias scalar. σ is the sigmoid activation function.
And designing a loss function according to the model characteristics. Predicting value of click rate of target short video through user
Figure BDA0002893654860000031
Calculating a predicted value
Figure BDA0002893654860000032
And the true value y, and the error is used to update the model parameters. We use a cross-entropy loss function to guide the update process of model parameters:
Figure BDA0002893654860000033
wherein y ∈ {0,1} is a true value representing whether the user clicked on the target short video. σ is a sigmoid function. We update the model parameters using Adam optimizer.
The invention has the following beneficial technical effects:
(1) the invention discloses a multi-behavior sequence characterization method. Different from the previous single behavior sequence characterization method, the method puts the two behavior sequences of click and affirmation into one behavior sequence for processing to generate the user interest vector characterization. Where the "click" behavior is sequential, and the "affirmative" behavior is unordered because the behavior events are more spaced apart.
(2) The invention combines non-local network (non-local network) and local network (local network). Wherein, the non-local network adopts an attention mechanism (attention mechanism), and learns the influence of all 'positive' behaviors on 'click' behaviors in the past period of time; the local network adopts a gated recurrent neural network (GRU) to learn the influence of the click behavior on the click behavior in the near period of time.
(3) The invention relates to a recurrent neural network based on a non-local attention mechanism, which can enable the network to simultaneously learn the influence of 'positive' behavior on 'click' behavior and the influence of 'click' behavior on 'click' behavior by improving the structure of an original neural network.
Drawings
FIG. 1 is a schematic flow chart of a short video recommendation method based on a non-local network and a local network according to the present invention;
fig. 2 is a model framework diagram of a short video recommendation method based on a non-local network and a local network according to the present invention.
Detailed Description
For further understanding of the present invention, the following describes a short video recommendation method based on non-local network and local network in detail with reference to specific embodiments, but the present invention is not limited thereto, and those skilled in the art can make insubstantial improvements and modifications under the core teaching of the present invention, and still fall within the scope of the present invention.
The short video click rate prediction task is to establish a model to predict the probability of the user clicking on the short video. The historical interactive short video sequence of the user is represented as
Figure BDA0002893654860000034
Wherein x isjRepresenting the jth short video, l is the length of the sequence. There are many kinds of user interaction behaviors including "click", "like", and "favorite". Different interactive behaviors represent different likelihoods. "click" indicates that the user would like to watch the short video, but the emotion is not strong; the 'like' and 'favorite' belong to the strong and definite emotions of the user, the 'like' indicates that the user likes the short video and is willing to watch the video of the same type, and the 'favorite' indicates that the user not only likes the short video currently but also wants to watch the video later. Short videos that are "liked" and "favorite" by the user are also "clicked" by the user at the same time. Thus, the short video click-through rate prediction problem can be expressed as: inputting user multi-behavior interaction sequences
Figure BDA0002893654860000035
And target short video xnewTo predict the user-to-target short video xnewThe click rate of (c).
Therefore, the invention provides a short video recommendation method based on a non-local network and a local network. The method predicts the click rate of the user on the target short video according to the multi-behavior interaction sequence of the user on the short video. The multiple behaviors here include "click," "like," and "favorite" behaviors of the user. In the method, "like" and "favorite" are summarized as the same type of behavior, i.e., "affirmative" behavior. At this time, the interaction sequence of the user has two kinds of interaction behaviors, one is a "click" behavior, and the other is an "affirmative" behavior. The original sequence recommendation method is directed to a sequence of interactive behaviors. Li et al applied the sequence recommendation method to the "click" behavior sequence and the "affirmative" behavior sequence, respectively, and finally experiments showed that the user interest characterization based on the "affirmative" behavior sequence had a very insignificant effect on improving the recommendation effect. The reasons are two: the time interval of the 'positive' action sequence of the user is longer, and the sequence is not strong; the "positive" behavior sequence is modeled separately, ignoring the effect of the "positive" behavior on subsequent "click" behavior. The method creatively provides a multi-behavior interaction sequence modeling method, two behavior sequences of click and affirmation are put into one behavior sequence to be processed, and the user interest vector representation is generated. Where the "click" behavior is sequential, and the "affirmative" behavior is unordered because the behavior events are more spaced apart. The method combines a non-local network (non-local network) and a local network (local network), wherein the non-local network adopts an attention mechanism (attention mechanism) and learns the influence of 'positive' behaviors on 'click' behaviors in the past period of time; the local network adopts a gated recurrent neural network (GRU) to learn the sequentiality of the click behavior. The method is a cyclic neural network based on a non-local attention mechanism, and the structure of the original neural network is improved, so that the network can learn the influence of the 'positive' behavior on the 'click' behavior and the influence of the 'click' behavior on the 'click' behavior at the same time.
The method consists essentially of three parts, as shown in FIG. 2. The first part is to adopt an attention mechanism method to obtain the influence of the 'positive' behavior on each 'click' behavior in the short video multi-behavior interaction sequence of the user. The second part is to generate a user interest characterization using a recurrent neural network based on a non-local attention mechanism. And the third part is that the click rate of the user on the target short video is predicted according to the user interest representation.
As shown in fig. 1, according to one embodiment of the present invention, the method comprises the steps of:
s100, obtaining the influence of the 'positive' behavior on each 'click' behavior in the short video multi-behavior interaction sequence of the user by adopting an attention mechanism method. Sequence of interactive actions for a user
Figure BDA0002893654860000041
Can be represented as X ═ X1,…,xl]Wherein
Figure BDA0002893654860000042
Is the feature vector of the cover picture of the short video, and d is the feature vector length. Wherein a sequence of "positive" behaviors is represented as
Figure BDA0002893654860000043
And X*Is a subset of X. The sequence of "click" actions is X ═ X1，…,xl]. The influence of the 'positive' behavior sequence on the 'click' behavior is obtained by using an attention mechanism method in a non-local network method. Typically, the last-click short video (last-click) in the sequence is used to represent the user's current click interest, so the attention mechanism is based on the last-click short video:
Figure BDA0002893654860000044
Figure BDA0002893654860000045
wherein the content of the first and second substances,
Figure BDA0002893654860000046
and
Figure BDA0002893654860000047
are parameters that the model needs to be trained. x is the number oftRepresenting the last short video vector representation in the click sequence,
Figure BDA0002893654860000048
the ith short video vector representation representing the "positive" sequence in the current "click" sequence. Sigma is sigmoid function.
Figure BDA0002893654860000049
Ith short video vector characterization representing a "positive" sequence in the current "click" sequence
Figure BDA00028936548600000410
The degree of importance of.
Figure BDA00028936548600000411
Is xtThe impact of "positive" behavior in the ending "click" behavior sequence on the current click interest.
And S200, generating a user interest characterization by adopting a recurrent neural network based on a non-local attention mechanism. The original gated recurrent neural network (GRU) can only handle single-action sequences, with the structure:
zt＝σ(Wxz·xt+Whz·ht-1)
rt＝σ(Wxr·xt+Whr·ht-1)
Figure BDA00028936548600000412
Figure BDA00028936548600000413
wherein r istIs a reset gate, ztTo update the gate (update gate),these two gating vectors determine which information can be used as the output of the gated loop unit.
Figure BDA00028936548600000414
Is the current memory content. x is the number oftIs the node input for the current layer.
Figure BDA00028936548600000415
Figure BDA00028936548600000416
And
Figure BDA00028936548600000417
respectively, control the update gate ztAnd a reset gate rtThe parameter (c) of (c).
Figure BDA00028936548600000418
And
Figure BDA00028936548600000419
is to control the pre-memory content
Figure BDA00028936548600000420
The parameter (c) of (c). As is the element-level matrix multiplication, σ is the sigmoid function.
Gated recurrent neural networks, however, do not apply to multi-behavior sequences. In order to be suitable for a multi-behavior sequence, the method improves an original gate control recurrent neural network, so that the selection of information by a gate control recurrent neural network unit (unit) not only considers the states of a short video and the last gate control recurrent neural network unit in the current sequence, but also considers the influence of 'positive' behavior, as follows:
Figure BDA00028936548600000431
Figure BDA00028936548600000432
Figure BDA00028936548600000421
Figure BDA00028936548600000422
wherein z istTo update the gate (update gate), rtIs a reset gate (reset gate) and these two gating vectors determine which information can be used as the output of the gated loop unit.
Figure BDA00028936548600000423
Is the current memory content. x is the number oftIs the node input at the current level,
Figure BDA00028936548600000424
is the effect of "positive" behavior.
Figure BDA00028936548600000425
Figure BDA00028936548600000426
And
Figure BDA00028936548600000427
respectively, control the update gate ztAnd a reset gate rtThe parameter (c) of (c).
Figure BDA00028936548600000428
And
Figure BDA00028936548600000429
is to control the pre-memory content
Figure BDA00028936548600000430
The parameter (c) of (c). As is the element-level matrix multiplication, σ is the sigmoid function. Hidden state h of last layer of gate control recurrent neural networktThe output of (a) is the user interest representation v.
S300, according to the userInterest representation for predicting target short video x of usernewClick rate of (2):
Figure BDA0002893654860000051
where v is a user interest representation, xnewIs the target short video.
Figure BDA0002893654860000052
And the predicted value of the click rate of the user on the target short video is shown.
Figure BDA0002893654860000053
Figure BDA0002893654860000054
And
Figure BDA0002893654860000055
is a matrix of transitions that is,
Figure BDA0002893654860000056
is an offset vector, b2Is a bias scalar. σ is the sigmoid activation function.
And S400, designing a loss function according to the model characteristics. Predicting value of click rate of target short video through user
Figure BDA0002893654860000057
Calculating a predicted value
Figure BDA0002893654860000058
And the true value y, and the error is used to update the model parameters. We use a cross-entropy loss function to guide the update process of model parameters:
Figure BDA0002893654860000059
wherein y ∈ {0,1} is a true value representing whether the user clicked on the target short video. σ is a sigmoid function. We update the model parameters using Adam optimizer.
The foregoing description of the embodiments is provided to facilitate understanding and application of the invention by those skilled in the art. It will be readily apparent to those skilled in the art that various modifications to the above-described embodiments may be made, and the generic principles defined herein may be applied to other embodiments without the use of inventive faculty. Therefore, the present invention is not limited to the above embodiments, and those skilled in the art should make improvements and modifications to the present invention based on the disclosure of the present invention within the protection scope of the present invention.

Claims (1)

1. A short video recommendation method based on a non-local network and a local network is characterized in that:
the method comprises the steps that an attention mechanism method is adopted, and the influence of 'positive' behaviors on each 'click' behavior in a short video multi-behavior interaction sequence of a user is obtained; sequence of interactive actions for a user
Figure FDA0003292152110000011
Can be represented as X ═ X1,…,xl]Wherein
Figure FDA0003292152110000012
Is the feature vector of the cover picture of the short video, d is the length of the feature vector; wherein a sequence of "positive" behaviors is represented as
Figure FDA0003292152110000013
And X*Is a subset of X; the sequence of "click" actions is X ═ X1,…,xl](ii) a Obtaining the influence of the 'affirmation' behavior sequence on the 'click' behavior by using an attention mechanism method in a non-local network method; the last-click short video (last-click) in the sequence is used to represent the user's current click interest, so the attention mechanism is based on the last-click short video:
Figure FDA0003292152110000014
Figure FDA0003292152110000015
wherein the content of the first and second substances,
Figure FDA0003292152110000016
and
Figure FDA0003292152110000017
is the parameter that the model needs to be trained; x is the number oftRepresenting the last short video vector representation in the click sequence,
Figure FDA0003292152110000018
the ith short video vector representation of the "positive" sequence in the current "click" sequence; sigma is sigmoid function;
Figure FDA0003292152110000019
ith short video vector characterization representing a "positive" sequence in the current "click" sequence
Figure FDA00032921521100000110
The degree of importance of;
Figure FDA00032921521100000111
is xtThe influence of the "positive" behavior in the ending "click" behavior sequence on the current click interest;
generating a user interest representation by adopting a recurrent neural network based on a non-local attention mechanism; the selection of information by a gated recurrent neural network element (unit) in a recurrent neural network based on a non-local attention mechanism takes into account not only the current short video in the sequence and the state of the last gated recurrent neural network element, but also the effect of "positive" behavior, as follows:
Figure FDA00032921521100000112
Figure FDA00032921521100000113
Figure FDA00032921521100000114
Figure FDA00032921521100000115
wherein z istTo update the gate (update gate), rtIs a reset gate (reset gate), and these two gating vectors determine which information can be used as the output of the gated loop unit;
Figure FDA00032921521100000116
is the current memory content; x is the number oftIs the node input at the current level,
Figure FDA00032921521100000117
is the effect of a "positive" behavior;
Figure FDA00032921521100000118
Figure FDA00032921521100000119
and
Figure FDA00032921521100000120
respectively, control the update gate ztAnd a reset gate rtThe parameters of (1);
Figure FDA00032921521100000121
and
Figure FDA00032921521100000122
is to control the pre-memory content
Figure FDA00032921521100000123
The parameters of (1); as a matrix multiplication at the element level, σ is a sigmoid function; hidden state h of last layer of gate control recurrent neural networktThe output of (a) is the user interest representation v;
predicting the target short video x of the user according to the user interest representationnewClick rate of (2):
Figure FDA00032921521100000124
where v is a user interest representation, xnewIs a target short video;
Figure FDA00032921521100000125
the target short video click rate prediction value is obtained by the user;
Figure FDA00032921521100000126
Figure FDA00032921521100000127
and
Figure FDA00032921521100000128
is a matrix of transitions that is,
Figure FDA00032921521100000129
is an offset vector, b2Is a bias scalar; σ is a sigmoid activation function;
designing a loss function according to the model characteristics; predicting value of click rate of target short video through user
Figure FDA00032921521100000130
Calculating a predicted value
Figure FDA00032921521100000131
And the true value y, and then using the error to update the model parameters; and guiding the updating process of the model parameters by adopting a cross entropy loss function:
Figure FDA00032921521100000132
wherein y is an actual value and represents whether the user clicks the target short video or not, wherein y belongs to {0,1 }; σ is a sigmoid function; model parameters were updated using Adam optimizer.
CN202110034609.0A 2021-01-12 2021-01-12 Short video recommendation method based on non-local network and local network Active CN112765401B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110034609.0A CN112765401B (en) 2021-01-12 2021-01-12 Short video recommendation method based on non-local network and local network

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110034609.0A CN112765401B (en) 2021-01-12 2021-01-12 Short video recommendation method based on non-local network and local network

Publications (2)

Publication Number Publication Date
CN112765401A CN112765401A (en) 2021-05-07
CN112765401B true CN112765401B (en) 2021-11-12

Family

ID=75701415

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110034609.0A Active CN112765401B (en) 2021-01-12 2021-01-12 Short video recommendation method based on non-local network and local network

Country Status (1)

Country Link
CN (1) CN112765401B (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114358364A (en) * 2021-11-20 2022-04-15 重庆邮电大学 Attention mechanism-based short video frequency click rate big data estimation method

Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112100440A (en) * 2020-08-21 2020-12-18 腾讯科技（深圳）有限公司 Video pushing method, device and medium

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US11562243B2 (en) * 2017-11-17 2023-01-24 Meta Platforms, Inc. Machine-learning models based on non-local neural networks

Patent Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112100440A (en) * 2020-08-21 2020-12-18 腾讯科技（深圳）有限公司 Video pushing method, device and medium

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
《融合项目嵌入表征与注意力机制的推荐算法》;都奕冰 等;《计算机工程与设计》;20200331;第41卷(第3期);第682-688页 *

Also Published As

Publication number Publication date
CN112765401A (en) 2021-05-07

Similar Documents

Publication Publication Date Title
CN110781321B (en) Multimedia content recommendation method and device
Chen et al. Deep reinforcement learning in recommender systems: A survey and new perspectives
CN113905391B (en) Integrated learning network traffic prediction method, system, equipment, terminal and medium
CN111382361B (en) Information pushing method, device, storage medium and computer equipment
CN111310040A (en) Artificial intelligence based recommendation method and device, electronic equipment and storage medium
CN107463701B (en) Method and device for pushing information stream based on artificial intelligence
CN112765480B (en) Information pushing method and device and computer readable storage medium
CN110162698A (en) A kind of user's representation data processing method, device and storage medium
CN112765461A (en) Session recommendation method based on multi-interest capsule network
CN112492396A (en) Short video click rate prediction method based on fine-grained multi-aspect analysis
CN112395504A (en) Short video click rate prediction method based on sequence capsule network
US11694165B2 (en) Key-value memory network for predicting time-series metrics of target entities
Wang et al. Attention-based CNN for personalized course recommendations for MOOC learners
CN115186097A (en) Knowledge graph and reinforcement learning based interactive recommendation method
CN112395505B (en) Short video click rate prediction method based on cooperative attention mechanism
CN111831895A (en) Network public opinion early warning method based on LSTM model
CN112765401B (en) Short video recommendation method based on non-local network and local network
CN112819575B (en) Session recommendation method considering repeated purchasing behavior
CN113704438B (en) Conversation recommendation method of abnormal picture based on layered attention mechanism
CN112307258B (en) Short video click rate prediction method based on double-layer capsule network
US20240037133A1 (en) Method and apparatus for recommending cold start object, computer device, and storage medium
CN113688306A (en) Recommendation strategy generation method and device based on reinforcement learning
CN113836406A (en) Information flow recommendation method and device
WO2024051707A1 (en) Recommendation model training method and apparatus, and resource recommendation method and apparatus
CN112199550A (en) Short video click rate prediction method based on emotion capsule network

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant
TR01 Transfer of patent right
TR01 Transfer of patent right

Effective date of registration: 20230404

Address after: Room 513, No. 940 Congyun Road, Yongping Street, Baiyun District, Guangzhou City, Guangdong Province, 510420

Patentee after: Guangzhou Runxin Intellectual Property Operation Co.,Ltd.

Address before: 310018, No. 258, source street, Xiasha Higher Education Park, Hangzhou, Zhejiang

Patentee before: China Jiliang University

Effective date of registration: 20230404

Address after: 2-2003, Floor 20, Building 2, Yard A13, Beiyuan Road, Chaoyang District, Beijing, 100012

Patentee after: Beijing Xindong Internet Information Technology Co.,Ltd.

Address before: Room 513, No. 940 Congyun Road, Yongping Street, Baiyun District, Guangzhou City, Guangdong Province, 510420

Patentee before: Guangzhou Runxin Intellectual Property Operation Co.,Ltd.