CN116610873A - Information recommendation method and device and storage medium - Google Patents
Information recommendation method and device and storage medium Download PDFInfo
- Publication number
- CN116610873A CN116610873A CN202310887568.9A CN202310887568A CN116610873A CN 116610873 A CN116610873 A CN 116610873A CN 202310887568 A CN202310887568 A CN 202310887568A CN 116610873 A CN116610873 A CN 116610873A
- Authority
- CN
- China
- Prior art keywords
- information
- characteristic information
- real
- user
- historical
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 78
- 230000006399 behavior Effects 0.000 claims abstract description 122
- 238000003062 neural network model Methods 0.000 claims description 37
- 238000004364 calculation method Methods 0.000 claims description 6
- 238000012163 sequencing technique Methods 0.000 claims 2
- 230000008569 process Effects 0.000 description 14
- 238000013528 artificial neural network Methods 0.000 description 7
- 238000010586 diagram Methods 0.000 description 6
- 230000009471 action Effects 0.000 description 4
- 238000013473 artificial intelligence Methods 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical compound [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000001413 cellular effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000004590 computer program Methods 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 229910021389 graphene Inorganic materials 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000008707 rearrangement Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9538—Presentation of query results
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5061—Partitioning or combining of resources
- G06F9/5072—Grid computing
-
- 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
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2209/00—Indexing scheme relating to G06F9/00
- G06F2209/50—Indexing scheme relating to G06F9/50
- G06F2209/502—Proximity
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Databases & Information Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Software Systems (AREA)
- Mathematical Physics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Information Transfer Between Computers (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The specification provides an information recommendation method and device and a storage medium, wherein the method comprises the following steps: acquiring real-time characteristic information based on the operation behavior of at least one first recommendation information issued by a user to the cloud node; transmitting the real-time characteristic information to an edge node; receiving and storing target characteristic information returned by the edge node, wherein the target characteristic information is characteristic information used for representing commonality between historical operation behaviors and real-time operation behaviors of a user; and based on the target characteristic information, reordering at least one piece of second recommended information issued by the cloud node and outputting the reordered second recommended information. According to the method and the device, the computing power of the edge node can be utilized to provide the target characteristic information for the client to more accurately represent the user operation behaviors, the richness of the characteristic information for representing the user operation behaviors on the client is expanded, the time delay of information recommendation can be reduced, the resources of each device of the end-edge cloud can be utilized more reasonably, and the usability of the end-edge cloud architecture is improved.
Description
Technical Field
One or more embodiments of the present disclosure relate to the field of data processing, and in particular, to an information recommendation method and apparatus, and a storage medium.
Background
At present, after receiving information recommended by cloud nodes, a client can sense operation behaviors of a user in real time, so that operations such as intelligent refreshing and content rearrangement are performed. However, due to the limitation of the computing power of the client, less characteristic information is available on the client for representing the operation behavior of the user, and the ranking of the recommendation information output by the final client is not high in matching degree with the user, so that the process needs to be optimized.
Disclosure of Invention
One or more embodiments of the present disclosure provide an information recommendation method and apparatus, and a storage medium.
In order to achieve the above object, one or more embodiments of the present disclosure provide the following technical solutions:
according to a first aspect of one or more embodiments of the present disclosure, an information recommendation method is provided, including:
based on the operation behavior of at least one first recommendation information issued by a user to a cloud node, obtaining real-time characteristic information, wherein the real-time characteristic information is characteristic information for representing the real-time operation behavior of the user;
transmitting the real-time characteristic information to an edge node;
receiving and storing target characteristic information returned by the edge node, wherein the target characteristic information is characteristic information used for representing commonalities between historical operation behaviors of the user and the real-time operation behaviors;
And based on the target characteristic information, reordering at least one piece of second recommended information issued by the cloud node and outputting the reordered second recommended information.
According to a second aspect of one or more embodiments of the present disclosure, there is provided an information recommendation method, including:
receiving real-time characteristic information sent by a client, wherein the real-time characteristic information is characteristic information for representing real-time operation behaviors of a user;
acquiring historical characteristic information of the user, which is provided by a cloud node, wherein the historical characteristic information is characteristic information used for representing historical operation behaviors of the user;
and based on the real-time characteristic information and the historical characteristic information, obtaining target characteristic information, and synchronizing the target characteristic information to the client and the cloud node, wherein the target characteristic information is characteristic information for representing commonalities between the historical operation behaviors of the user and the real-time operation behaviors.
According to a third aspect of one or more embodiments of the present specification, there is provided an information recommendation method, including:
based on historical characteristic information of a user, at least one first recommendation information is issued to a client so that the client obtains real-time characteristic information, wherein the historical characteristic information is characteristic information used for representing historical operation behaviors of the user, and the real-time characteristic information is characteristic information used for representing real-time operation behaviors of the user on at least one first recommendation information;
Transmitting the historical characteristic information of the user to an edge node so that the edge node obtains target characteristic information based on the real-time characteristic information and the historical characteristic information, wherein the target characteristic information is characteristic information representing commonalities between the historical operation behaviors of the user and the real-time operation behaviors;
based on the target characteristic information sent by the edge node, synchronously updating the history characteristic information of the user;
and based on the updated historical characteristic information, at least one piece of second recommendation information is issued to the client so that the client reorders and outputs the at least one piece of second recommendation information based on the target characteristic information.
According to a fourth aspect of one or more embodiments of the present specification, there is provided an information recommendation apparatus, comprising:
the cloud node comprises a first processing module, a second processing module and a third processing module, wherein the first processing module is used for obtaining real-time characteristic information based on the operation behavior of at least one first recommendation information issued by a user to a cloud node, and the real-time characteristic information is used for representing the real-time operation behavior of the user;
the first sending module is used for sending the real-time characteristic information to an edge node;
The first receiving module is used for receiving and storing target characteristic information returned by the edge node, wherein the target characteristic information is characteristic information representing commonality between the historical operation behavior of the user and the real-time operation behavior;
and the second processing module is used for reordering at least one piece of second recommended information issued by the cloud node based on the target characteristic information and outputting the reordered second recommended information.
According to a fifth aspect of one or more embodiments of the present specification, there is provided an information recommendation apparatus, comprising:
the second receiving module is used for receiving real-time characteristic information sent by the client, wherein the real-time characteristic information is characteristic information used for representing real-time operation behaviors of the user;
the third receiving module is used for acquiring historical characteristic information of the user, which is provided by the cloud node, wherein the historical characteristic information is characteristic information used for representing historical operation behaviors of the user;
and the third processing module is used for obtaining target characteristic information based on the real-time characteristic information and the historical characteristic information and then synchronizing the target characteristic information to the client and the cloud node, wherein the target characteristic information is characteristic information for representing commonality between the historical operation behavior of the user and the real-time operation behavior.
According to a sixth aspect of one or more embodiments of the present specification, there is provided an information recommendation apparatus, comprising:
the second sending module is used for sending at least one first recommendation message to the client based on the historical characteristic information of the user so that the client can obtain real-time characteristic information, wherein the historical characteristic information is characteristic information used for representing the historical operation behavior of the user, and the real-time characteristic information is characteristic information used for representing the real-time operation behavior of the user on at least one first recommendation message;
the third sending module is used for sending the historical characteristic information of the user to an edge node so that the edge node obtains target characteristic information based on the real-time characteristic information and the historical characteristic information, wherein the target characteristic information is characteristic information representing the commonality between the historical operation behavior of the user and the real-time operation behavior;
the updating module is used for synchronously updating the history characteristic information of the user based on the target characteristic information sent by the edge node;
and the fourth sending module is used for sending at least one piece of second recommendation information to the client based on the updated historical characteristic information so that the client reorders and outputs the at least one piece of second recommendation information based on the target characteristic information.
According to a seventh aspect of one or more embodiments of the present specification, there is provided a client, comprising:
a processor;
a memory for storing processor-executable instructions;
wherein the processor implements the information recommendation method of any of the first aspects by executing the executable instructions.
According to a fourth aspect of one or more embodiments of the present specification, there is provided a method of providing an edge node, comprising:
a processor;
a memory for storing processor-executable instructions;
wherein the processor implements the information recommendation method of any of the second aspects by executing the executable instructions.
According to a ninth aspect of one or more embodiments of the present specification, there is provided a cloud node, comprising:
a processor;
a memory for storing processor-executable instructions;
wherein the processor implements the information recommendation method of any of the third aspects by executing the executable instructions.
According to a tenth aspect of one or more embodiments of the present specification, there is provided an information recommendation system, including:
a client for implementing the information recommendation method according to any one of the first aspects;
An edge node for implementing the information recommendation method according to the second aspect;
a cloud node for implementing the information recommendation method according to any one of the third aspects.
According to an eleventh aspect of one or more embodiments of the present specification, there is provided a computer readable storage medium having stored thereon computer instructions which, when executed by a processor, implement the steps of the information recommendation method as described in any of the above.
The technical scheme provided by the embodiment of the specification can comprise the following beneficial effects:
in the method, the computing power of the edge node can be utilized to provide the target feature information for the client to more accurately represent the user operation behavior, the richness of the feature information for representing the user operation behavior on the client is expanded, and the time delay of information recommendation can be reduced because the end node is close to the client.
Drawings
Fig. 1 is a schematic diagram of an information recommendation system according to an exemplary embodiment.
Fig. 2 is a flowchart of an information recommendation method according to an exemplary embodiment.
Fig. 3 is a flowchart of another information recommendation method according to an exemplary embodiment.
Fig. 4 is a flowchart of another information recommendation method according to an exemplary embodiment.
Fig. 5 is a flowchart of another information recommendation method according to an exemplary embodiment.
Fig. 6 is a block diagram of an information recommendation apparatus according to an exemplary embodiment.
Fig. 7 is a block diagram of another information recommendation device provided by an exemplary embodiment.
Fig. 8 is a block diagram of another information recommendation device provided by an exemplary embodiment.
Fig. 9 is a schematic structural diagram of an electronic device according to an exemplary embodiment.
Detailed Description
Reference will now be made in detail to exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, the same numbers in different drawings refer to the same or similar elements, unless otherwise indicated. The implementations described in the following exemplary embodiments do not represent all implementations consistent with one or more embodiments of the present specification. Rather, they are merely examples of apparatus and methods consistent with aspects of one or more embodiments of the present description as detailed in the accompanying claims.
It should be noted that: in other embodiments, the steps of the corresponding method are not necessarily performed in the order shown and described in this specification. In some other embodiments, the method may include more or fewer steps than described in this specification. Furthermore, individual steps described in this specification, in other embodiments, may be described as being split into multiple steps; while various steps described in this specification may be combined into a single step in other embodiments.
Before introducing the information recommendation scheme provided by the present disclosure, an information recommendation system based on a terminal edge cloud architecture in the present disclosure is first introduced, where the system is shown in fig. 1, and includes:
the client 101 (fig. 1 includes the client 101-1, the client 101-2, and the client … …) is responsible for collecting data and performing simple data processing according to needs, where the client 101 may be a terminal device, such as a mobile phone, an intelligent wearable device, a notebook computer, a personal assistant, and the disclosure is not limited thereto;
edge nodes 102 (fig. 1 includes edge nodes 102-1, 102-2, … …), which may also be referred to as content delivery networks (Content Delivery Network, CDNs), are typically deployed near clients 101, forwarding information between cloud nodes 103 and clients 101 with computational power intermediate between clients 101 and cloud nodes 103, and may handle relatively complex mathematical computation and data analysis tasks, and edge nodes 102 may be, for example, edge servers;
Yun Jiedian 103, which may also be referred to as an internet data center (Internet Data Center, IDC), is a central node of cloud computing and a management end of the edge node 102, has strong computing power and mass storage resources, and can be responsible for processing complex computing tasks.
In the present disclosure, the computing power of the edge node 102 or the cloud node 103 may be used to provide the data needed by the client 101 for the client 101 in consideration of the limited computing power of the client 101, and further, the cloud node 103 is far from the client 101 and has a larger network delay in consideration of the high computing power, so the present disclosure mainly considers the use of the computing power of the edge node 102 to provide the data needed by the client 101.
Fig. 2 is a flowchart of an information recommendation method according to an exemplary embodiment. Referring to fig. 2, the method may be performed by a client, the method including:
in step 201, real-time feature information is obtained based on an operation behavior of the user on at least one first recommendation information issued by the cloud node.
In the embodiment of the disclosure, the cloud node may determine, through the content recommendation system, one or more first recommendation information for the user based on the total historical feature information of the user stored on the cloud node, and further, the client receives the one or more first recommendation information issued by the cloud node.
In one example, the historical feature information is feature information that characterizes historical operational behavior of the user.
In one example, the full-scale historical feature information may include feature information composed of all of the user's historical operating behaviors.
In one example, the first recommendation information may be information that may be of interest to the user, video, audio, merchandise, interests, links, and the like.
In one example, the first recommendation information may be personalized recommendation information determined by the cloud node based on the user's full amount of historical operational data.
The client can acquire the real-time operation behavior information of the user on the first recommended information in real time, and perform characteristic calculation on the first recommended information to obtain the real-time characteristic information.
In one example, the real-time feature information may be feature information characterizing real-time operational behavior of the user.
In one example, the user's action on the first recommendation information in real time may include, but is not limited to, browsing, clicking, converting, etc. actions on the first recommendation information, and the converting action may include, but is not limited to, praise, negatively evaluating, forwarding, collecting, etc.
In one example, feature computation may include, but is not limited to, a classification process and/or a filtering process.
The client may filter the operation behavior information, and filter out a part of information having a smaller association with the operation behavior, so as to obtain real-time feature information that can more accurately represent the real-time operation behavior of the user.
For example, the client may classify the above-mentioned operation behavior information according to at least one of a preset period, a type of the operation behavior, and a type of the first recommendation information.
For example, the preset period is 2 hours (of course, other durations may also be used), and the client may perform classification processing on the operation behavior information of the user according to the 2 hours as a period.
For another example, the client classifies the operation behavior information according to the behavior of the user's approval and negative evaluation.
For another example, the client performs classification processing on the operation behavior information according to video, audio and commodities according to the type of the first recommendation information issued by the cloud node.
The foregoing is merely exemplary, and the client may perform other feature calculations on the above operation behavior information to obtain the real-time feature information, which is not limited in this disclosure.
In step 202, the real-time feature information is sent to an edge node.
In the embodiment of the disclosure, the client can provide the real-time characteristic information to the edge node so that the edge node can obtain the target characteristic information by utilizing own computing power.
In one example, the target characteristic information is characteristic information for characterizing commonalities between the user's historical operational behavior and the real-time operational behavior.
In one example, the target feature information may be a generic ebedding feature extracted by the edge node with an embedded service (embedding service) of the neural network that is capable of covering the user's historical and real-time operational behavior.
The embedded service of the neural network may refer to an operation performed by an embedded layer of the neural network, and may specifically be general feature extraction of an input value. In the embodiment of the disclosure, general features between the historical operation behaviors and the real-time operation behaviors are extracted.
In step 203, the target feature information returned by the edge node is received and stored.
In the embodiment of the disclosure, the client may receive the target feature information returned by the edge node and store the target feature information. Because the target characteristic information filters out the characteristic information which does not have commonality between the historical operation behavior and the real-time operation behavior, the operation behavior of the user can be represented more accurately, the occupied memory space is smaller, and compared with the case that the client stores the real-time characteristic information, the memory space of the client can be effectively saved.
In step 204, based on the target feature information, at least one second recommendation information issued by the cloud node is output after being reordered.
In the embodiment of the disclosure, after the edge node obtains the target characteristic information, the target characteristic information can be synchronized to the cloud node, so that the cloud node can update the full-scale historical characteristic information of the user stored by the cloud node in time. Further, the cloud node determines one or more second recommendation information for the user based on the updated historical characteristic information and then sends the second recommendation information to the client.
The client can input the target characteristic information provided by the edge node into a target neural network model on the client, obtain an output result of the target neural network model, and reorder and output at least one piece of second recommended information based on the output result and the type of each piece of second recommended information issued by the cloud node.
In one example, the target neural network model is used to optimally order the multiple referral information according to the user's operational behavior.
In one example, the second recommendation information may be information that may be of interest to the user, video, audio, merchandise, interests, links, and the like.
In one example, the second recommendation information may be personalized recommendation information determined by the cloud node based on the user's most recent full-scale historical operational behavior.
In one example, the type of the second recommendation information may be the same as or different from the type of the first recommendation information. For example, the cloud node recommends, based on the historical operation behavior, the first recommendation information for the client is a news-like video, and the second recommendation information may be a news-like video, or may be another type of video.
In one example, the client may reorder the second recommendation information according to an optimal ordering provided by the target neural network model, and a type of one or more second recommendation information newly issued by the cloud node.
For example, the client determines that the user has the greatest interest in a certain type of video or commodity based on the output result, and may preferentially recommend the video or commodity corresponding to the type in the second recommendation information.
Illustratively, on a certain page, for example, the first page of the website, the first page of the application program, the transaction class page of the application program, etc., the page may be refreshed after the second recommendation information is reordered.
For example, the second recommendation information is a link key on the page, the client may refresh the page after reordering the plurality of link keys, and the link key on the refreshed page is one or more link keys that are of most interest to the user.
In the embodiment of the disclosure, the initial neural network is trained by the client itself in consideration of limited computing power of the client and limited stored sample information, so that the target neural network model can occupy a large amount of client resources, and the finally obtained target neural network model is not accurate enough. Therefore, the training process of the target neural network model can be completed on the cloud node, and the network parameters of the target neural network model after the training process are provided for the client.
The client may set network parameters of the initial target neural network model to network parameters provided by the cloud nodes, thereby constructing the target neural network model on the client.
In addition, the cloud node can periodically update the network parameters of the target neural network model and provide the updated network parameters to the client so that the client synchronously updates the target neural network model of the cloud node.
In the above embodiment, the computing power of the edge node can be used for providing the target feature information for the client to more accurately represent the user operation behavior, so that the richness of the feature information for representing the user operation behavior on the client is expanded, and the time delay of information recommendation can be reduced because the end node is close to the client.
Fig. 3 is a flowchart of an information recommendation method according to an exemplary embodiment. Referring to fig. 3, the method may be performed by an edge node, the method comprising:
in step 301, real-time feature information sent by a client is received.
In the embodiment of the disclosure, the client may collect the real-time operation behavior information of the user on the first recommended information in real time, perform feature calculation on the operation behavior information, obtain the real-time feature information, and send the real-time feature information to the edge node.
The process of obtaining the real-time feature information by the client is similar to the above step 201, and will not be described herein.
In step 302, historical feature information of the user provided by a cloud node is obtained.
In the embodiment of the disclosure, the edge node may pull the historical characteristic information of the user from the cloud node.
In one example, the historical feature information may correspond to the real-time feature information described above, feature information that characterizes the user's historical operating behavior.
In one example, the edge node may pull the user's full amount of historical feature information from the cloud node. The full-scale historical feature information may include feature information composed of all of the historical operating behaviors of the user.
In step 303, based on the real-time feature information and the historical feature information, target feature information is obtained and then sent to the client and the cloud node.
In one example, the target characteristic information is characteristic information for characterizing commonalities between the user's historical operational behavior and the real-time operational behavior.
In one example, the edge node may invoke an embedding service to extract a generic enabling feature from the real-time feature information and the historical feature information that can cover the user's historical operational behavior and real-time operational behavior, resulting in the target feature information.
The embedded service of the neural network may refer to an operation performed by an embedded layer of the neural network, and may specifically be general feature extraction of an input value. In the embodiment of the disclosure, general characteristics of historical operation behaviors and real-time operation behaviors are extracted.
In one example, the neural network model employed by the edge node to obtain the target feature information may be a large artificial intelligence (Artificial Intelligence, AI) model. The AI large model can increase the depth and width of the model by storing more parameters, so that the expressive capacity of the model is improved, the parameters start from billions, a large amount of data is trained, and a high-quality prediction result is generated.
In one possible implementation, the edge node may send the target feature information to the client, so that the client uses the target feature information as an input value, and inputs the input value into a target neural network model on the client, to obtain an optimal ranking for the multiple analogy information according to the operation behavior of the user.
In one possible implementation, the edge node may send the target feature information to Yun Jiedian so that the cloud node synchronizes the user's full history feature information based on the target feature information.
In the embodiment, the computing power of the edge node can be utilized to provide the target characteristic information for representing the user operation behavior for the client, the resources of each device of the end-edge cloud can be utilized more reasonably, and the usability of the end-edge cloud architecture is improved.
Fig. 4 is a flowchart of an information recommendation method according to an exemplary embodiment. Referring to fig. 4, the method may be performed by a cloud node, the method comprising:
in step 401, at least one first recommendation information is issued to the client based on the historical feature information of the user.
In one example, the historical feature information is feature information that characterizes historical operational behavior of the user.
In the embodiment of the disclosure, the cloud node may determine, based on the total historical feature information of the user stored in the cloud node, first recommendation information issued to the client and then issue the first recommendation information to the client.
In one example, the full-scale historical feature information may include feature information composed of all of the user's historical operating behaviors.
In one example, the first recommendation information may be information that may be of interest to the user, video, audio, merchandise, interests, links, and the like.
In one example, the first recommendation information may be personalized recommendation information determined by the cloud node based on the user's full amount of historical operational data.
In step 402, the historical feature information of the user is sent to an edge node.
In the embodiment of the disclosure, the cloud node may provide the total historical feature information of the user to the edge node, so that the edge node obtains the target feature information based on the real-time feature information and the historical feature information and sends the target feature information to the client and the cloud node. The specific process of obtaining the target feature information by the edge node is similar to the process of step 303, and will not be described herein.
In step 403, the history feature information of the user is synchronously updated based on the target feature information sent by the edge node.
In the embodiment of the disclosure, the cloud node updates the full-scale historical characteristic information of the user stored on the cloud node based on the target characteristic information sent by the edge node.
In step 404, at least one second recommendation information is issued to the client based on the updated historical feature information.
In one example, the second recommendation information may be information that may be of interest to the user, video, audio, merchandise, interests, links, and the like.
In one example, the second recommendation information may be personalized recommendation information determined by the cloud node based on the updated full-scale historical operational data.
After receiving the second recommendation information, the client may execute step 204, and reorder and output at least one second recommendation information issued by the cloud node based on the target feature information. The specific process is not described here in detail.
In one possible implementation, the cloud node may perform supervised training on the initial neural network model based on the full-scale historical feature information of the user, and obtain the target neural network model when the end training condition is satisfied (e.g., the difference between the output result and the sample label is no longer reduced or reaches the expected number of iterative training). Further, the cloud node may issue network parameters of the target neural network model to the client so that the client builds the target neural network model.
In one possible implementation manner, the cloud node may periodically update and train the target neural network model based on the updated full-scale historical feature information, obtain updated network parameters, and send the updated network parameters to the client, so that the client updates the target neural network model in time.
In the above embodiment, the cloud node may provide the historical feature information of the user stored in the cloud node to the edge node, so that the edge node calculates the target feature information, and the target feature information may more accurately represent the operation behavior of the user. Resources of all devices of the end-edge cloud are reasonably utilized, and the availability of an end-edge cloud architecture is improved.
The above-described process is further illustrated below.
Fig. 5 is a flowchart of an information recommendation method according to an exemplary embodiment. Referring to fig. 5, the method is applicable to the system shown in fig. 1, and may include the following steps:
in step 501, the client 101 receives at least one first feature information issued by the cloud node 103.
The implementation of step 501 is similar to that of step 401 described above, and will not be described in detail here.
In step 502, the client 101 obtains real-time feature information based on the operation behavior of the user on at least one first recommendation information issued by the cloud node 103.
For example, the terminal feature module on the client 101 may collect the operation behavior information of the user in real time, and perform feature calculation on the operation behavior information to obtain real-time feature information.
The specific implementation of step 501 is similar to that of step 201 described above, and will not be described again here.
In step 503, the client 101 sends real-time feature information to the edge node 102.
In step 504, after the edge node 102 pulls the full-scale historical feature information of the user from Yun Jiedian, the target feature information is obtained based on the real-time feature information and the historical feature information.
For example, embedded service computing enhanced feature information may be invoked.
Illustratively, the edge node 102 uses the AI large model to obtain the target feature information.
The specific implementation of step 504 is similar to that of step 303, and is not described here again.
In step 505, the edge node 102 synchronizes the target feature information to the cloud node 103.
Cloud node 103 may synchronously update its own stored full-scale historical feature information based on the target feature information.
In step 506, the edge node 102 synchronizes the target feature information to the client 101.
In step 507, the client 101 reorders and outputs at least one second recommendation information.
The specific implementation of step 507 is similar to that of step 204 described above, and will not be described again here.
In step 508, the target neural network model is trained and updated by the cloud node 103, and the client 101 and the cloud node 103 maintain synchronization of network parameters of the target neural network model.
In the above embodiment, when the complexity of the intelligent scene of the page end, the data volume and the like are limited due to the computational effort of the client end, and the end feature processing needs to be degraded at certain moments or in certain scenes, the strong edge computational effort of the edge node is introduced, so that the computational effort short board of the end feature processing is compensated. The richness of the end features is expanded, the edge nodes can combine the total historical feature information of the user on the cloud nodes and the real-time feature information provided by the client in real time to jointly determine the general feature information, the basic features of the user are covered by the general feature information, the general features can be directly processed by a plurality of industry tasks, and an embedded layer in a neural network model can be directly used, so that the implementation is simple and convenient. The repeated calculation amount of the neural network model on the client to the feature processing is reduced, and the processing capacity of the neural network is improved.
Referring to fig. 6, the information recommendation device may be applied to a client to implement the technical solution of the present specification. Wherein, the information recommending apparatus may include:
the first processing module 601 is configured to obtain real-time feature information based on an operation behavior of at least one first recommendation information issued by a user to a cloud node, where the real-time feature information is feature information for characterizing the real-time operation behavior of the user;
A first sending module 602, configured to send the real-time feature information to an edge node;
a first receiving module 603, configured to receive and store target feature information returned by the edge node, where the target feature information is feature information that characterizes commonalities between the historical operation behavior of the user and the real-time operation behavior;
and the second processing module 604 is configured to reorder and output at least one second recommendation information issued by the cloud node based on the target feature information.
Referring to fig. 7, the information recommending apparatus may be applied to an edge node to implement the technical solution of the present specification. Wherein, the information recommending apparatus may include:
the second receiving module 701 is configured to receive real-time feature information sent by the client, where the real-time feature information is feature information used to characterize real-time operation behavior of the user;
a third receiving module 702, configured to obtain historical feature information of the user provided by a cloud node, where the historical feature information is feature information used to characterize a historical operation behavior of the user;
and a third processing module 703, configured to obtain target feature information based on the real-time feature information and the historical feature information, and then synchronize the target feature information with the client and the cloud node, where the target feature information is feature information that characterizes commonalities between the historical operation behavior and the real-time operation behavior of the user.
Referring to fig. 8, the information recommendation device may be applied to a cloud node to implement the technical solution of the present specification. Wherein, the information recommending apparatus may include:
a second sending module 801, configured to send at least one first recommendation information to a client based on historical feature information of a user, so that the client obtains real-time feature information, where the historical feature information is feature information used to characterize a historical operation behavior of the user, and the real-time feature information is feature information used to characterize a real-time operation behavior of the user on at least one first recommendation information;
a third sending module 802, configured to send, to an edge node, historical feature information of the user, so that the edge node obtains target feature information based on the real-time feature information and the historical feature information, where the target feature information is feature information that characterizes commonalities between a historical operation behavior of the user and the real-time operation behavior;
an updating module 803, configured to synchronously update history feature information of the user based on the target feature information sent by the edge node;
and a fourth sending module 804, configured to issue at least one second recommendation information to the client based on the updated historical feature information, so that the client reorders and outputs at least one second recommendation information based on the target feature information.
The embodiment of the disclosure further provides an information recommendation system, for example, as shown in fig. 1, including:
a client 101, where the client 101 is configured to implement an information recommendation method as shown in fig. 2;
an edge node 102, where the edge node 102 is configured to implement an information recommendation method as shown in fig. 3;
yun Jiedian 103, wherein the cloud node 103 is used for implementing the information recommendation method shown in fig. 4.
The system, apparatus, module or unit set forth in the above embodiments may be implemented in particular by a computer chip or entity, or by a product having a certain function. A typical implementation device is a computer, which may be in the form of a personal computer, laptop computer, cellular telephone, camera phone, smart phone, personal digital assistant, media player, navigation device, email device, game console, tablet computer, wearable device, or a combination of any of these devices.
In a typical configuration, a computer includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.
The memory may include volatile memory in a computer-readable medium, random Access Memory (RAM) and/or nonvolatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). Memory is an example of computer-readable media.
Computer readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of storage media for a computer include, but are not limited to, phase change memory (PRAM), static Random Access Memory (SRAM), dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), read Only Memory (ROM), electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, read only compact disc read only memory (CD-ROM), digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic disk storage, quantum memory, graphene-based storage or other magnetic storage devices, or any other non-transmission medium, which can be used to store information that can be accessed by the computing device. Computer-readable media, as defined herein, does not include transitory computer-readable media (transmission media), such as modulated data signals and carrier waves.
Fig. 9 is a schematic structural diagram of an electronic device provided in an exemplary embodiment. Referring to fig. 9, at a hardware level, the device includes a processor 902, an internal bus 904, a network interface 906, a memory 908, and a nonvolatile storage 910, although other hardware required by other services is also possible. One or more embodiments of the present description may be implemented in a software-based manner, such as by the processor 902 reading a corresponding computer program from the non-volatile memory 910 into the memory 908 and then running. Of course, in addition to software implementation, one or more embodiments of the present disclosure do not exclude other implementation manners, such as a logic device or a combination of software and hardware, etc., that is, the execution subject of the following processing flow is not limited to each logic unit, but may also be hardware or a logic device.
The electronic device may be the client 101, the edge node 102, or the cloud node 103 described above, which is not limited by the present disclosure.
It should also be noted that 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 one … …" does not exclude the presence of other like elements in a process, method, article or apparatus that comprises the element.
The foregoing describes specific embodiments of the present disclosure. Other embodiments are within the scope of the following claims. In some cases, the actions or steps recited in the claims can be performed in a different order than in the embodiments and still achieve desirable results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
The terminology used in the one or more embodiments of the specification is for the purpose of describing particular embodiments only and is not intended to be limiting of the one or more embodiments of the specification. As used in this specification, one or more embodiments and the appended claims, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It should also be understood that the term "and/or" as used herein refers to and encompasses any or all possible combinations of one or more of the associated listed items.
It should be understood that although the terms first, second, third, etc. may be used in one or more embodiments of the present description to describe various information, these information should not be limited to these terms. These terms are only used to distinguish one type of information from another. For example, first information may also be referred to as second information, and similarly, second information may also be referred to as first information, without departing from the scope of one or more embodiments of the present description. The word "if" as used herein may be interpreted as "at … …" or "at … …" or "responsive to a determination", depending on the context.
The foregoing description of the preferred embodiment(s) is (are) merely intended to illustrate the embodiment(s) of the present invention, and it is not intended to limit the embodiment(s) of the present invention to the particular embodiment(s) described.
Claims (18)
1. An information recommendation method, comprising:
based on the operation behavior of at least one first recommendation information issued by a user to a cloud node, obtaining real-time characteristic information, wherein the real-time characteristic information is characteristic information for representing the real-time operation behavior of the user;
transmitting the real-time characteristic information to an edge node;
receiving and storing target characteristic information returned by the edge node, wherein the target characteristic information is characteristic information used for representing commonalities between historical operation behaviors of the user and the real-time operation behaviors;
and based on the target characteristic information, reordering at least one piece of second recommended information issued by the cloud node and outputting the reordered second recommended information.
2. The method of claim 1, wherein the obtaining real-time feature information based on the operation behavior of the user on the at least one first recommendation information issued by the cloud node includes:
Collecting real-time operation behavior information of the user on at least one piece of first recommended information;
and performing feature calculation on the operation behavior information to obtain the real-time feature information.
3. The method of claim 2, wherein the performing feature computation on the operational behavior information comprises:
and classifying the real-time characteristic information according to at least one of a preset period, the type of the operation behavior and the type of the first recommendation information.
4. The method of claim 1, wherein the reordering the at least one second recommendation information issued by the cloud node based on the target feature information, and outputting the reordered at least one second recommendation information, comprises:
inputting the target characteristic information into a target neural network model on a client to obtain an output result of the target neural network model, wherein the target neural network model is used for optimally sequencing the multiple analogy information according to the operation behaviors of the user;
and reordering at least one piece of second recommended information based on the output result and the type of each piece of second recommended information, and outputting the reordered second recommended information.
5. The method according to claim 4, wherein the method further comprises:
Receiving network parameters of the target neural network model sent by a cloud node;
and constructing or synchronously updating the target neural network model on the client based on the network parameters.
6. An information recommendation method, comprising:
receiving real-time characteristic information sent by a client, wherein the real-time characteristic information is characteristic information for representing real-time operation behaviors of a user;
acquiring historical characteristic information of the user, which is provided by a cloud node, wherein the historical characteristic information is characteristic information used for representing historical operation behaviors of the user;
and based on the real-time characteristic information and the historical characteristic information, obtaining target characteristic information, and synchronizing the target characteristic information to the client and the cloud node, wherein the target characteristic information is characteristic information for representing commonalities between the historical operation behaviors of the user and the real-time operation behaviors.
7. The method of claim 6, wherein the deriving the target feature information based on the real-time feature information and the historical feature information comprises:
and extracting common embedded characteristic information from the real-time characteristic information and the historical characteristic information to obtain the target characteristic information.
8. An information recommendation method, comprising:
based on historical characteristic information of a user, at least one first recommendation information is issued to a client so that the client obtains real-time characteristic information, wherein the historical characteristic information is characteristic information used for representing historical operation behaviors of the user, and the real-time characteristic information is characteristic information used for representing real-time operation behaviors of the user on at least one first recommendation information;
transmitting the historical characteristic information of the user to an edge node so that the edge node obtains target characteristic information based on the real-time characteristic information and the historical characteristic information, wherein the target characteristic information is characteristic information representing commonalities between the historical operation behaviors of the user and the real-time operation behaviors;
based on the target characteristic information sent by the edge node, synchronously updating the history characteristic information of the user;
and based on the updated historical characteristic information, at least one piece of second recommendation information is issued to the client so that the client reorders and outputs the at least one piece of second recommendation information based on the target characteristic information.
9. The method of claim 8, wherein the method further comprises:
training an initial neural network model on a cloud node based on the historical characteristic information of the user to obtain a target neural network model, wherein the target neural network model is used for optimally sequencing the multiple analogy information according to the operation behaviors of the user;
and sending the network parameters of the target neural network model to the client so that the client establishes the target neural network model.
10. The method according to claim 9, wherein the method further comprises:
updating and training the target neural network model on the cloud node based on the updated historical characteristic information;
and sending the updated network parameters of the target neural network model to the client so that the client synchronously updates the target neural network model on the client.
11. An information recommendation device, characterized by comprising:
the cloud node comprises a first processing module, a second processing module and a third processing module, wherein the first processing module is used for obtaining real-time characteristic information based on the operation behavior of at least one first recommendation information issued by a user to a cloud node, and the real-time characteristic information is used for representing the real-time operation behavior of the user;
The first sending module is used for sending the real-time characteristic information to an edge node;
the first receiving module is used for receiving and storing target characteristic information returned by the edge node, wherein the target characteristic information is characteristic information representing commonality between the historical operation behavior of the user and the real-time operation behavior;
and the second processing module is used for reordering at least one piece of second recommended information issued by the cloud node based on the target characteristic information and outputting the reordered second recommended information.
12. An information recommendation device, characterized by comprising:
the second receiving module is used for receiving real-time characteristic information sent by the client, wherein the real-time characteristic information is characteristic information used for representing real-time operation behaviors of the user;
the third receiving module is used for acquiring historical characteristic information of the user, which is provided by the cloud node, wherein the historical characteristic information is characteristic information used for representing historical operation behaviors of the user;
and the third processing module is used for obtaining target characteristic information based on the real-time characteristic information and the historical characteristic information and then synchronizing the target characteristic information to the client and the cloud node, wherein the target characteristic information is characteristic information for representing commonality between the historical operation behavior of the user and the real-time operation behavior.
13. An information recommendation device, characterized by comprising:
the second sending module is used for sending at least one first recommendation message to the client based on the historical characteristic information of the user so that the client can obtain real-time characteristic information, wherein the historical characteristic information is characteristic information used for representing the historical operation behavior of the user, and the real-time characteristic information is characteristic information used for representing the real-time operation behavior of the user on at least one first recommendation message;
the third sending module is used for sending the historical characteristic information of the user to an edge node so that the edge node obtains target characteristic information based on the real-time characteristic information and the historical characteristic information, wherein the target characteristic information is characteristic information representing the commonality between the historical operation behavior of the user and the real-time operation behavior;
the updating module is used for synchronously updating the history characteristic information of the user based on the target characteristic information sent by the edge node;
and the fourth sending module is used for sending at least one piece of second recommendation information to the client based on the updated historical characteristic information so that the client reorders and outputs the at least one piece of second recommendation information based on the target characteristic information.
14. A client, comprising:
a processor;
a memory for storing processor-executable instructions;
wherein the processor implements the information recommendation method of any of claims 1-5 by executing the executable instructions.
15. An edge node, comprising:
a processor;
a memory for storing processor-executable instructions;
wherein the processor implements the information recommendation method of claim 6 or 7 by executing the executable instructions.
16. A cloud node, comprising:
a processor;
a memory for storing processor-executable instructions;
wherein the processor is configured to implement the information recommendation method of any of claims 8-10 by executing the executable instructions.
17. An information recommendation system, comprising:
a client for implementing the information recommendation method according to any one of claims 1-5;
an edge node for implementing the information recommendation method according to claim 6 or 7;
cloud node for implementing the information recommendation method according to any of claims 8-10.
18. A computer readable storage medium having stored thereon computer instructions which, when executed by a processor, implement the steps of the information recommendation method according to any of claims 1-10.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310887568.9A CN116610873B (en) | 2023-07-19 | 2023-07-19 | Information recommendation method and device and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310887568.9A CN116610873B (en) | 2023-07-19 | 2023-07-19 | Information recommendation method and device and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116610873A true CN116610873A (en) | 2023-08-18 |
| CN116610873B CN116610873B (en) | 2023-09-26 |
Family
ID=87683920
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310887568.9A Active CN116610873B (en) | 2023-07-19 | 2023-07-19 | Information recommendation method and device and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116610873B (en) |
Citations (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105320706A (en) * | 2014-08-05 | 2016-02-10 | 阿里巴巴集团控股有限公司 | Processing method and device of search result |
| CN106951496A (en) * | 2017-03-15 | 2017-07-14 | 微鲸科技有限公司 | A kind of list ordering device and method |
| CN107203530A (en) * | 2016-03-16 | 2017-09-26 | 北大方正集团有限公司 | Information recommendation method |
| KR20180121466A (en) * | 2017-04-06 | 2018-11-07 | 네이버 주식회사 | Personalized product recommendation using deep learning |
| CN110941773A (en) * | 2019-10-11 | 2020-03-31 | 北京华跃博弈科技有限公司 | Information recommendation method based on Internet of things |
| CN112084413A (en) * | 2020-09-16 | 2020-12-15 | 腾讯科技（深圳）有限公司 | Information recommendation method and device and storage medium |
| CN112650946A (en) * | 2020-12-30 | 2021-04-13 | 中国联合网络通信集团有限公司 | Product information recommendation method, device and system and storage medium |
| CN113536103A (en) * | 2020-04-21 | 2021-10-22 | 北京达佳互联信息技术有限公司 | Information recommendation method and device, electronic equipment and storage medium |
| CN113642990A (en) * | 2021-08-10 | 2021-11-12 | 江苏大学 | Agricultural machinery intelligent management and scheduling system and method based on cloud edge cooperation |
| CN113722588A (en) * | 2021-08-12 | 2021-11-30 | 北京达佳互联信息技术有限公司 | Resource recommendation method and device and electronic equipment |
| CN113766016A (en) * | 2021-08-30 | 2021-12-07 | 中国银联股份有限公司 | Information pushing method and server |
| WO2022016409A1 (en) * | 2020-07-22 | 2022-01-27 | 宁波吉利汽车研究开发有限公司 | Virtual vehicle key management method and system based on internet of vehicles |
| CN114548723A (en) * | 2022-02-15 | 2022-05-27 | 南京邮电大学 | Real-time vehicle and goods matching and path recommendation method and system based on cloud edge cooperation |
| WO2022142519A1 (en) * | 2020-12-30 | 2022-07-07 | 北京锐安科技有限公司 | Information recommendation method and apparatus, and electronic device and storage medium |
| WO2022258789A1 (en) * | 2021-06-09 | 2022-12-15 | Dover Europe Sàrl | Methods, systems and devices for implementing remote services |
| CN116108267A (en) * | 2022-12-19 | 2023-05-12 | 华为技术有限公司 | Recommendation method and related equipment |
| WO2023115807A1 (en) * | 2021-12-21 | 2023-06-29 | 北京达佳互联信息技术有限公司 | Information processing method and apparatus |
-
2023
- 2023-07-19 CN CN202310887568.9A patent/CN116610873B/en active Active
Patent Citations (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105320706A (en) * | 2014-08-05 | 2016-02-10 | 阿里巴巴集团控股有限公司 | Processing method and device of search result |
| CN107203530A (en) * | 2016-03-16 | 2017-09-26 | 北大方正集团有限公司 | Information recommendation method |
| CN106951496A (en) * | 2017-03-15 | 2017-07-14 | 微鲸科技有限公司 | A kind of list ordering device and method |
| KR20180121466A (en) * | 2017-04-06 | 2018-11-07 | 네이버 주식회사 | Personalized product recommendation using deep learning |
| CN110941773A (en) * | 2019-10-11 | 2020-03-31 | 北京华跃博弈科技有限公司 | Information recommendation method based on Internet of things |
| CN113536103A (en) * | 2020-04-21 | 2021-10-22 | 北京达佳互联信息技术有限公司 | Information recommendation method and device, electronic equipment and storage medium |
| WO2022016409A1 (en) * | 2020-07-22 | 2022-01-27 | 宁波吉利汽车研究开发有限公司 | Virtual vehicle key management method and system based on internet of vehicles |
| CN112084413A (en) * | 2020-09-16 | 2020-12-15 | 腾讯科技（深圳）有限公司 | Information recommendation method and device and storage medium |
| CN112650946A (en) * | 2020-12-30 | 2021-04-13 | 中国联合网络通信集团有限公司 | Product information recommendation method, device and system and storage medium |
| WO2022142519A1 (en) * | 2020-12-30 | 2022-07-07 | 北京锐安科技有限公司 | Information recommendation method and apparatus, and electronic device and storage medium |
| WO2022258789A1 (en) * | 2021-06-09 | 2022-12-15 | Dover Europe Sàrl | Methods, systems and devices for implementing remote services |
| CN113642990A (en) * | 2021-08-10 | 2021-11-12 | 江苏大学 | Agricultural machinery intelligent management and scheduling system and method based on cloud edge cooperation |
| CN113722588A (en) * | 2021-08-12 | 2021-11-30 | 北京达佳互联信息技术有限公司 | Resource recommendation method and device and electronic equipment |
| CN113766016A (en) * | 2021-08-30 | 2021-12-07 | 中国银联股份有限公司 | Information pushing method and server |
| WO2023115807A1 (en) * | 2021-12-21 | 2023-06-29 | 北京达佳互联信息技术有限公司 | Information processing method and apparatus |
| CN114548723A (en) * | 2022-02-15 | 2022-05-27 | 南京邮电大学 | Real-time vehicle and goods matching and path recommendation method and system based on cloud edge cooperation |
| CN116108267A (en) * | 2022-12-19 | 2023-05-12 | 华为技术有限公司 | Recommendation method and related equipment |
Non-Patent Citations (2)
| Title |
|---|
| PANG, GUANGYAO: "Efficient Deep Reinforcement Learning-Enabled Recommendation", IEEE, pages 871 - 886 * |
| 熊丽荣;王玲燕;黄玉柱;: "一种联合LTR和社交网络的Top-k推荐方法", 小型微型计算机系统, no. 12, pages 2577 - 2584 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116610873B (en) | 2023-09-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20150278706A1 (en) | Method, Predictive Analytics System, and Computer Program Product for Performing Online and Offline Learning | |
| CN110413867B (en) | Method and system for content recommendation | |
| CN116757297A (en) | Method and system for selecting features of machine learning samples | |
| CN112380449B (en) | Information recommendation method, model training method and related device | |
| CN111783810B (en) | Method and device for determining attribute information of user | |
| CN111898698B (en) | Object processing method and device, storage medium and electronic equipment | |
| CN113034246A (en) | Financial product recommendation method and device | |
| CN116882520A (en) | Prediction method and system for predetermined prediction problem | |
| CN111783873A (en) | Incremental naive Bayes model-based user portrait method and device | |
| CN110598084A (en) | Object sorting method, commodity sorting device and electronic equipment | |
| US20230069999A1 (en) | Method and apparatus for updating recommendation model, computer device and storage medium | |
| CN111582341A (en) | User abnormal operation prediction method and device | |
| CN114297509A (en) | User interest vector generation method, product recommendation method, device, equipment and medium | |
| CN111353103A (en) | Method and apparatus for determining user community information | |
| CN116610873B (en) | Information recommendation method and device and storage medium | |
| CN111709766A (en) | User behavior prediction method and device, storage medium and electronic equipment | |
| CN113836388A (en) | Information recommendation method and device, server and storage medium | |
| CN109241249B (en) | Method and device for determining burst problem | |
| CN116245603A (en) | User tag processing method and device, cloud server and readable storage medium | |
| CN111292171B (en) | Financial product pushing method and device | |
| CN111368864A (en) | Identification method, availability evaluation method and device, electronic equipment and storage medium | |
| CN110232393B (en) | Data processing method and device, storage medium and electronic device | |
| CN112417290A (en) | Training method of book sorting push model, electronic equipment and storage medium | |
| CN112036418A (en) | Method and device for extracting user features | |
| CN116610868B (en) | Sample labeling method, end-edge cloud cooperative training method and device |
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 |