CN116629924A - Method and device for optimizing multimedia resource delivery, computer equipment and storage medium - Google Patents

Method and device for optimizing multimedia resource delivery, computer equipment and storage medium Download PDF

Info

Publication number
CN116629924A
CN116629924A CN202210121804.1A CN202210121804A CN116629924A CN 116629924 A CN116629924 A CN 116629924A CN 202210121804 A CN202210121804 A CN 202210121804A CN 116629924 A CN116629924 A CN 116629924A
Authority
CN
China
Prior art keywords
candidate
region
attribute information
area
variables
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
CN202210121804.1A
Other languages
Chinese (zh)
Inventor
王茹
朱静涛
谢茵
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.)
SF Technology Co Ltd
Original Assignee
SF Technology 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 SF Technology Co Ltd filed Critical SF Technology Co Ltd
Priority to CN202210121804.1A priority Critical patent/CN116629924A/en
Publication of CN116629924A publication Critical patent/CN116629924A/en
Pending legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising
    • G06Q30/0201—Market modelling; Market analysis; Collecting market data
    • G06Q30/0203—Market surveys; Market polls
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising
    • G06Q30/0201—Market modelling; Market analysis; Collecting market data

Landscapes

  • Business, Economics & Management (AREA)
  • Strategic Management (AREA)
  • Engineering & Computer Science (AREA)
  • Accounting & Taxation (AREA)
  • Development Economics (AREA)
  • Finance (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Game Theory and Decision Science (AREA)
  • Data Mining & Analysis (AREA)
  • Economics (AREA)
  • Marketing (AREA)
  • Physics & Mathematics (AREA)
  • General Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The application provides a multimedia resource release optimizing method, a device, a computer device and a storage medium, wherein the method comprises the following steps: acquiring attribute information of each candidate region; constructing intervention variables and confusion variables of each candidate area according to the attribute information; analyzing intervention variables and confusion variables of each candidate region to obtain causal effect estimation values of each candidate region; and screening out a target area to carry out multimedia resource delivery according to the causal effect estimation value. By adopting the method, the delivery accuracy of the multimedia resources can be effectively improved.

Description

Method and device for optimizing multimedia resource delivery, computer equipment and storage medium
Technical Field
The embodiment of the application relates to the technical field of big data, in particular to a method and a device for optimizing multimedia resource delivery, computer equipment and a storage medium.
Background
With the development of mobile internet and big data, more and more enterprises or merchants choose to push or put marketing content to users through online or downward channels so as to improve the business achievement rate.
At present, most enterprises or merchants complete user description and positioning to form a user portrait by analyzing online behaviors of users, and then push or put appropriate content to the users according to the user portrait. Or alternatively, off-line random pushing or throwing is performed. However, online behavior analysis does not avoid the problem of single analysis dimension, but offline random delivery lacks perfect channel attribution, so that the actual effect difference of similar marketing contents of each channel is difficult to distinguish, and finally, the service achievement rate cannot be improved, and the actual demands of users cannot be met.
Therefore, the existing marketing content delivery method has the technical problem of low delivery accuracy.
Disclosure of Invention
The application aims to provide a method, a device, computer equipment and a storage medium for optimizing multimedia resource release, which are used for improving the release accuracy of multimedia resources.
In a first aspect, the present application provides a method for optimizing multimedia resource delivery, including:
acquiring attribute information of each candidate region;
constructing intervention variables and confusion variables of each candidate area according to the attribute information;
analyzing intervention variables and confusion variables of each candidate region to obtain causal effect estimation values of each candidate region;
and screening out a target area to carry out multimedia resource delivery according to the causal effect estimation value.
In some embodiments of the present application, the attribute information includes user attribute information and region attribute information, and constructing intervention variables and confusion variables for each candidate region according to the attribute information includes: analyzing the user attribute information to obtain brand competitiveness values of the candidate areas; constructing intervention variables of each candidate area according to the brand competitiveness value; and constructing confusion variables of each candidate area according to the area attribute information.
In some embodiments of the present application, analyzing user attribute information to obtain brand competitiveness values for candidate regions includes: analyzing the user attribute information to obtain the number of users belonging to the target brand in each candidate area, and obtaining a first user number; wherein the target brands belong to the target industry; aiming at the candidate area, acquiring the number of users belonging to the target industry to obtain a second number of users; and calculating the quotient of the first user number and the second user number to obtain the brand competitiveness value.
In some embodiments of the application, constructing intervention variables for each candidate region based on brand competitiveness values, comprises: sequencing each candidate region according to the brand competitiveness value to obtain a competitiveness value sequence; screening out a positive example area and a negative example area in each candidate area according to the competitive force value sequence; the positive and negative regions are constructed as intervention variables.
In some embodiments of the present application, the region attribute information includes AOI attribute information and crowd attribute information, and constructing confusion variables for each candidate region according to the region attribute information includes: acquiring region type information, region price information and region peripheral information of each candidate region according to the AOI attribute information, and taking the region type information, the region price information and the region peripheral information of each candidate region as AOI attribute variables; according to the crowd attribute information, counting crowd characteristic values, accumulated consumption frequency and accumulated consumption amount of each candidate area to be used as crowd characteristic variables; and constructing an AOI attribute variable and a crowd characteristic variable as confusion variables.
In some embodiments of the present application, analyzing the intervention variable and the confusion variable for each candidate region to obtain a causal effect estimate for each candidate region includes: inputting the intervention variable and the confusion variable of each candidate region into the trend score model, and outputting the trend score value of each candidate region; layering processing is carried out on each candidate region according to each tendency score value, and a target region layer comprising a positive example region and a negative example region is obtained through screening; the brand competitiveness value corresponding to the positive example area is a first competitiveness value, and the brand competitiveness value corresponding to the negative example area is a second competitiveness value; and aiming at the target area layer, calculating the difference between the first competitive value and the second competitive value to obtain the causal effect estimated value of each candidate area.
In some embodiments of the present application, screening out a target area for multimedia resource delivery according to a causal effect estimation value includes: if the causal effect estimated value is larger than a preset causal effect threshold value, determining a positive example area and a negative example area contained in the target area layer as the target area to carry out multimedia resource delivery.
In a second aspect, the present application provides a multimedia resource delivery optimizing apparatus, including:
The information acquisition module is used for acquiring attribute information of each candidate region;
the variable construction module is used for constructing intervention variables and confusion variables of each candidate area according to the attribute information;
the variable analysis module is used for analyzing the intervention variable and the confusion variable of each candidate area to obtain a causal effect estimated value of each candidate area;
and the resource release module is used for screening out a target area to release multimedia resources according to the causal effect estimation value.
In a third aspect, the present application also provides a computer device comprising:
one or more processors;
a memory; and one or more applications, wherein the one or more applications are stored in the memory and configured to be executed by the processor to implement the above-described multimedia asset delivery optimization method.
In a fourth aspect, the present application also provides a computer readable storage medium having stored thereon a computer program, the computer program being loaded by a processor for performing the steps of the method for optimizing delivery of multimedia resources.
In a fifth aspect, embodiments of the present application provide a computer program product or computer program comprising computer instructions stored in a computer readable storage medium. The processor of the computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device performs the method provided in the first aspect.
According to the method, the device, the computer equipment and the storage medium for optimizing the multimedia resource delivery, the server acquires the attribute information of each candidate region, constructs the intervention variable and the confusion variable of each candidate region according to the attribute information, analyzes the intervention variable and the confusion variable of each candidate region to obtain the causal effect estimated value of each candidate region, and finally screens out the target region according to the causal effect estimated value for multimedia resource delivery. Therefore, the effect result of each candidate area relative to the multimedia resource is analyzed on line in a mode of constructing multidimensional variable modeling, so that the delivery strategy of the multimedia resource is optimized timely, and the delivery accuracy of the multimedia resource can be effectively improved.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are needed in the description of the embodiments will be briefly described below, it being obvious that the drawings in the following description are only some embodiments of the present application, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a schematic view of a scenario of a method for optimizing multimedia resource delivery provided in an embodiment of the present application;
Fig. 2 is a flow chart of a method for optimizing multimedia resource delivery according to an embodiment of the present application;
fig. 3 is a schematic structural diagram of a multimedia resource delivery optimizing device provided in an embodiment of the present application;
fig. 4 is a schematic structural diagram of a computer device according to an embodiment of the present application.
Detailed Description
The following description of the embodiments of the present application will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present application, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.
In the description of the present application, the terms "first," "second," and the like are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defining "a first" or "a second" may explicitly or implicitly include one or more of the described features. In the description of the present application, the meaning of "a plurality" is two or more, unless explicitly defined otherwise.
In the description of the present application, the term "for example" is used to mean "serving as an example, instance, or illustration. Any embodiment described as "for example" in this disclosure is not necessarily to be construed as preferred or advantageous over other embodiments. The following description is presented to enable any person skilled in the art to make and use the application. In the following description, details are set forth for purposes of explanation. It will be apparent to one of ordinary skill in the art that the present application may be practiced without these specific details. In other instances, well-known structures and processes have not been described in detail so as not to obscure the description of the application with unnecessary detail. Thus, the present application is not intended to be limited to the embodiments shown, but is to be accorded the widest scope consistent with the principles and features disclosed herein.
In the embodiment of the application, the method for optimizing the multimedia resource delivery provided by the embodiment of the application can be applied to a multimedia resource delivery optimizing system shown in fig. 1. The multimedia resource delivery optimization system comprises a terminal 102 and a server 104. The terminal 102 may be a device that includes both receive and transmit hardware, i.e., a device having receive and transmit hardware capable of performing bi-directional communications over a bi-directional communication link. Such a device may include: a cellular or other communication device having a single-line display or a multi-line display or a cellular or other communication device without a multi-line display. The terminal 102 may be a desktop terminal or a mobile terminal, and the terminal 102 may be one of a mobile phone, a tablet computer, and a notebook computer. The server 104 may be a stand-alone server, or may be a server network or a server cluster of servers, including but not limited to a computer, a network host, a single network server, a set of multiple network servers, or a cloud server of multiple servers. Wherein the Cloud server is composed of a large number of computers or web servers based on Cloud Computing (Cloud Computing). In addition, the terminal 102 and the server 104 establish a communication connection through a network, and the network may specifically be any one of a wide area network, a local area network, and a metropolitan area network.
It will be appreciated by those skilled in the art that the application environment shown in fig. 1 is only one application scenario suitable for the present solution, and is not limited to the application scenario of the present solution, and other application environments may include more or fewer devices than those shown in fig. 1. For example, only 1 server is shown in fig. 1. It will be appreciated that the multimedia asset delivery optimization system may also include one or more other devices, and is not limited in this regard. In addition, the multimedia asset delivery optimization system may further comprise a memory for storing data, such as storing attribute information of each candidate region.
It should be noted that, the schematic view of the scenario of the multimedia resource delivery optimizing system shown in fig. 1 is only an example, and the multimedia resource delivery optimizing system and scenario described in the embodiments of the present application are for more clearly describing the technical solution of the embodiments of the present application, and do not constitute a limitation on the technical solution provided by the embodiments of the present application, and as a person of ordinary skill in the art can know that, with the evolution of the multimedia resource delivery optimizing system and the appearance of a new service scenario, the technical solution provided by the embodiments of the present application is equally applicable to similar technical problems.
Referring to fig. 2, an embodiment of the present application provides a method for optimizing multimedia resource delivery, and the embodiment is mainly exemplified by the method being applied to the server 104 in fig. 1, and the method includes steps S201 to S204, specifically as follows:
s201, obtaining attribute information of each candidate area.
The candidate area is AOI (Area of Interest) of the selected multimedia resource delivery strategy to be optimized, and the AOI is an information plane, which is also called an interest plane. "AOI" refers to a regional geographic entity in map data. For example, the candidate area "AOI" includes a certain cell, a certain business area, etc., and the type of the candidate area "AOI" according to the embodiments of the present application will be described in detail below.
Wherein the attribute information includes, but is not limited to, user attribute information and area attribute information, the user attribute information is derived from service data, and the area attribute information is derived from service data and geographic data. The service data is taken from a service scene, but it can be understood that different service scenes exist based on different application fields, and the embodiment of the application is mainly exemplified by a brand promotion scene, but is not limited to the brand promotion scene.
In a specific implementation, in a first step of starting a task of optimizing multimedia resource delivery, the server 104 needs to obtain attribute information of each candidate area, where the attribute information may be pre-stored in the server 104 locally or may be obtained through a request of other devices. In addition, the attribute information may be obtained from the acquisition, i.e. may have a subsequent processing requirement, or may be obtained after a certain preprocessing method, where the preprocessing method includes, but is not limited to: deduplication, debugging, format conversion, etc.
Further, if the attribute information is requested to be acquired through other devices, the attribute information includes, but is not limited to, one of the following ways: 1. in a common network architecture, the server 104 receives attribute information from the terminal 102 or other cloud devices with network connections established; 2. in a preset blockchain network, the server 104 can synchronously acquire attribute information from other terminal nodes or server nodes, and the blockchain network can be a public chain, a private chain and the like; 3. in the preset tree structure, the server 104 may request attribute information from a superior server or may poll attribute information from a subordinate server.
S202, constructing intervention variables and confusion variables of each candidate area according to the attribute information.
Wherein the intervention variable is a variable that exists in the independent variable to dependent variable causal link that causes a change in the dependent variable that itself changes with the independent variable, such variable being statistically related to both the independent variable and the dependent variable. Confusion variables are variables that are related to both independent and dependent variables that create false relationships between the independent and dependent variables, and are uncontrolled variables, which may also be referred to as additional variables.
In a specific implementation, since the multimedia resource delivery is performed based on the AOI selected from the target city as a unit, the purpose is to bring more consumers to the delivery party so as to improve the service achievement rate. However, the multimedia resources of the relevant products are usually positively influenced by the brand of the relevant products and other brands in the industry to which the relevant products belong, so that the optimization strategy provided by the embodiment of the application also needs to consider the effect brought to the industry.
Specifically, the embodiment of the present application proposes that the brand competitiveness value (Y) of each candidate area "AOI" is calculated first, then, in a selected range, a candidate area with a high brand competitiveness value (Y) is taken as a put area (hereinafter referred to as a positive example area), and other candidate areas with insufficient brand competitiveness value (Y) are taken as non-put areas (hereinafter referred to as negative example areas), finally, intervention variables and confusion variables are constructed according to the positive example area, the negative example area and other attribute information, and specific construction steps of the intervention variables and the confusion variables will be described in detail below.
In one embodiment, the attribute information includes user attribute information and region attribute information, and the step includes: analyzing the user attribute information to obtain brand competitiveness values of the candidate areas; constructing intervention variables of each candidate area according to the brand competitiveness value; and constructing confusion variables of each candidate area according to the area attribute information.
The user attribute information includes the number of relevant users of each brand and the number of relevant users of each industry, where the number of relevant users may refer to the number of registered users, or may refer to the number of receiving users or the number of sending users in the logistics field. It should be noted that, when the "related user number" refers to the number of the receiving users or the sending users in the logistics field, the objective of the multimedia resource delivery optimization is to improve the service achievement rate of a certain target brand by delivering the multimedia resource suitable for the logistics scene, especially for the target brand. For example, a mailing coupon for milk brand A, a mailing coupon for milk brand B, etc., and both milk brand A and milk brand B are attributed to the fresh industry.
In particular implementations, the attribute information obtained by the server 104 in the preceding step includes user attribute information and region attribute information that should be used to facilitate execution of subsequent steps. Specifically, the user attribute information including the number of users related to each brand and the number of users related to each industry can be used for analyzing and obtaining brand competitiveness values (Y) of candidate areas, and the brand competitiveness values (Y) can be used for further analyzing and screening out positive example areas and negative example areas to obtain intervention variables. The confusion variable is the region attribute information that needs to be analyzed for inclusion in the attribute information, as will be described in further detail below.
In one embodiment, analyzing the user attribute information to obtain brand competitiveness values for each candidate region includes: analyzing the user attribute information to obtain the number of users belonging to the target brand in each candidate area, and obtaining a first user number; wherein the target brands belong to the target industry; aiming at the candidate area, acquiring the number of users belonging to the target industry to obtain a second number of users; and calculating the quotient of the first user number and the second user number to obtain the brand competitiveness value.
In particular implementations, the first number of users may be expressed as "y 1 ", the second user number can be expressed as" y 2 ", thus, the brand competitiveness value is expressed as (y=y 1 /y 2 ). For example, in a candidate Area (AOI), server 104 obtains that 200 logistics tasks are generated in the AOI based on statistics of a preset period, that is, there are 200 number of receiving users and corresponding number of sending users (analysis based on the number of receiving users in this embodiment), wherein the number of receiving users related to a target brand is 80, the number of receiving users of other brands is 120, and the target brand and other brands belong to the same industry, and then the brand competitiveness value y=80/200=0.4 of the candidate area. It can be understood that only a certain candidate area is exemplified here, and in the actual application process, each brand existing in each candidate area should be analyzed one by one, and the preset time period can also be set according to the actual service requirement, and the length of the time period is not limited by the application.
In one embodiment, constructing intervention variables for each candidate region based on brand competitiveness values includes: sequencing each candidate region according to the brand competitiveness value to obtain a competitiveness value sequence; screening out a positive example area and a negative example area in each candidate area according to the competitive force value sequence; the positive and negative regions are constructed as intervention variables.
Wherein N is more than or equal to 1; the positive example area may be determined as a delivery area of the multimedia resource, and the negative example area may be determined as a non-delivery area of the multimedia resource.
In particular implementations, after server 104 analyzes brand competitiveness values (Y) for each candidate region (AOI) using the methods described in the embodiments above, each brand competitiveness value (Y) may be ordered according to a certain rule, such as ascending and descending. When the brand-competitive value sequence is obtained by ascending arrangement, the server 104 may obtain the last N brand-competitive values (Y) in the brand-competitive value sequence as target brand-competitive values; when the sequence of brand competitiveness values is obtained by descending order, server 104 may obtain the top N brand competitiveness values (Y) in the sequence of brand competitiveness values as target brand competitiveness values. Then, the candidate region corresponding to the target brand competitiveness value is taken as a positive example region, and the candidate region corresponding to the non-target brand competitiveness value is taken as a negative example region. And finally, taking the positive example area and the negative example area as intervention variables. It can be appreciated that the value of N can be determined depending on the actual business requirements.
In one embodiment, the region attribute information includes AOI attribute information and crowd attribute information, and constructing a confusion variable for each candidate region according to the region attribute information includes: acquiring region type information, region price information and region peripheral information of each candidate region according to the AOI attribute information, and taking the region type information, the region price information and the region peripheral information of each candidate region as AOI attribute variables; according to the crowd attribute information, counting crowd characteristic values, accumulated consumption frequency and accumulated consumption amount of each candidate area to be used as crowd characteristic variables; and constructing an AOI attribute variable and a crowd characteristic variable as confusion variables.
The AOI attribute information is a set of information such as region type information, region price information and region peripheral information; the crowd attribute information includes, but is not limited to, user portrait information such as gender, age, and wedding status of the user, and consumption content information of the user.
In a specific implementation, the region type information is generally divided into 10 classes, which are respectively: (1) ordinary houses, (2) high-end houses, (3) other houses, (4) office buildings, (5) business and living dual-purpose buildings, (6) industrial parks, (7) catering and shopping centers, (8) primary and secondary schools, (9) universities and adult education, and residential infrastructure (government banking hospitals and scientific and teaching places) and others. For analysis of the region type information, the server 104 needs to form a unique identification code for desensitization of each type, and perform MD5 code translation on the identification code, and then perform subsequent analysis by using the translated unique identification code.
Further, the server 104 may obtain the room price information of each candidate area by accessing the external interface of the real estate block, and if the room price information is not available, the room price information of the real estate block or the administrative area where the candidate area is located is used for complementing the room price median, and if the room price information is not available, the room price information is replaced by the city trading room price median. The regional peripheral information is mainly obtained by analyzing the positions of candidate regions, namely judging whether target brands of stores and other brands of stores in the industry exist in a specific commercial radiation range, and further respectively taking the target brands of stores and the other brands of stores as two dichotomous variables to enter a model, namely if the target brands of stores exist, the target brands of stores are marked as '1', and if the target brands of stores do not exist, the target brands of stores are marked as '0'. For example, healthcare products have a store radiation distance in the range of 2km, while household appliances have a store radiation distance in the range of 10 km. It can be understood that commercial radiation distances of various industries are preset, and the mode entering forms of the two classification variables comprise the following four types: [1, 1], [1, 0], [0, 1], [0, 0].
Further, the server 104 may count the specified crowd characteristics in each candidate region according to the crowd attribute information, and analyze and model the crowd characteristics corresponding to the crowd characteristics (also referred to as TGI values), including but not limited to: gender, age, wedding condition, etc. The TGI value of each candidate region "AOI" may be used as a crowd feature variable, and the TGI value may be calculated as follows:
For example, a certain characteristic population (sex is male, age 20 years and above, married) in a certain candidate region "AOI" occupies 250 people, and the city in which the candidate region "AOI" is located is "city a", whereas the same characteristic population (sex is male, age 20 years and above, married) in "city a" occupies 800 people, and the TGI value of the candidate region "AOI" is "0.3125=250/800".
Still further, the first part of the crowd characteristic variable is the analyzed crowd characteristic value, and the rest part of the crowd characteristic variable also comprises the accumulated consumption frequency and the accumulated consumption amount. The accumulated consumption frequency and the accumulated consumption amount are derived from statistics of consumption contents of users, the consumption contents are divided into consumption frequency and consumption amount related to the put-in product, and then the accumulated consumption frequency and the accumulated consumption amount are the sum of consumption frequency and consumption amount of a certain characteristic group in a certain candidate area 'AOI' in a preset period. It should be noted that, the counted delivered item should be the smallest parent item unit of the delivered item, and the length of the preset period depends on the service requirement.
For example, the smallest parent of the lipstick is lip make-up, and the consumption frequency can be judged by counting the purchase quantity of the lip make-up, so that the consumption preference of the user is reflected. The amount of consumption is determined by the amount of the order of the "lip make-up" category, reflecting the purchasing power of the user.
It should be noted that, the "modeling" refers to a trend score model, and in the embodiment of the present application, the trend score model selects a logic model, and variables finally input to the trend score model (logic model) are intervention variables (positive example area and negative example area) and confusion variables (AOI attribute variables and crowd feature variables).
S203, analyzing the intervention variable and the confusion variable of each candidate area to obtain the causal effect estimated value of each candidate area.
In particular implementations, after the server 104 analyzes the intervention variables and the confounding variables, the intervention variables and the confounding variables may be input into the trend score model, so that the model first ranks the candidate regions based on the intervention variables and the confounding variables of each candidate region, and then calculates a causal effect estimate (ATE) based on the positive and negative example regions included in each hierarchy. The trend score model may be a logic model, where the logic model (also referred to as a "assessment model", "classification assessment model", and Logistic regression "logistic regression") is one of the discrete selection method models, and the logic model is the earliest discrete selection model and is the most widely used model at present.
In one embodiment, the step includes: inputting the intervention variable and the confusion variable of each candidate region into the trend score model, and outputting the trend score value of each candidate region; layering processing is carried out on each candidate region according to each tendency score value, and a target region layer comprising a positive example region and a negative example region is obtained through screening; the brand competitiveness value corresponding to the positive example area is a first competitiveness value, and the brand competitiveness value corresponding to the negative example area is a second competitiveness value; and aiming at the target area layer, calculating the difference between the first competitive value and the second competitive value to obtain the causal effect estimated value of each candidate area.
The tendency score model is applied with a tendency score layering algorithm which is an effective means for controlling influences of confounding factors, and the tendency score layering algorithm is combined with the Logit model, so that the confounding bias can be controlled more effectively, and the putting accuracy of multimedia resources is improved.
In particular implementations, after the server 104 inputs the intervention variable and the confounding variable for each candidate region into the trend score model, the intervention variable will serve as a dependent variable for the model, the confounding variable will serve as an independent variable for the model, the trend score model will analyze each dependent variable and independent variable to output a trend score value for each candidate region, and the trend score value is in the range of 0 to 1. Then, the server 104 may divide each candidate region into n layers according to the trend score value, and determine the actual value of n by using the degree of balance of the trend score value in each layer, and finally discard the region containing only the positive example region or the negative example regionInstead, a target area layer containing both the positive and negative areas is screened out, and then the causal effect estimation value (ATE) of the target area layer is calculated, wherein the ATE=y 3 -y 4 ”，“y 3 "is brand competitive value corresponding to the positive example area in the target area layer," y 4 "is the brand competitiveness value corresponding to the counterexample area in the target area layer.
It should be noted that, at this time, not all candidate areas may correspond to the causal effect estimation "ATE", and the causal effect estimation "ATE" is only specific to the positive example area and the negative example area in the target area layer, that is, the steps described in this embodiment are preliminary screening of the candidate areas.
S204, screening out a target area to carry out multimedia resource delivery according to the causal effect estimation.
In a specific implementation, after the server 104 obtains the causal effect estimation value "ATE" of the target area layer, each causal effect estimation value "ATE" may be compared, and then the final target area is screened out according to the comparison result to perform multimedia resource delivery, and the other screened candidate areas do not need to perform multimedia resource delivery. It should be noted that, this step is aimed at discarding the region with low estimation value or negative effect to perform selective delivery, and at the same time, the intervention variable and the confusion variable can be updated at regular time according to the model result to perform dynamic adjustment by re-modeling, so as to stop delivery when the estimated causal effect of the model evaluation is negative, and obtain the maximized effect by minimizing the resource.
In one embodiment, the step includes: if the causal effect estimated value is larger than a preset causal effect threshold value, determining a positive example area and a negative example area contained in the target area layer as the target area to carry out multimedia resource delivery.
Wherein the causal effect threshold may be a value according to actual traffic requirements, e.g. the causal effect threshold is denoted "ATEn", "ATEn ≡0".
In a specific implementation, after all the causal effect estimates "ATE" are obtained by the server 104, each causal effect estimate "ATE" may be compared with a causal effect threshold "ATEn", where "ATE" greater than the causal effect threshold "ATEn" is selected as a target causal effect estimate, and then a target area layer corresponding to the target causal effect estimate is determined as a final desired result, that is, both the positive example area and the negative example area to which the final desired result points are regarded as target areas.
For example, the server 104 performs layering processing on the candidate region according to the tendency score value, to obtain three region layers: A. b, C where the region layer "A" has only one positive region (candidate region whose brand competitiveness value Y satisfies the condition), the region layer "B" has only one negative region (candidate region whose brand competitiveness value Y does not satisfy the condition), and the region layer "C" contains one positive region (Y) 3 =0.7) and a counterexample area (y 4 =0.5), then zone layer "C" is the target zone layer, and the causal effect estimate "ate=0.7-0.5=0.2" for the target zone layer "C". At this time, because of the causal effect threshold "aten=0", the positive example area and the negative example area in the target area layer "C" are both target areas, and the multimedia resources can be continuously put in.
According to the multimedia resource delivery optimization method in the embodiment, the server acquires the attribute information of each candidate region, constructs the intervention variable and the confusion variable of each candidate region according to the attribute information, analyzes the intervention variable and the confusion variable of each candidate region to obtain the causal effect estimated value of each candidate region, and finally screens out the target region according to the causal effect estimated value to carry out multimedia resource delivery. Therefore, the effect result of each candidate area relative to the multimedia resource is analyzed on line in a mode of constructing multidimensional variable modeling, so that the delivery strategy of the multimedia resource is optimized timely, and the delivery accuracy of the multimedia resource can be effectively improved.
It should be understood that, although the steps in the flowchart of fig. 2 are shown in sequence as indicated by the arrows, the steps are not necessarily performed in sequence as indicated by the arrows. The steps are not strictly limited to the order of execution unless explicitly recited herein, and the steps may be executed in other orders. Moreover, at least some of the steps in fig. 2 may include multiple sub-steps or stages that are not necessarily performed at the same time, but may be performed at different times, nor do the order in which the sub-steps or stages are performed necessarily performed in sequence, but may be performed alternately or alternately with at least a portion of the sub-steps or stages of other steps or other steps.
In order to better implement the method for optimizing multimedia resource delivery provided by the embodiment of the present application, on the basis of the method for optimizing multimedia resource delivery provided by the embodiment of the present application, the embodiment of the present application further provides a device for optimizing multimedia resource delivery, as shown in fig. 3, where the device 300 for optimizing multimedia resource delivery includes:
an information obtaining module 310, configured to obtain attribute information of each candidate region;
a variable construction module 320, configured to construct intervention variables and confusion variables of each candidate region according to the attribute information;
the variable analysis module 330 is configured to analyze the intervention variable and the confusion variable of each candidate region to obtain a causal effect estimation value of each candidate region;
and the resource release module 340 is configured to screen out the target area for multimedia resource release according to the causal effect estimation.
In one embodiment, the attribute information includes user attribute information and region attribute information, and the variable construction module 320 is further configured to analyze the user attribute information to obtain brand competitiveness values of each candidate region; constructing intervention variables of each candidate area according to the brand competitiveness value; and constructing confusion variables of each candidate area according to the area attribute information.
In one embodiment, the variable construction module 320 is further configured to analyze the user attribute information to obtain a number of users belonging to the target brand in each candidate region, so as to obtain a first number of users; wherein the target brands belong to the target industry; aiming at the candidate area, acquiring the number of users belonging to the target industry to obtain a second number of users; and calculating the quotient of the first user number and the second user number to obtain the brand competitiveness value.
In one embodiment, the variable construction module 320 is further configured to sort the candidate regions according to the brand competitiveness values, to obtain a sequence of competitiveness values; screening out a positive example area and a negative example area in each candidate area according to the competitive force value sequence; the positive and negative regions are constructed as intervention variables.
In one embodiment, the region attribute information includes AOI attribute information and crowd attribute information, and the variable construction module 320 is further configured to obtain, as AOI attribute variables, region type information, region room price information, and region peripheral information of each candidate region according to the AOI attribute information; according to the crowd attribute information, counting crowd characteristic values, accumulated consumption frequency and accumulated consumption amount of each candidate area to be used as crowd characteristic variables; and constructing an AOI attribute variable and a crowd characteristic variable as confusion variables.
In one embodiment, the variable analysis module 330 is further configured to input the intervention variable and the confusion variable of each candidate region into the trend score model, and output the trend score value of each candidate region; layering processing is carried out on each candidate region according to each tendency score value, and a target region layer comprising a positive example region and a negative example region is obtained through screening; the brand competitiveness value corresponding to the positive example area is a first competitiveness value, and the brand competitiveness value corresponding to the negative example area is a second competitiveness value; and aiming at the target area layer, calculating the difference between the first competitive value and the second competitive value to obtain the causal effect estimated value of each candidate area.
In one embodiment, the resource delivery module 340 is further configured to determine the positive example area and the negative example area included in the target area layer as the target area for delivering the multimedia resource if the causal effect estimated value is greater than the preset causal effect threshold.
In the above embodiment, the method of constructing the multidimensional variable modeling is provided in the present application, and the effect result of each candidate area relative to the multimedia resource is analyzed on line, so as to optimize the delivery strategy of the multimedia resource in good time, and effectively improve the delivery accuracy of the multimedia resource.
It should be noted that, the specific limitation of the multimedia resource delivery optimizing device may be referred to the limitation of the multimedia resource delivery optimizing method, and will not be described herein. The modules in the multimedia resource delivery optimizing device can be realized by all or part of software, hardware and combination thereof. The above modules may be embedded in hardware or independent of a processor in the electronic device, or may be stored in software in a memory in the electronic device, so that the processor may call and execute operations corresponding to the above modules.
In some embodiments of the present application, the multimedia asset delivery optimization device 300 may be implemented in the form of a computer program that may be run on a computer device as shown in fig. 4. The memory of the computer device may store various program modules constituting the multimedia resource delivery optimizing apparatus 300, such as the information acquisition module 310, the variable construction module 320, the variable analysis module 330, and the resource delivery module 340 shown in fig. 3; the computer program comprising the respective program modules causes the processor to execute the steps in the method for optimizing multimedia asset delivery according to the respective embodiments of the present application described in the present specification. For example, the computer device shown in fig. 4 may perform step S201 through the information acquisition module 310 in the multimedia asset delivery optimization device 300 shown in fig. 3. The computer device may perform step S202 through the variable construction module 320. The computer device may perform step S203 through the variable analysis module 330. The computer device may perform step S204 through the resource provisioning module 340. The computer device includes a processor, a memory, and a network interface coupled by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The network interface of the computer device is used for communicating with an external computer device through a network connection. The computer program, when executed by a processor, implements a method of optimizing delivery of multimedia resources.
It will be appreciated by persons skilled in the art that the architecture shown in fig. 4 is merely a block diagram of some of the architecture relevant to the present inventive arrangements and is not limiting as to the computer device to which the present inventive arrangements are applicable, and that a particular computer device may include more or fewer components than shown, or may combine some of the components, or have a different arrangement of components.
In some embodiments of the application, a computer device is provided that includes one or more processors; a memory; and one or more applications, wherein the one or more applications are stored in the memory and configured to be executed by the processor to perform the steps of the above-described multimedia asset delivery optimization method. The steps of the method for optimizing multimedia resources may be the steps of the method for optimizing multimedia resources in the foregoing embodiments.
In some embodiments of the present application, a computer readable storage medium is provided, in which a computer program is stored, where the computer program is loaded by a processor, so that the processor performs the steps of the above-mentioned multimedia resource delivery optimization method. The steps of the method for optimizing multimedia resources may be the steps of the method for optimizing multimedia resources in the foregoing embodiments.
Those of ordinary skill in the art will appreciate that implementing all or part of the above-described methods may be accomplished by way of a computer program stored on a non-transitory computer readable storage medium, which when executed, may comprise the steps of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in embodiments provided herein can include at least one of non-volatile and volatile memory. The nonvolatile Memory may include Read-Only Memory (ROM), magnetic tape, floppy disk, flash Memory, optical Memory, or the like. Volatile memory can include random access memory (Random Access Memory, RAM) or external cache memory. By way of illustration, and not limitation, RAM can take many forms, such as static random access memory (Static Random Access Memory, SRAM) or dynamic random access memory (Dynamic Random Access Memory, DRAM), among others.
The technical features of the above embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.
The foregoing describes in detail a method, apparatus, computer device and storage medium for optimizing multimedia resource delivery, and specific examples are applied to illustrate the principles and embodiments of the present application, and the description of the foregoing examples is only used to help understand the method and core idea of the present application; meanwhile, as those skilled in the art will have variations in the specific embodiments and application scope in light of the ideas of the present application, the present description should not be construed as limiting the present application.

Claims (10)

1. The method for optimizing the multimedia resource delivery is characterized by comprising the following steps of:
acquiring attribute information of each candidate region;
constructing intervention variables and confusion variables of each candidate area according to the attribute information;
analyzing intervention variables and confusion variables of each candidate region to obtain causal effect estimation values of each candidate region;
and screening out a target area to carry out multimedia resource delivery according to the causal effect estimated value.
2. The method of claim 1, wherein the attribute information includes user attribute information and region attribute information, and the constructing the intervention variable and the confusion variable for each of the candidate regions based on the attribute information comprises:
Analyzing the user attribute information to obtain brand competitiveness values of the candidate areas;
constructing intervention variables of the candidate areas according to the brand competitiveness value;
and constructing confusion variables of the candidate areas according to the area attribute information.
3. The method of claim 2, wherein said analyzing said user attribute information to obtain brand competitiveness values for each of said candidate areas comprises:
analyzing the user attribute information to obtain the number of users belonging to the target brand in each candidate area, and obtaining a first user number; wherein the target brand is attributed to a target industry;
aiming at the candidate area, acquiring the number of users belonging to the target industry to obtain a second number of users;
and calculating the quotient of the first user number and the second user number to obtain the brand competitiveness value.
4. The method of claim 2, wherein said constructing an intervention variable for each of said candidate regions based on said brand competitiveness value comprises:
sequencing the candidate areas according to the brand competitiveness value to obtain a competitiveness value sequence;
screening out a positive example area and a negative example area in each candidate area according to the competitive value sequence;
And constructing the positive example area and the negative example area as the intervention variables.
5. The method of claim 2, wherein the region attribute information includes AOI attribute information and crowd attribute information, and the constructing a confusion variable for each of the candidate regions based on the region attribute information comprises:
acquiring region type information, region price information and region peripheral information of each candidate region according to the AOI attribute information, and taking the region type information, the region price information and the region peripheral information of each candidate region as AOI attribute variables;
according to the crowd attribute information, counting crowd characteristic values, accumulated consumption frequency and accumulated consumption amount of each candidate area to be used as crowd characteristic variables;
and constructing the AOI attribute variable and the crowd characteristic variable as the confusion variable.
6. The method of any one of claims 1-5, wherein said analyzing intervention variables and confounding variables for each of said candidate regions to obtain causal effect estimates for each of said candidate regions comprises:
inputting the intervention variable and the confusion variable of each candidate region into a trend score model, and outputting the trend score value of each candidate region;
layering processing is carried out on each candidate region according to each tendency score value, and a target region layer comprising a positive example region and a negative example region is obtained through screening; the brand competitiveness value corresponding to the positive example area is a first competitiveness value, and the brand competitiveness value corresponding to the negative example area is a second competitiveness value;
And calculating the difference between the first competitive value and the second competitive value aiming at the target area layer to obtain causal effect estimated values of the candidate areas.
7. The method of claim 6, wherein screening the target area for multimedia resource delivery based on the causal effect estimate comprises:
and if the causal effect estimated value is larger than a preset causal effect threshold value, determining a positive example area and a negative example area contained in the target area layer as the target area to carry out multimedia resource delivery.
8. A multimedia resource delivery optimizing apparatus, comprising:
the information acquisition module is used for acquiring attribute information of each candidate region;
the variable construction module is used for constructing intervention variables and confusion variables of the candidate areas according to the attribute information;
the variable analysis module is used for analyzing the intervention variable and the confusion variable of each candidate region to obtain a causal effect estimated value of each candidate region;
and the resource release module is used for screening out a target area to release multimedia resources according to the causal effect estimation value.
9. A computer device, the computer device comprising:
One or more processors;
a memory; and
one or more applications, wherein the one or more applications are stored in the memory and configured to be executed by the processor to implement the multimedia asset delivery optimization method of any of claims 1 to 7.
10. A computer readable storage medium, having stored thereon a computer program, the computer program being loaded by a processor to perform the steps of the method of optimizing delivery of a multimedia resource as claimed in any of claims 1 to 7.
CN202210121804.1A 2022-02-09 2022-02-09 Method and device for optimizing multimedia resource delivery, computer equipment and storage medium Pending CN116629924A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210121804.1A CN116629924A (en) 2022-02-09 2022-02-09 Method and device for optimizing multimedia resource delivery, computer equipment and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210121804.1A CN116629924A (en) 2022-02-09 2022-02-09 Method and device for optimizing multimedia resource delivery, computer equipment and storage medium

Publications (1)

Publication Number Publication Date
CN116629924A true CN116629924A (en) 2023-08-22

Family

ID=87636962

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210121804.1A Pending CN116629924A (en) 2022-02-09 2022-02-09 Method and device for optimizing multimedia resource delivery, computer equipment and storage medium

Country Status (1)

Country Link
CN (1) CN116629924A (en)

Similar Documents

Publication Publication Date Title
US9152969B2 (en) Recommendation ranking system with distrust
US10116758B2 (en) Delivering notifications based on prediction of user activity
Cao et al. Customer demand prediction of service-oriented manufacturing incorporating customer satisfaction
Javari et al. Recommender systems based on collaborative filtering and resource allocation
US20140188994A1 (en) Social Neighborhood Determination
CN109993627B (en) Recommendation method, recommendation model training device and storage medium
Kim et al. The accuracy of tourism forecasting and data characteristics: a meta-analytical approach
CN111429161B (en) Feature extraction method, feature extraction device, storage medium and electronic equipment
US10313461B2 (en) Adjusting pacing of notifications based on interactions with previous notifications
US20170236226A1 (en) Computerized systems, processes, and user interfaces for globalized score for a set of real-estate assets
US20130325861A1 (en) Data Clustering for Multi-Layer Social Link Analysis
CN112989169A (en) Target object identification method, information recommendation method, device, equipment and medium
CN113781139A (en) Item recommendation method, item recommendation device, equipment and medium
CN111475744B (en) Personalized position recommendation method based on ensemble learning
US20200286022A1 (en) Platform for In-Memory Analysis of Network Data Applied to Site Selection with Current Market Information, Demand Estimates, and Competitor Information
Mulders et al. Inference of node attributes from social network assortativity
Vicari et al. Multivariate linear regression for heterogeneous data
Shafik et al. Recommendation system comparative analysis: internet of things aided networks
CN110720099A (en) System and method for providing recommendation based on seed supervised learning
CN111340522A (en) Resource recommendation method, device, server and storage medium
Pachamanova et al. Uncertainty representation and risk management for direct segmented marketing
CN112150184A (en) Click rate estimation method and system, computer system and computer readable medium
Martínez-Torres et al. Identifying the features of reputable users in eWOM communities by using Particle Swarm Optimization
CN112100511A (en) Preference degree data obtaining method and device and electronic equipment
CN113760521A (en) Virtual resource allocation 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