CN111445289A - Resource delivery method, device, equipment and storage medium - Google Patents

Resource delivery method, device, equipment and storage medium Download PDF

Info

Publication number
CN111445289A
CN111445289A CN202010246080.4A CN202010246080A CN111445289A CN 111445289 A CN111445289 A CN 111445289A CN 202010246080 A CN202010246080 A CN 202010246080A CN 111445289 A CN111445289 A CN 111445289A
Authority
CN
China
Prior art keywords
resource
information
user
tracking
target
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202010246080.4A
Other languages
Chinese (zh)
Inventor
郑文琛
杨强
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.)
WeBank Co Ltd
Original Assignee
WeBank Co Ltd
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 WeBank Co Ltd filed Critical WeBank Co Ltd
Priority to CN202010246080.4A priority Critical patent/CN111445289A/en
Publication of CN111445289A publication Critical patent/CN111445289A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00Commerce
    • G06Q30/02Marketing; Price estimation or determination; Fundraising
    • G06Q30/0241Advertisements
    • G06Q30/0251Targeted advertisements
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00Pattern recognition
    • G06F18/20Analysing
    • G06F18/21Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
    • G06F18/214Generating training patterns; Bootstrap methods, e.g. bagging or boosting
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06NCOMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00Machine learning

Abstract

The invention discloses a resource delivery method, a device, equipment and a storage medium, wherein the method comprises the following steps: acquiring user characteristic information obtained by tracking a target resource screen by a user; obtaining exposed resource information corresponding to the user characteristic information from a tracking information base in a matching manner; and determining the resource delivery content of the target resource screen according to the user characteristic information and the exposed resource information. The invention realizes the purpose of performing progressive resource exposure on the user in front of the resource screen in a targeted manner, and provides a progressive resource impression for the user instead of performing resource delivery without destination, thereby improving the effect of resource delivery, improving the user conversion rate of resources and reducing the cost of resource delivery.

Description

Resource delivery method, device, equipment and storage medium
Technical Field
The invention relates to the field of artificial intelligence, in particular to a resource delivery method, a resource delivery device, resource delivery equipment and a storage medium.
Background
With the development of computer technology, more and more technologies (big data, distributed, Blockchain, artificial intelligence, etc.) are applied to the financial field, and the traditional financial industry is gradually changing to financial technology (Fintech), but higher requirements are also put forward on the technologies due to the requirements of security and real-time performance of the financial industry.
The resource screen is mostly used for resource release for off-line resource release, and the released resource contents are mostly switched according to a fixed cyclic sequence, for example, the resource 1 is played, the resource 2 is played, the resource 3 is played, the resource is played in a cyclic manner, or a plurality of resources are played randomly. No matter which releasing mode is adopted, the existing offline resource releasing is not targeted, so that a large amount of cost is probably input for resource popularization, but very low resource reverberation is obtained, namely, the user conversion rate of resource releasing is low.
Disclosure of Invention
The invention mainly aims to provide a resource delivery method, a resource delivery device and a storage medium, and aims to solve the problem of low user conversion rate of the current offline resource delivery mode.
In order to achieve the above object, the present invention provides a resource delivering method, which comprises the following steps:
acquiring user characteristic information obtained by tracking a target resource screen by a user;
obtaining exposed resource information corresponding to the user characteristic information from a preset tracking information base in a matching manner;
and determining the resource delivery content of the target resource screen according to the user characteristic information and the exposed resource information.
Optionally, the exposed resource information includes historical release contents corresponding to the user characteristic information and historical interaction information corresponding to the historical release contents, and after the step of determining the resource release contents of the target resource screen according to the user characteristic information and the exposed resource information, the method further includes:
acquiring the current interaction information of the user corresponding to the user characteristic information based on the resource delivery content;
and associating the user characteristic information, the resource delivery content and the current interaction information, and then adding the associated information to the tracking information base so as to update the tracking information base.
Optionally, the step of determining the resource delivery content of the target resource screen according to the user characteristic information and the exposed resource information includes:
and inputting the user characteristic information and the exposed resource information into a target resource release selection model obtained by pre-training to obtain the resource release content of the target resource screen.
Optionally, before the step of inputting the user characteristic information and the exposed resource information into a target resource launching selection model obtained through pre-training to obtain the resource launching content of the target resource screen, the method further includes:
acquiring an experience data set of a reinforcement learning framework based on the tracking information base;
and performing reinforcement learning training on the resource delivery selection model to be trained based on the empirical data set to obtain a target resource delivery selection model.
Optionally, the step of obtaining an experience data set of a reinforcement learning framework based on the tracking information base includes:
acquiring target user characteristic information and target exposed resource information corresponding to the target user characteristic information from the tracking information base as state data;
acquiring target resource release content determined based on the target user characteristic information and the target exposed resource information from the tracking information base as action data;
target interaction information of a user corresponding to the target user characteristic information based on the target resource delivery content is acquired from the tracking information base and is used as reward data;
and taking the state data, the action data and the reward data as a piece of experience data, and obtaining an experience data set based on a plurality of pieces of experience data.
Optionally, after the step of determining the resource delivery content of the target resource screen according to the user characteristic information and the exposed resource information, the method further includes:
detecting whether the tracking information to be deleted exists in the tracking information base according to a preset rule;
and if the tracking information to be deleted exists, deleting the tracking information to be deleted from the tracking information base so as to update the tracking information base.
Optionally, the step of obtaining user characteristic information obtained by performing user tracking on the target resource screen includes:
acquiring tracking data acquired by user tracking equipment corresponding to the target resource screen;
and performing data analysis on the tracking data to obtain the user characteristic information.
Optionally, the user tracking device comprises at least a camera device.
In order to achieve the above object, the present invention further provides a resource delivering device, including:
the acquisition module is used for acquiring user characteristic information obtained by tracking a target resource screen by a user;
the matching module is used for matching and obtaining the exposed resource information corresponding to the user characteristic information from a preset tracking information base;
and the determining module is used for determining the resource releasing content of the target resource screen according to the user characteristic information and the exposed resource information.
In order to achieve the above object, the present invention further provides a resource delivering apparatus, including: a memory, a processor and a resource placement program stored on the memory and executable on the processor, the resource placement program when executed by the processor implementing the steps of the resource placement method as described above.
In addition, to achieve the above object, the present invention further provides a computer readable storage medium, which stores a resource delivering program, and when the resource delivering program is executed by a processor, the resource delivering program implements the steps of the resource delivering method as described above.
In the invention, the user characteristic information obtained by tracking the target resource screen by the user is obtained, the exposed resource information corresponding to the user characteristic information is obtained in the tracking information base in a matching way, and then the resource release content of the target resource screen is determined based on the user characteristic information and the exposed resource information. According to the resource delivery method and the resource delivery system, what the resource content to be exposed to the user next step is determined according to the user characteristic information of the user in front of the resource screen and the exposed resource information of the user, so that progressive resource delivery is achieved, namely, for the user in front of the resource screen, progressive resource exposure is performed in a targeted manner, a progressive resource impression is given to the user instead of the resource delivery performed without destination, and therefore the resource delivery effect is improved, the user conversion rate of the resource is improved, and the resource delivery cost is reduced.
Drawings
FIG. 1 is a schematic diagram of a hardware operating environment according to an embodiment of the present invention;
fig. 2 is a schematic flowchart of a resource delivering method according to a first embodiment of the present invention;
FIG. 3 is a block diagram of a resource delivering apparatus according to a preferred embodiment of the present invention.
The implementation, functional features and advantages of the objects of the present invention will be further explained with reference to the accompanying drawings.
Detailed Description
It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
As shown in fig. 1, fig. 1 is a schematic device structure diagram of a hardware operating environment according to an embodiment of the present invention.
It should be noted that, the resource delivering device in the embodiment of the present invention may be a smart phone, a personal computer, a server, and other devices, and is not limited herein.
As shown in fig. 1, the resource delivery apparatus may include: a processor 1001, such as a CPU, a network interface 1004, a user interface 1003, a memory 1005, a communication bus 1002. Wherein a communication bus 1002 is used to enable connective communication between these components. The user interface 1003 may include a Display screen (Display), an input unit such as a Keyboard (Keyboard), and the optional user interface 1003 may also include a standard wired interface, a wireless interface. The network interface 1004 may optionally include a standard wired interface, a wireless interface (e.g., WI-FI interface). The memory 1005 may be a high-speed RAM memory or a non-volatile memory (e.g., a magnetic disk memory). The memory 1005 may alternatively be a storage device separate from the processor 1001.
Those skilled in the art will appreciate that the configuration of the apparatus shown in fig. 1 does not constitute a limitation of the resource delivery apparatus and may include more or fewer components than those shown, or some components may be combined, or a different arrangement of components.
As shown in fig. 1, a memory 1005, which is a kind of computer storage medium, may include therein an operating system, a network communication module, a user interface module, and a resource delivery program. The operating system is a program for managing and controlling hardware and software resources of the device, and supports the running of a resource release program and other software or programs.
In the device shown in fig. 1, the user interface 1003 is mainly used for data communication with a client; the network interface 1004 is mainly used for establishing communication connection with a server; and the processor 1001 may be configured to call the resource delivery program stored in the memory 1005 and perform the following operations:
acquiring user characteristic information obtained by tracking a target resource screen by a user;
obtaining exposed resource information corresponding to the user characteristic information from a tracking information base in a matching manner;
and determining the resource delivery content of the target resource screen according to the user characteristic information and the exposed resource information.
Further, after the step of determining the resource delivery content of the target resource screen according to the user characteristic information and the exposed resource information, the processor 1001 may be further configured to call a resource delivery program stored in the memory 1005 to perform the following operations:
acquiring the current interaction information of the user corresponding to the user characteristic information based on the resource delivery content;
and associating the user characteristic information, the resource delivery content and the current interaction information, and then adding the associated information to the tracking information base so as to update the tracking information base.
Further, the step of determining the resource delivery content of the target resource screen according to the user characteristic information and the exposed resource information includes:
and inputting the user characteristic information and the exposed resource information into a target resource release selection model obtained by pre-training to obtain the resource release content of the target resource screen.
Further, before the step of inputting the user characteristic information and the exposed resource information into a target resource delivery selection model obtained by pre-training to obtain the resource delivery content of the target resource screen, the method further includes:
acquiring an experience data set of a reinforcement learning framework based on the tracking information base;
and performing reinforcement learning training on the resource delivery selection model to be trained based on the empirical data set to obtain a target resource delivery selection model.
Further, the step of obtaining an experience data set of a reinforcement learning framework based on the tracking information base comprises:
acquiring target user characteristic information and target exposed resource information corresponding to the target user characteristic information from the tracking information base as state data;
acquiring target resource release content determined based on the target user characteristic information and the target exposed resource information from the tracking information base as action data;
target interaction information of a user corresponding to the target user characteristic information based on the target resource delivery content is acquired from the tracking information base and is used as reward data;
and taking the state data, the action data and the reward data as a piece of experience data, and obtaining an experience data set based on a plurality of pieces of experience data.
Further, after the step of determining the resource delivery content of the target resource screen according to the user characteristic information and the exposed resource information, the processor 1001 may be further configured to call a resource delivery program stored in the memory 1005, and perform the following operations:
detecting whether the tracking information to be deleted exists in the tracking information base according to a preset rule;
and if the tracking information to be deleted exists, deleting the tracking information to be deleted from the tracking information base so as to update the tracking information base.
Further, the step of obtaining the user characteristic information obtained by tracking the target resource screen by the user includes:
acquiring tracking data acquired by user tracking equipment corresponding to the target resource screen;
and performing data analysis on the tracking data to obtain the user characteristic information.
Further, the user tracking apparatus includes at least an image pickup apparatus.
Based on the above structure, various embodiments of the resource delivery method are provided.
Referring to fig. 2, fig. 2 is a flowchart illustrating a resource delivering method according to a first embodiment of the present invention.
While a logical order is shown in the flow chart, in some cases, the steps shown or described may be performed in an order different than that shown. The executing subject of each embodiment of the resource delivery method of the present invention may be a device such as a smart phone, a personal computer, and a server, and the device may be referred to as a resource delivery device. In this embodiment, the resource delivery method includes:
step S10, user characteristic information obtained by tracking the target resource screen by the user is obtained;
in this embodiment, the resources displayed by various devices such as the screen may be various, for example, public service advertisements, commercial advertisements, important news information, notification messages that need to be advertised, and the like. It will be appreciated that whatever type of resource has a particular presentation purpose and will therefore also correspond to a range of target populations. For example, a commercial for displaying children's products, the target population may mainly include children of suitable age, teachers, parents, etc. For example, the target group of news information related to the leave-retirement policy may mainly include the elderly and children. For another example, the target group of public service advertisement for advocating green travel may be people in all communities.
In this embodiment, an off-line resource screen may be installed in a shopping mall, a street or other places suitable for resource delivery, the resource screen may be used to deliver resource information, and the specific type of resource screen is not limited in this embodiment, for example, an L ED (L bright Emitting Diode) display screen may be used.
Specifically, if resource delivery needs to be performed on a certain resource screen and resource content needs to be confirmed, the resource screen can be used as a target resource screen.
The user tracking can be performed on each resource screen, the user tracking can be a user who determines to pass through the resource screen, watch the resource content of the resource screen or interact with the resource content, the user characteristic information of the users can be obtained through the user tracking, and the user characteristic information can comprise information of the gender, the age, the dressing, ornaments, the same-row crowd, the face, the action, the gait and the like of the user; interaction information of a user watching certain resource content can be acquired, and the interaction information can include information of whether interaction is available and what interaction is performed. The information obtained by the user tracking may be collectively referred to as user tracking information.
Then, for the target resource screen, the user tracking may be performed on the target resource screen in the above-mentioned user tracking manner to obtain user characteristic information, and then the user characteristic information may include characteristic information of a user staying in front of or passing by the target resource screen.
It should be noted that, in order to ensure the privacy security of the user, it may be limited that the obtained personal features of the user are only used for cross-screen tracking in a specific shopping mall environment (for example, according to different faces, motions, and gaits captured by two cameras, whether the user is the same user), but are strictly prohibited to be used for communicating with other user databases (for example, a face identity card system), and the privacy of the user is strictly protected (that is, the system cannot determine the personal identity information of one user).
Further, the step S10 includes:
step S101, acquiring tracking data acquired by user tracking equipment corresponding to the target resource screen;
and step S102, carrying out data analysis on the tracking data to obtain the user characteristic information.
The user tracking mode may be that some user tracking devices, such as a camera, a distance sensor, an infrared sensor, or other sensor devices, are installed in front of the resource screen, or a camera device in a market is adopted, and images, sensor data, and the like in the front range of the resource screen are collected by the user tracking devices to obtain tracking data. The tracking data may be obtained from the user tracking device, and may be obtained in real time for real-time data. And then analyzing the tracking data to obtain user tracking information. The data analysis mode has various modes, if the data is image data, a target detection model based on a neural network can be adopted to identify character features and actions, so as to obtain user feature information, interaction information and the like, for example, a camera can be adopted to capture the attention of a user, including but not limited to whether the user watches a resource screen for more than a certain time, whether the user interacts with the content of the resource screen in an action (for example, scanning a code) and the like; if the data collected by the distance sensor are available, the data can be used for assisting image data, and the distance between each user and the screen is analyzed to serve as interactive information.
Step S20, obtaining exposed resource information corresponding to the user characteristic information from a preset tracking information base in a matching way;
the user tracking information obtained by tracking the user of each resource screen can be stored in a library form to obtain a tracking information library, the tracking information library can comprise a plurality of pieces of data, each piece of data can comprise user characteristic information, resource release content and interaction information, each piece of data can also correspond to one recording time, and one piece of data represents how a user interacts with certain resource release content at a certain time.
After the user characteristic information corresponding to the target resource screen is obtained, the user characteristic information is matched with each user characteristic information recorded in the tracking information base, if the user characteristic information is matched with each user characteristic information, the resource release content and the interaction information corresponding to the matched user characteristic information are obtained from the tracking information base, the corresponding recording time can also be obtained, and the obtained information can be used as the exposed resource information corresponding to the user characteristic information of the target resource screen. The method for matching the user characteristic information may be as follows: and comparing the user characteristic information with the user characteristic information corresponding to the target resource screen one by one for certain user characteristic information in the tracking information base to determine the matching degree of each characteristic, and if the matching degree of each characteristic is greater than the preset matching degree, determining that the user characteristic information is successfully matched with the user characteristic information corresponding to the target resource screen. It is understood that, as a result of the matching, a plurality of user feature information with successful matching may be obtained, indicating that the user corresponding to the user feature information is traced a plurality of times.
Take the advertisement screen of the market as an example for explanation: an advertisement screen is arranged on each of 3 floors of a certain shopping mall, and user tracking is carried out on each advertisement screen; when the user A passes through the first floor, tracking that the user corresponding to the user characteristic information 1 passes through the advertisement screen 1 of the first floor at the time t1, watching the advertisement 1, and staying for one minute, wherein the data is recorded in a tracking information base; then, the user A goes to the second floor, the user corresponding to the user characteristic information 2 is tracked to pass through the advertisement screen 2 of the second floor at the time t2, the advertisement 2 is watched, code scanning interaction is carried out, and the data is recorded into a tracking information base; then, the user a goes to the third floor, performs user tracking on the advertisement screen of the third floor, acquires the user characteristic information 3 at the time t3, performs matching in the tracking information base, and successfully matches both the user characteristic information 1 and the user characteristic information 2 with the user characteristic information 3, so that the data corresponding to the user characteristic information 1 and the data corresponding to the user characteristic information 2 can be acquired from the tracking information base as the exposed information corresponding to the user characteristic information 3, including: the user corresponding to user profile 3 watched ad 1 and stayed at time t1 for 1 minute, and watched ad 2 and performed a code-scanning interaction at time t 2.
If the user characteristic information is not matched in the tracking information base, it indicates that the user passes through the resource screen for the first time, and the exposed resource information of the user may be determined to be empty.
Step S30, determining the resource delivery content of the target resource screen according to the user characteristic information and the exposed resource information.
After the exposed resource information corresponding to the user characteristic information of the target resource screen is acquired, determining the resource release content of the target resource screen according to the user characteristic information and the exposed resource information, and releasing the resource release content on the target resource screen, that is, directly outputting the resource release content to the target resource screen, or sending a control instruction to the target resource screen, so that the target resource screen acquires the resource release content from a local or remote server of the target resource screen according to the control instruction and outputs the resource release content to the target resource screen. Specifically, a rule for determining resource delivery content from a resource library based on user characteristic information and exposed resource information may be preset, where the resource library has each resource previously manufactured; for example, when the exposed resource information is empty, it is determined that the resource release content to be released next is a resource corresponding to the user characteristic information, for example, if the user characteristic information indicates that the user is an adult, the fitness resource can be released, and if the user characteristic information indicates that the user is an adult carrying a child, the extracurricular training resource can be released; for another example, when the number of times that the user watches a certain resource is greater than the preset number of times in the exposed resource information, it is determined that the content of the resource release that needs to be released next is another resource related to the watched resource, and the content of the other resource may be a resource that is more detailed than the watched resource, so that the user can further know the resource information. When the exposed resource information further includes the interaction information, the rule may be set to be more complex, for example, when the exposed resource information indicates that the user only passes through the resource when viewing the resource without interaction, it is determined that the content of the resource to be released next is still the resource, and when the exposed resource information indicates that the user stays for a period of time when viewing the resource or performs interaction, it is determined that the content of the resource to be released next is a resource which is more detailed than the resource. That is, what the resource content to be exposed to the user next is determined according to the user characteristic information of the user and the exposed resource information of the user, so that progressive resource delivery is realized, that is, for the user in front of the resource screen, progressive resource exposure is performed in a targeted manner, and a progressive resource impression is given to the user instead of performing resource delivery without destination, so that the effect of resource delivery is improved, the user conversion rate of the resource is improved, and the resource delivery cost is reduced.
For example, taking resources as an advertisement as an example, if it is determined that the user is a parent, a child is carried, and the user is just entering a market soon based on the user characteristic information, it is determined that the advertisement delivery content of the next step is "8 education training institutions in the mall"; if the user has accepted multiple exposures and there is positive feedback (i.e., "sweep code know more" etc. interactive interaction) for "there are 8 education training institutions in mall", then the next step of advertising is determined to be "langerhan provides the best personalized lesson for your children".
Further, it should be noted that, when the target resource screen is tracked by the user, user characteristic information of multiple users may be acquired, that is, multiple users may be in front of the target resource screen. Then a user profile may be locked from the plurality of user profiles and the resource placement may be determined based on the user profile. Specifically, exposed resource information corresponding to each user feature information can be matched from the tracking information base, exposure times and interaction information of each resource corresponding to each user feature information are determined according to each exposed resource information, one user feature information with the largest exposure times is locked from each user feature, or more user feature information is locked for interaction, that is, users most likely to be interested in the resource are mined from users in front of the current resource screen, and next-step resource delivery is performed for the users, so that user conversion is realized. Alternatively, a portion of the user characteristic information may be locked from the plurality of user characteristic information, and the resource placement content may be determined based on the portion of the user characteristic information. Specifically, a part of the user characteristic information that matches the type of the resource in the resource library may be selected from the plurality of user characteristic information, for example, if the resource library is a resource for male, the user characteristic information of the male user may be selected from the plurality of user characteristic information.
In this embodiment, the user characteristic information obtained by performing user tracking on the target resource screen is acquired, the exposed resource information corresponding to the user characteristic information is obtained by matching in the tracking information base obtained by performing user tracking, and then the resource delivery content of the target resource screen is determined based on the user characteristic information and the exposed resource information. In this embodiment, what the resource content to be exposed to the user next is determined according to the user characteristic information of the user in front of the resource screen and the exposed resource information of the user, so as to implement progressive resource delivery, that is, for the user in front of the resource screen, progressive resource exposure is performed on a targeted basis, and a progressive resource impression is given to the user instead of resource delivery performed on a destination-free basis, so that the effect of resource delivery is improved, the user conversion rate of the resource is improved, and the resource delivery cost is reduced.
Further, based on the first embodiment described above, a second embodiment of the resource delivering method of the present invention is provided, in this embodiment, the exposed resource information includes historical delivering contents corresponding to the user characteristic information, and historical interaction information corresponding to the historical delivering contents, respectively, and after step S30, the method further includes:
step S40, obtaining the current interaction information of the user corresponding to the user characteristic information based on the resource delivery content;
in this embodiment, the obtained exposed resource information corresponding to the user characteristic information may include historical release contents corresponding to the user characteristic information and historical interaction information corresponding to the historical release contents, that is, include each resource release content corresponding to the user characteristic information and interaction information corresponding to each resource release content, which are matched from the tracking information base.
After determining the resource releasing content of the target resource screen and releasing the resource releasing content on the target resource screen, acquiring the current interaction information of the user corresponding to the user characteristic information based on the resource releasing content for the user tracking based on the target resource screen to obtain the user characteristic information. Specifically, the user tracking device corresponding to the target resource screen may be used to track the user, acquire data acquired by the user tracking device, and analyze the acquired data to acquire the current interaction information. For example, during the period of delivering the resource delivery content, the multi-frame image data collected by the camera for a period of time may be analyzed respectively, and it is determined that all the multi-frame images identify certain user characteristic information, it indicates that the user corresponding to the user characteristic information stays in front of the target resource screen for a period of time, and this may be used as the current interaction information of the user corresponding to the user characteristic information based on the resource delivery content. For another example, during the period of delivering the resource delivery content, the image data acquired by the camera is analyzed, and a code scanning action of a user corresponding to certain user characteristic information in the image is identified, so that the code scanning action can be used as the current interaction information of the user corresponding to the user characteristic information based on the resource delivery content.
And step S50, adding the user characteristic information, the resource release content and the current interaction information into the tracking information base after associating, so as to update the tracking information base.
And associating the user characteristic information, the resource delivery content and the current interaction information, and adding the user characteristic information, the resource delivery content and the current interaction information as a piece of data to a tracking information base so as to update the tracking information base. Then, when the next resource release content of the target resource screen needs to be determined, or the next resource release content of other resource screens needs to be determined, the next resource release content may be determined based on the updated tracking information base in the manner described in the first embodiment.
Further, after step S30, the method further includes:
step S60, detecting whether the tracking information to be deleted exists in the tracking information base according to a preset rule;
when the number of resource screens is large, the number of users is large, and the types of the tracking information are large, the tracking information base is large, so that a large-capacity storage space is needed for storage, some useless tracking information in the tracking information base can be deleted, the occupation of the storage space is reduced, and the storage cost is reduced. Specifically, a detection rule may be set in advance as needed, and whether the tracking information to be deleted exists in the tracking information base or not may be detected according to the detection rule. The detection rule may be a rule that detection is performed periodically, for example, detection is performed every other day; the condition for determining the tracking information to be deleted in each detection may be that the recording time of the information exceeds a certain time length from the detection time, for example, the information recorded one week ago is determined as the information to be deleted; or, when the interaction information corresponding to a certain user feature information is non-interactive, it indicates that the user corresponding to the user feature information is not interested in the resource content, and the user feature information and the information associated with the user feature information may be both used as the tracking information to be deleted.
Step S70, if the tracking information to be deleted exists, deleting the tracking information to be deleted from the tracking information base to update the tracking information base.
If the tracking information to be deleted exists in the tracking information base, the determined tracking information to be deleted can be deleted from the tracking information base so as to update the tracking information base. Then, when the next resource release content of the target resource screen needs to be determined, or the next resource release content of other resource screens needs to be determined, the next resource release content may be determined based on the updated tracking information base in the manner described in the first embodiment.
Further, based on the first embodiment and the second embodiment, a third embodiment of the resource delivering method of the present invention is provided, in this embodiment, the step S30 includes:
step S301, inputting the user characteristic information and the exposed resource information into a target resource release selection model obtained by pre-training to obtain resource release contents of the target resource screen.
The method comprises the steps that a target resource release selection model can be trained in advance, the model can adopt a common machine learning model structure, input data of the target resource release selection model can be user characteristic information and exposed resource information corresponding to the user characteristic information, and output data can be resource release content. The training data used in the training process may be some data set by a human being according to experience, for example, a piece of training data may include a piece of user characteristic information, exposed resource information and resource release contents to be released, which are manually drawn up. The training mode of the model can adopt the existing machine learning model training method, and details are not repeated herein.
After user characteristic information corresponding to the target resource screen and exposed resource information corresponding to the user characteristic information are obtained, the two kinds of information are input into a target resource release selection model, and resource release content of the target resource screen is obtained.
In this embodiment, a target resource delivery selection model is trained in advance, and the content of resource delivery to be delivered next is determined through the model, so that the content of resource delivery is determined according to user characteristic information and exposed resource information without manual work, automation in a progressive resource delivery scheme is realized, the efficiency of resource delivery is improved, and the labor cost and the cost of resource delivery are reduced.
Further, before the step S30, the method further includes:
step S80, acquiring an experience data set of a reinforcement learning framework based on the tracking information base;
further, the target resource placement selection model may be a model obtained based on reinforcement learning training. Reinforcement learning, also known as refitting learning and evaluation learning, is an important machine learning method. In this embodiment, the experience data set of the reinforcement learning framework may be acquired based on a tracking information base obtained by tracking users for each resource screen.
Further, the step S80 includes:
step S801, acquiring target user characteristic information and target exposed resource information corresponding to the target user characteristic information from the tracking information base as state data;
step S802, obtaining target resource release content determined based on the target user characteristic information and the target exposed resource information from the tracking information base as action data;
step S803, target interaction information of a user corresponding to the target user characteristic information based on the target resource delivery content is acquired from the tracking information base and is used as reward data;
step S804, using the state data, the action data, and the reward data as a piece of experience data, and obtaining an experience data set based on a plurality of pieces of experience data.
Specifically, according to the characteristics of reinforcement learning, one piece of experience data may include states, actions and rewards; the method comprises the steps that target user characteristic information and target exposed resource information corresponding to the target user characteristic information can be obtained from a user tracking information base and serve as state data; acquiring target resource release content determined based on the target user characteristic information and the target exposed resource information from a user tracking information base as action data; target interaction information of a user corresponding to the target user characteristic information and based on the target resource delivery content is acquired from a user tracking information base and is used as reward data; taking the state data, the action data and the reward data as experience data; in a similar manner, a plurality of empirical data are obtained to form an empirical data set.
And step S90, performing reinforcement learning training on the resource delivery selection model to be trained based on the empirical data set to obtain a target resource delivery selection model.
And performing reinforcement learning training on the resource delivery selection model to be trained by adopting an empirical data set to obtain a target resource delivery selection model. The resource putting selection model to be trained can adopt a common reinforcement learning model. The way of performing the reinforcement learning training based on the experience data set may adopt a commonly used reinforcement learning training way, for example, an Actor-crititc algorithm of a reinforcement learning framework may be adopted, an action selection based on a state set and a reward set in the experience data set is used as the Actor, a reward (i.e., user interaction) brought by the selected action (i.e., what resource content is exposed) is used as the crititc, and the Actor is scored, so that the Actor is updated, so that the Actor is selected to optimize each step of the longest-term reward (i.e., user purchase conversion).
It should be noted that the target resource delivery selection model can be optimized periodically based on the information in the tracking information base, so that the resource delivery strategy can be continuously optimized in the using process, and the user conversion rate is continuously improved.
In addition, an embodiment of the present invention further provides a resource delivering device, and referring to fig. 3, the resource delivering device includes:
the acquisition module 10 is configured to acquire user characteristic information obtained by performing user tracking on a target resource screen;
the matching module 20 is configured to match the exposed resource information corresponding to the user feature information from a preset tracking information base;
and the determining module 30 is configured to determine the resource delivery content of the target resource screen according to the user characteristic information and the exposed resource information.
Further, the exposed resource information includes historical release contents corresponding to the user feature information, and historical interaction information corresponding to the historical release contents, respectively, and the obtaining module 10 is further configured to: acquiring interaction information of a user corresponding to the user characteristic information based on the resource delivery content;
the resource delivering device further comprises:
and the adding module is used for associating the user characteristic information, the resource delivery content and the interaction information and then adding the associated information to the tracking information base so as to update the tracking information base.
Further, the determining module 30 includes:
and the input unit is used for inputting the user characteristic information and the exposed resource information into a target resource release selection model obtained by pre-training to obtain the resource release content of the target resource screen.
Further, the obtaining module 10 is further configured to: acquiring an experience data set of a reinforcement learning framework based on the tracking information base;
the resource delivering device further comprises:
and the training module is used for carrying out reinforcement learning training on the resource delivery selection model to be trained based on the experience data set to obtain a target resource delivery selection model.
Further, the obtaining module 10 includes:
the first acquisition unit is used for acquiring target user characteristic information and target exposed resource information corresponding to the target user characteristic information from the tracking information base as state data;
a second obtaining unit, configured to obtain, from the tracking information base, target resource delivery content determined based on the target user feature information and the target exposed resource information as action data;
a third obtaining unit, configured to obtain, from the tracking information base, target interaction information that is performed by a user corresponding to the target user feature information based on the target resource delivery content, as reward data;
and the fourth acquisition unit is used for taking the state data, the action data and the reward data as one piece of experience data and obtaining an experience data set based on a plurality of pieces of experience data.
Further, the resource delivering apparatus further includes:
the detection module is used for detecting whether the tracking information to be deleted exists in the tracking information base according to a preset rule;
and the deleting module is used for deleting the tracking information to be deleted from the tracking information base if the tracking information to be deleted exists so as to update the tracking information base.
Further, the obtaining module 10 includes:
a fifth obtaining unit, configured to obtain tracking data collected by a user tracking device corresponding to the target resource screen;
and the analysis unit is used for carrying out data analysis on the tracking data to obtain the user characteristic information.
Further, the user tracking apparatus includes at least an image pickup apparatus.
The specific implementation of the resource delivering apparatus of the present invention is basically the same as the embodiments of the resource delivering method, and is not described herein again.
In addition, an embodiment of the present invention further provides a computer-readable storage medium, where a resource delivery program is stored on the storage medium, and when executed by a processor, the resource delivery program implements the steps of the resource delivery method as described below.
The embodiments of the resource delivery device and the computer-readable storage medium of the present invention can refer to the embodiments of the resource delivery method of the present invention, and are not described herein again.
It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.
The above-mentioned serial numbers of the embodiments of the present invention are merely for description and do not represent the merits of the embodiments.
Through the above description of the embodiments, those skilled in the art will clearly understand that the method of the above embodiments can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware, but in many cases, the former is a better implementation manner. Based on such understanding, the technical solutions of the present invention may be embodied in the form of a software product, which is stored in a storage medium (such as ROM/RAM, magnetic disk, optical disk) and includes instructions for enabling a terminal device (such as a mobile phone, a computer, a server, an air conditioner, or a network device) to execute the method according to the embodiments of the present invention.
The above description is only a preferred embodiment of the present invention, and not intended to limit the scope of the present invention, and all modifications of equivalent structures and equivalent processes, which are made by using the contents of the present specification and the accompanying drawings, or directly or indirectly applied to other related technical fields, are included in the scope of the present invention.

Claims (10)

1. A resource delivery method is characterized by comprising the following steps:
acquiring user characteristic information obtained by tracking a target resource screen by a user;
obtaining exposed resource information corresponding to the user characteristic information from a preset tracking information base in a matching manner;
and determining the resource delivery content of the target resource screen according to the user characteristic information and the exposed resource information.
2. The resource delivering method according to claim 1, wherein the exposed resource information includes historical delivering contents corresponding to the user characteristic information and historical interaction information corresponding to the historical delivering contents, and after the step of determining the resource delivering contents of the target resource screen according to the user characteristic information and the exposed resource information, the method further comprises:
acquiring the current interaction information of the user corresponding to the user characteristic information based on the resource delivery content;
and associating the user characteristic information, the resource delivery content and the current interaction information, and then adding the associated information to the tracking information base so as to update the tracking information base.
3. The resource delivery method of claim 1, wherein the step of determining the resource delivery content of the target resource screen based on the user characteristic information and the exposed resource information comprises:
and inputting the user characteristic information and the exposed resource information into a target resource release selection model obtained by pre-training to obtain the resource release content of the target resource screen.
4. The resource delivery method according to claim 3, wherein before the step of inputting the user characteristic information and the exposed resource information into a pre-trained target resource delivery selection model to obtain the resource delivery content of the target resource screen, the method further comprises:
acquiring an experience data set of a reinforcement learning framework based on the tracking information base;
and performing reinforcement learning training on the resource delivery selection model to be trained based on the empirical data set to obtain a target resource delivery selection model.
5. The method of resource placement according to claim 4, wherein the step of obtaining an empirical data set of a reinforcement learning framework based on the tracking information base comprises:
acquiring target user characteristic information and target exposed resource information corresponding to the target user characteristic information from the tracking information base as state data;
acquiring target resource release content determined based on the target user characteristic information and the target exposed resource information from the tracking information base as action data;
target interaction information of a user corresponding to the target user characteristic information based on the target resource delivery content is acquired from the tracking information base and is used as reward data;
and taking the state data, the action data and the reward data as a piece of experience data, and obtaining an experience data set based on a plurality of pieces of experience data.
6. The method for resource delivery according to claim 1, wherein after the step of determining the content of the resource delivery of the target resource screen according to the user characteristic information and the exposed resource information, the method further comprises:
detecting whether the tracking information to be deleted exists in the tracking information base according to a preset rule;
and if the tracking information to be deleted exists, deleting the tracking information to be deleted from the tracking information base so as to update the tracking information base.
7. The resource delivery method according to any one of claims 1 to 6, wherein the step of obtaining the user characteristic information obtained by user tracking of the target resource screen comprises:
acquiring tracking data acquired by user tracking equipment corresponding to the target resource screen;
and performing data analysis on the tracking data to obtain the user characteristic information.
8. A resource delivery apparatus, comprising:
the acquisition module is used for acquiring user characteristic information obtained by tracking a target resource screen by a user;
the matching module is used for matching and obtaining the exposed resource information corresponding to the user characteristic information from a preset tracking information base;
and the determining module is used for determining the resource releasing content of the target resource screen according to the user characteristic information and the exposed resource information.
9. A resource delivery device, characterized in that the resource delivery device comprises: memory, a processor and a resource placement program stored on the memory and executable on the processor, which when executed by the processor implements the steps of the resource placement method according to any of claims 1 to 7.
10. A computer-readable storage medium, having a resource placement program stored thereon, which when executed by a processor implements the steps of the resource placement method according to any one of claims 1 to 7.
CN202010246080.4A 2020-03-31 2020-03-31 Resource delivery method, device, equipment and storage medium Pending CN111445289A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010246080.4A CN111445289A (en) 2020-03-31 2020-03-31 Resource delivery method, device, equipment and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010246080.4A CN111445289A (en) 2020-03-31 2020-03-31 Resource delivery method, device, equipment and storage medium

Publications (1)

Publication Number Publication Date
CN111445289A true CN111445289A (en) 2020-07-24

Family

ID=71652731

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010246080.4A Pending CN111445289A (en) 2020-03-31 2020-03-31 Resource delivery method, device, equipment and storage medium

Country Status (1)

Country Link
CN (1) CN111445289A (en)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2012141683A (en) * 2010-12-28 2012-07-26 Yahoo Japan Corp Advertisement information providing device
CN107679899A (en) * 2017-09-26 2018-02-09 北京智云联众科技有限公司 The content put-on method and device of a kind of advertisement screen
CN109451038A (en) * 2018-12-06 2019-03-08 北京达佳互联信息技术有限公司 A kind of information-pushing method, device, server and computer readable storage medium
CN110413893A (en) * 2019-07-31 2019-11-05 腾讯科技(北京)有限公司 Object Push method, apparatus, computer equipment and storage medium

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2012141683A (en) * 2010-12-28 2012-07-26 Yahoo Japan Corp Advertisement information providing device
CN107679899A (en) * 2017-09-26 2018-02-09 北京智云联众科技有限公司 The content put-on method and device of a kind of advertisement screen
CN109451038A (en) * 2018-12-06 2019-03-08 北京达佳互联信息技术有限公司 A kind of information-pushing method, device, server and computer readable storage medium
CN110413893A (en) * 2019-07-31 2019-11-05 腾讯科技(北京)有限公司 Object Push method, apparatus, computer equipment and storage medium

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
李益群;张文生;杨柳;刘琰琼;: "基于标签的强化学习推荐算法研究与应用", 计算机应用研究, no. 08, pages 2845 - 28852 *

Similar Documents

Publication Publication Date Title
US20190079977A1 (en) Method for determining user behavior preference, and method and device for presenting recommendation information
US7921036B1 (en) Method and system for dynamically targeting content based on automatic demographics and behavior analysis
JP6267861B2 (en) Usage measurement techniques and systems for interactive advertising
KR101643915B1 (en) Construction system for big data reflect regional characteristics tailored ads
JP5224360B2 (en) Electronic advertising device, electronic advertising method and program
CN106296264A (en) A kind of pushing intelligent advertisements system based on recognition of face
CN108156209B (en) Media pushing method and system
TW201702956A (en) Smart billboards
WO2010053191A1 (en) Electronic advertisement system, electronic advertisement distribution apparatus, electronic advertisement distribution method, and recording medium
JP2003271084A (en) Apparatus and method for providing information
JP2012252613A (en) Customer behavior tracking type video distribution system
KR101607617B1 (en) System of providing real-time moving picture for tourist attraction
CN109048888B (en) Intelligent service method of robot
JP5408739B2 (en) Information processing apparatus, information processing method, information processing system, and information processing program
KR20130052883A (en) Advertisement of providing system and the method
KR102166044B1 (en) System for providing kids cafe management service using freepass membership
WO2019159044A1 (en) Method for placing a virtual advertising object for display to a user
US9930424B2 (en) Proxy channels for viewing audiences
JP6816091B2 (en) Analytical system
JP2008243095A (en) Face detection system and face detection method
CN111445289A (en) Resource delivery method, device, equipment and storage medium
CN112165514A (en) System and method for determining display content
CN111143698B (en) Content recommendation method and system
CN113297476A (en) Information processing method, display method and equipment
KR20200092630A (en) Method for providing cleaning academy service turning authenticated sanitary worker out using systematized and formalized education

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