CN111401593A - Value combination determination method and device, readable storage medium and electronic equipment - Google Patents
Value combination determination method and device, readable storage medium and electronic equipment Download PDFInfo
- Publication number
- CN111401593A CN111401593A CN201811644344.0A CN201811644344A CN111401593A CN 111401593 A CN111401593 A CN 111401593A CN 201811644344 A CN201811644344 A CN 201811644344A CN 111401593 A CN111401593 A CN 111401593A
- Authority
- CN
- China
- Prior art keywords
- value
- target
- combination
- display information
- proposed
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 36
- 238000012549 training Methods 0.000 claims description 42
- 238000004590 computer program Methods 0.000 claims description 9
- 238000012545 processing Methods 0.000 claims description 3
- 230000008901 benefit Effects 0.000 abstract description 4
- 238000013528 artificial neural network Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 5
- 238000004891 communication Methods 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 2
- 230000002860 competitive effect Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000013480 data collection Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- 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
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
-
- 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
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/01—Social networking
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Economics (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Tourism & Hospitality (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Software Systems (AREA)
- Operations Research (AREA)
- Game Theory and Decision Science (AREA)
- Development Economics (AREA)
- Computing Systems (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Quality & Reliability (AREA)
- Artificial Intelligence (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- User Interface Of Digital Computer (AREA)
Abstract
The disclosure relates to a value combination determination method, a value combination determination device, a readable storage medium and an electronic device. The method comprises the following steps: generating a plurality of first value combinations corresponding to target display information, wherein the value combinations corresponding to the display information comprise the proposed value of the display information and the expected value of the display information; inputting the first value combination into a target prediction model to obtain a prediction return value corresponding to the first value combination; and determining the drawn value in the first value combination with the maximum predicted return value as the target drawn value of the target display information, and determining the expected value in the first value combination with the maximum predicted return value as the target expected value of the target display information. Therefore, the labor can be saved, the efficiency can be improved, and meanwhile, the benefit of the user can be maximized.
Description
Technical Field
The present disclosure relates to the field of computer technologies, and in particular, to a value combination determination method and apparatus, a readable storage medium, and an electronic device.
Background
Information flows (Feeds) are one of the most important innovation points in social media mobile applications, namely presenting a user with a series of information when the user loads a display interface. The series of information is a series of display information, each piece of display information has a set value, and corresponding values can be obtained by displaying the display information, so that the display information is usually displayed according to the set value, namely the higher the set value is, the higher the display priority is. In practice, different display information providers give a proposed value and a desired value for the display information, and the desired value of the display information can be regarded as the total value of the display information required to be spent in one display period under the proposed value. The display information will get a reward value after being actually displayed, i.e. the reward obtained during the display, so the final goal of the display information display is to get a high reward value. In the prior art, the determination of the proposed value and the expected value requires manual data collection and then manual analysis and determination, so that on one hand, a large amount of manpower, material resources, time and the like are required to be invested, the efficiency is low, on the other hand, the determined proposed value and the determined expected value cannot meet the requirements constantly due to the fact that the manual analysis mode is limited by factors such as experience, environment and the like, and the accuracy is insufficient.
Disclosure of Invention
The invention aims to provide a value combination determination method, a value combination determination device, a readable storage medium and electronic equipment, so as to automatically generate a value combination.
In order to achieve the above object, according to a first aspect of the present disclosure, there is provided a value combination determination method including:
generating a plurality of first value combinations corresponding to target display information, wherein the value combinations corresponding to the display information comprise the proposed value of the display information and the expected value of the display information;
inputting the first value combination into a target prediction model to obtain a prediction return value corresponding to the first value combination;
and determining the drawn value in the first value combination with the maximum predicted return value as the target drawn value of the target display information, and determining the expected value in the first value combination with the maximum predicted return value as the target expected value of the target display information.
Optionally, the target prediction model is obtained by:
acquiring a training sample corresponding to the target display information, wherein the training sample at least comprises a second value combination corresponding to the target display information and an actual return value corresponding to the second value combination, and the second value combination is a value combination corresponding to the target display information in historical actual display;
and training a first prediction model according to the training samples to obtain the target prediction model.
Optionally, the training sample further includes a third value combination corresponding to historical display information and an actual return value corresponding to the third value combination, where the historical display information is display information having a corresponding relationship with the target display information.
Optionally, the generating a plurality of first value combinations corresponding to the target display information includes:
acquiring an expected value range and a proposed value range corresponding to the target display information;
generating the plurality of first value combinations according to the expected value range and the proposed value range.
Optionally, the determining the proposed value in the first value combination with the largest predicted return value as the target proposed value of the target exhibition information and the expected value in the first value combination with the largest predicted return value as the target expected value of the target exhibition information includes:
if the first value combination with the largest predicted return value is multiple, determining the first value combination with the lowest expected value as a target first value combination;
determining a proposed value in the target first value combination as a target proposed value of the target presentation information, and determining an expected value in the target first value combination as the target expected value of the target presentation information.
According to a second aspect of the present disclosure, there is provided a value combination determination apparatus, the apparatus comprising:
the generating module is used for generating a plurality of first value combinations corresponding to the target display information, wherein the value combinations corresponding to the display information comprise the proposed value of the display information and the expected value of the display information;
the processing module is used for inputting the first value combination into a target prediction model to obtain a prediction return value corresponding to the first value combination;
and the determining module is used for determining the proposed value in the first value combination with the maximum predicted return value as the target proposed value of the target display information, and determining the expected value in the first value combination with the maximum predicted return value as the target expected value of the target display information.
Optionally, the target prediction model is obtained by:
acquiring a training sample corresponding to the target display information, wherein the training sample at least comprises a second value combination corresponding to the target display information and an actual return value corresponding to the second value combination, and the second value combination is a value combination corresponding to the target display information in historical actual display;
and training a first prediction model according to the training samples to obtain the target prediction model.
Optionally, the training sample further includes a third value combination corresponding to historical display information and an actual return value corresponding to the third value combination, where the historical display information is display information having a corresponding relationship with the target display information.
Optionally, the generating module includes:
the acquisition submodule is used for acquiring an expected value range and a proposed value range corresponding to the target display information;
a generating sub-module for generating the plurality of first value combinations according to the expected value range and the proposed value range.
Optionally, the determining module includes:
a first determining sub-module, configured to determine, if the first value combination with the largest predicted return value is multiple, the first value combination with the lowest expected value as a target first value combination;
a second determining sub-module, configured to determine the proposed value in the target first value combination as a target proposed value of the target presentation information, and determine an expected value in the target first value combination as the target expected value of the target presentation information.
According to a third aspect of the present disclosure, there is provided a computer readable storage medium having stored thereon a computer program which, when executed by a processor, performs the steps of the method of the first aspect of the present disclosure.
According to a fourth aspect of the present disclosure, there is provided an electronic device comprising:
a memory having a computer program stored thereon;
a processor for executing the computer program in the memory to implement the steps of the method of the first aspect of the disclosure.
According to the technical scheme, aiming at the target display information, a plurality of first value combinations corresponding to the target display information are firstly generated, then the first value combinations are input into a target prediction model to obtain prediction return values corresponding to the first value combinations, then the proposed value in the first value combination with the largest prediction return value is determined as the target proposed value of the target display information from the plurality of prediction return values, and the expected value in the first value combination with the largest prediction return value is determined as the target expected value of the target display information. In this way, the return value of the target display information is predicted through the target prediction model, and the drawn value and the expected value corresponding to the combination of the first value with the maximum predicted return value are determined as the target drawn value and the target expected value corresponding to the target display information, so that the target drawn value and the target expected value can be automatically generated aiming at the target display information, the step of manually determining the drawn value and the expected value is omitted, the labor is saved, the efficiency is improved, meanwhile, the generated target drawn value and the target expected value can obtain the optimal return value as much as possible, and the benefit of a user is maximized.
Additional features and advantages of the disclosure will be set forth in the detailed description which follows.
Drawings
The accompanying drawings, which are included to provide a further understanding of the disclosure and are incorporated in and constitute a part of this specification, illustrate embodiments of the disclosure and together with the description serve to explain the disclosure without limiting the disclosure. In the drawings:
FIG. 1 is a flow chart of a value combination determination method provided according to one embodiment of the present disclosure;
FIG. 2 is a flow diagram of one exemplary implementation of the step of generating a plurality of first value combinations corresponding to target presentation information in a value combination determination method provided in accordance with the present disclosure;
FIG. 3 is a block diagram of a value combination determination device provided in accordance with one embodiment of the present disclosure;
FIG. 4 is a block diagram illustrating an electronic device in accordance with an example embodiment.
Detailed Description
The following detailed description of specific embodiments of the present disclosure is provided in connection with the accompanying drawings. It should be understood that the detailed description and specific examples, while indicating the present disclosure, are given by way of illustration and explanation only, not limitation.
It is noted that the terms "first," "second," "third," and the like in the description and claims of the present disclosure and in the above-described drawings are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order.
Fig. 1 is a flowchart of a value combination determination method provided according to an embodiment of the present disclosure. As shown in fig. 1, the method may include the following steps.
In step 11, a plurality of first value combinations corresponding to the target presentation information are generated.
Illustratively, the presentation information may be, for example, text, pictures, video, and the like. The value combination corresponding to the presentation information may include a proposed value of the presentation information and a desired value of the presentation information. The proposed value of the presentation information may be considered as the value that it takes to click or view the presentation information once after being presented at the booth, which may be understood as a virtual or actual carrier for presenting the presentation information, and the expected value of the presentation information may be considered as the total value that it takes to click or view the presentation information for the expected number of times (i.e., one presentation period) at the proposed value. When an idle exhibition position exists, the user can give a proposed value and a budget value, and the currently highest proposed value has the highest display priority at present according to the proposed values and the budget values given by the users.
For the target presentation information, a plurality of first value combinations corresponding to the target presentation information may be generated, each first value combination including a proposed value and a desired value. Alternatively, as can be seen from the above-mentioned concepts of the proposed value and the expected value, in general, the expected value in the same value combination is greater than or equal to the proposed value, and therefore, this factor may be considered when generating the first value combination, that is, in generating a plurality of first value combinations corresponding to the target exhibition information, a condition that the expected value is greater than or equal to the proposed value in the same first value combination may be set, for example, so that the generated first value combinations are all usable, improving the generation efficiency of the first value combination.
In one possible embodiment, the plurality of first value combinations may be generated directly for the target presentation information, i.e. the first value combinations are generated randomly. Illustratively, the proposed value and the expected value may be randomly chosen from positive real numbers.
In another possible implementation, as shown in fig. 2, the generating of the plurality of first value combinations corresponding to the target exhibition information in step 11 may include the following steps.
In step 21, an expected value range and a proposed value range corresponding to the target display information are obtained.
For example, the provider of the presentation information may estimate the value of various presentation information in advance for different presentation information, that is, may set respective expected value ranges and drawn value ranges for different presentation information in advance. The expected value range and the drawn value range can be set artificially according to the content of the display information, the historical drawn value of the display information and the historical expected value. Therefore, for the target exhibition information, the expected value range and the proposed value range corresponding to the target exhibition information can be acquired.
In step 22, a plurality of first value combinations are generated based on the desired value range and the proposed value range.
A plurality of first value combinations can be generated according to the expected value range and the proposed value range corresponding to the target display information. For example, a desired value may be randomly selected from a range of desired values, a proposed value may be randomly selected from a range of proposed values, a first value combination composed of the desired value and the proposed value may be generated, and a plurality of first value combinations may be generated as described above until the number of first value combinations satisfies a predetermined requirement. As another example, selectable desired values and selectable proposed values within the desired value range and proposed value range may be traversed against the desired value range and proposed value range to obtain a plurality of first value combinations. Alternatively, the generated plurality of combinations of first values may be mutually different.
Through the mode, the plurality of first value combinations are generated according to the expected value range and the proposed value range corresponding to the target display information, so that the generated first value combinations are more targeted, more suitable for the target display information and strong in practicability.
Returning now to FIG. 1, in step 12, the first value combination is input into the target predictive model to obtain a corresponding return-to-forecast value for the first value combination.
The target prediction model may generate and output a predicted return value corresponding to the first value combination from the input first value combination. The manner in which the target prediction model is obtained will first be illustrated below.
Illustratively, the target prediction model may be obtained by:
acquiring a training sample corresponding to the target display information;
and training the first prediction model according to the training samples to obtain a target prediction model.
In one possible embodiment, the training sample may include the second value combination corresponding to the target presentation information and the actual return value corresponding to the second value combination. Wherein the second value combination can be a value combination corresponding to the target display information in the actual display of the history.
The training sample corresponding to the target presentation information may include a plurality of second value combinations corresponding to the target presentation information and actual return values corresponding to the second value combinations, each of the second value combinations corresponding to an actual return value. The second value combination corresponding to the target presentation information and the actual return value corresponding to the second value combination may be a value combination of the target presentation information in the actual presentation of the history and an actual return value of the value combination in the actual presentation of the history. For example, the second value combination corresponding to the target presentation information may be the historical value combination of the target presentation information in the recent period, that is, all value combinations used by the target presentation information in the actual presentation in the past period, and the actual return value corresponding to the second value combination is the actual return value of the target presentation information corresponding to all the used value combinations in the actual presentation in the past period. For another example, the target exhibition information may have a corresponding booth, and the target exhibition information may have been exhibited in a plurality of different booths in the past, so that the target booth to which the target exhibition information currently corresponds may be determined, and thus, the second value combination corresponding to the target exhibition information may be a value combination of which the target exhibition information corresponds to the target booth, that is, all value combinations of which the target exhibition information has been used in the actual exhibition of the target booth in the past, and the actual return value corresponding to the second value combination is an actual return value of which the target exhibition information corresponds to all the value combinations used in the actual exhibition of the target booth. For another example, the target display information may have a corresponding display period, where the display period is a certain period of the day, and the target display information may have been displayed in a plurality of different display periods in the past, so that the target display period currently corresponding to the target display information may be determined, and thus, the second value combination corresponding to the target display information may be a value combination of the target display information corresponding to the target display period, that is, all value combinations of the target display information used in the actual display of the target display period in the past, and the actual return value corresponding to the second value combination is an actual return value of the target display information corresponding to all the value combinations used in the actual display of the target display period. For another example, a target exhibition time period and a target exhibition position corresponding to the target exhibition information at present may be determined, so that the second value combination corresponding to the target exhibition information may be a value combination of the target exhibition information corresponding to the target exhibition position and the target exhibition time period, that is, all value combinations used by the target exhibition information in the past in the target exhibition time period and in the actual exhibition of the target exhibition position, and the actual return value corresponding to the second value combination is an actual return value of the target exhibition information corresponding to all the value combinations used in the target exhibition time period and in the time exhibition of the target exhibition position.
For example, the training samples may be directly obtained from a data storage space corresponding to the presentation information. After the training sample corresponding to the target display information is obtained, the first prediction model may be trained according to the training sample to obtain the target prediction model. Illustratively, the first prediction model may be trained by a machine learning algorithm (e.g., neural network learning) to obtain the target prediction model.
Illustratively, the target prediction model may be obtained using a neural network learning method. The following will describe in detail a process of constructing a target prediction model by using a supervised neural network training method, but the method provided by the present disclosure is not limited to the above learning method, and is not limited to this training method, and the following embodiments are only exemplary.
A set of proposed values and expected values (i.e., a first value combination) in the training sample are input into the first prediction model, and the connection weights of the neural network in the model are adjusted according to the difference between the actual output (the predicted return value generated by the first prediction model for the input first value combination) and the expected output (the actual return value corresponding to the first value combination) of the first prediction model. Initially, the connection weights of the neural network within the first predictive model may be randomly determined. And then, performing the operation on each first value combination in the training sample until the difference between the actual output and the expected output is smaller than a preset difference threshold value. When the difference between the actual output and the expected output of the first prediction model is smaller than the preset difference threshold, the current prediction of the first prediction model can reach a certain accuracy, and therefore the first prediction model can be determined as the target prediction model. In a possible embodiment, the target prediction model may be updated in real time, that is, in the process of using the target prediction model, the target display information is actually displayed once more, after the display, the actual return value corresponding to the actual display is collected, the value combination and the actual return value corresponding to the actual display are used as new training data, and the target prediction model is continuously corrected, so that the target decision model is more and more accurate.
Through the method, the first prediction model can be trained by utilizing the second value combination corresponding to the target display information and the actual return value corresponding to the second value combination to obtain the target prediction model. The second value combination corresponding to the target display information and the actual return value corresponding to the second value combination are easy to obtain, so that data acquisition is facilitated, and meanwhile, a target prediction model is convenient to obtain.
In another possible implementation, the training sample may include a third value combination corresponding to the historical display information and an actual reward value corresponding to the third value combination, in addition to the second value combination corresponding to the target display information and the actual reward value corresponding to the second value combination. The historical display information may be display information having a corresponding relationship with the target display information. The presentation information corresponding to the target presentation information may be, for example, presentation information of the same type as the target presentation information or having a competitive relationship with the target presentation information. For example, the same type may be the same type of information, such as displaying information for text, displaying information for video, displaying information for picture, and the like. For another example, the existence of the competitive relationship may mean that the target audience for displaying the information is the same, such as science and technology videos, and fun pictures.
The description of the limitation and the acquisition of the training samples, and the training of the first prediction model by using the training samples to obtain the target prediction model are similar to the above description and have the same principle, and are not repeated here.
Through the mode, besides the second value combination and the actual return value related to the target display information, the third value combination of the historical display information and the actual return value of the third value combination corresponding to the target display information are also considered, so that the collected data are more comprehensive, the historical display information can reflect the display information display condition of the global market to a certain extent, and compared with the actual return value of the second value combination and the second value combination corresponding to the target display information, the obtained target prediction model is more accurate.
It should be noted that, both the method and the process for constructing the model by using the machine learning algorithm are well known to those skilled in the art, and a brief description is given above for one possible case for the convenience of understanding, but the method for constructing the model in the present disclosure is not limited thereto, and is not described in detail for other implementation manners.
In step 13, the proposed value in the first value combination with the largest predicted return value is determined as the target proposed value of the target exhibition information, and the expected value in the first value combination with the largest predicted return value is determined as the target expected value of the target exhibition information.
For the obtained plurality of predicted return values, the proposed value in the largest first value combination can be determined as the target proposed value of the target presentation information, and the expected value in the first value combination with the largest predicted return value can be determined as the target expected value of the target presentation information.
In one possible embodiment, there may be a first value combination with the largest predicted return value, and then the proposed value corresponding to the largest predicted return value may be directly determined as the target proposed value and the expected value corresponding to the target proposed value may be determined as the target expected value, and then the set of target proposed value and target expected value may be used to compete for the exhibition right of the exhibition.
In another possible embodiment, there may be a plurality of first value combinations with the largest predicted return value, and step 13 may include the following steps:
if the first value combination with the largest predicted return value is multiple, determining the first value combination with the lowest expected value as a target first value combination;
and determining the proposed value in the target first value combination as the target proposed value of the target display information, and determining the expected value in the target first value combination as the target expected value of the target display information.
In practical application, the lower the expected value is, the lower the value of the current cost can be ensured under the condition that the same return value can be obtained, and the current expected value can be reduced as much as possible under the condition that the return value is ensured, so that the expected value of the current required cost is reduced, namely the cost of the current period is reduced. Therefore, if the first value combination having the largest predicted return value is plural, the first value combination having the lowest expected value can be determined as the target first value combination. And then, determining the proposed value in the target first value combination as the target proposed value of the target exhibition information, and determining the expected value in the target first value combination as the target expected value of the target exhibition information, so as to compete for exhibition permission of the exhibition position by using the set of the target proposed value and the target expected value.
In this way, it is possible to guarantee that the expected value currently spent is the lowest when the expected value is given, while guaranteeing a high return value.
In another possible embodiment, if the first value combination with the largest predicted return value is multiple, one of the first value combinations may be randomly selected, the drawn value of the value combination may be used as the target drawn value, the expected value of the value combination may be used as the target expected value, and then the set of target drawn value and the target expected value may be used to compete for the exhibition right of the exhibition position.
According to the scheme, aiming at the target display information, firstly, a plurality of first value combinations corresponding to the target display information are generated, then the first value combinations are input into a target prediction model to obtain prediction return values corresponding to the first value combinations, then the proposed value in the first value combination with the maximum prediction return value is determined as the target proposed value of the target display information from the plurality of prediction return values, and the expected value in the first value combination with the maximum prediction return value is determined as the target expected value of the target display information. In this way, the return value of the target display information is predicted through the target prediction model, and the drawn value and the expected value corresponding to the combination of the first value with the maximum predicted return value are determined as the target drawn value and the target expected value corresponding to the target display information, so that the target drawn value and the target expected value can be automatically generated aiming at the target display information, the step of manually determining the drawn value and the expected value is omitted, the labor is saved, the efficiency is improved, meanwhile, the generated target drawn value and the target expected value can obtain the optimal return value as much as possible, and the benefit of a user is maximized.
Fig. 3 is a block diagram of a value combination determination apparatus provided according to an embodiment of the present disclosure. As shown in fig. 3, the apparatus 30 may include:
a generating module 31, configured to generate a plurality of first value combinations corresponding to target display information, where a value combination corresponding to display information includes a proposed value of the display information and an expected value of the display information;
the processing module 32 is configured to input the first value combination into a target prediction model to obtain a prediction return value corresponding to the first value combination;
and the determining module 33 is configured to determine the proposed value in the first value combination with the largest predicted return value as the target proposed value of the target exhibition information, and determine the expected value in the first value combination with the largest predicted return value as the target expected value of the target exhibition information.
Optionally, the target prediction model is obtained by:
acquiring a training sample corresponding to the target display information, wherein the training sample at least comprises a second value combination corresponding to the target display information and an actual return value corresponding to the second value combination, and the second value combination is a value combination corresponding to the target display information in historical actual display;
and training a first prediction model according to the training samples to obtain the target prediction model.
Optionally, the training sample further includes a third value combination corresponding to historical display information and an actual return value corresponding to the third value combination, where the historical display information is display information having a corresponding relationship with the target display information.
Optionally, the generating module 31 includes:
the acquisition submodule is used for acquiring an expected value range and a proposed value range corresponding to the target display information;
a generating sub-module for generating the plurality of first value combinations according to the expected value range and the proposed value range.
Optionally, the determining module 33 includes:
a first determining sub-module, configured to determine, if the first value combination with the largest predicted return value is multiple, the first value combination with the lowest expected value as a target first value combination;
a second determining sub-module, configured to determine the proposed value in the target first value combination as a target proposed value of the target presentation information, and determine an expected value in the target first value combination as the target expected value of the target presentation information.
With regard to the apparatus in the above-described embodiment, the specific manner in which each module performs the operation has been described in detail in the embodiment related to the method, and will not be elaborated here.
FIG. 4 is a block diagram illustrating an electronic device in accordance with an example embodiment. For example, the electronic device 1900 may be provided as a server. Referring to fig. 4, an electronic device 1900 includes a processor 1922, which may be one or more in number, and a memory 1932 for storing computer programs executable by the processor 1922. The computer program stored in memory 1932 may include one or more modules that each correspond to a set of instructions. Further, the processor 1922 may be configured to execute the computer program to perform the value combination determination method described above.
Additionally, the electronic device 1900 may further include a power component 1926 and a communication component 1950, the power component 1926 may be configured to perform power management of the electronic device 1900, and the communication component 1950 may be configured to enable communication, e.g., wired or wireless communication, of the electronic device 1900. in addition, the electronic device 1900 may further include an input/output (I/O) interface 1958 the electronic device 1900 may operate based on an operating system stored in a memory 1932, e.g., WindowsServerTM, Mac OS XTM, UnixTM, &lTtTtTranslation = L &"," [ L &l Tt/T &gTt inxTM, and so on.
In another exemplary embodiment, there is also provided a computer-readable storage medium including program instructions which, when executed by a processor, implement the steps of the value combination determination method described above. For example, the computer readable storage medium may be the memory 1932 described above that includes program instructions executable by the processor 1922 of the electronic device 1900 to perform the value combination determination method described above.
The preferred embodiments of the present disclosure are described in detail with reference to the accompanying drawings, however, the present disclosure is not limited to the specific details of the above embodiments, and various simple modifications may be made to the technical solution of the present disclosure within the technical idea of the present disclosure, and these simple modifications all belong to the protection scope of the present disclosure.
It should be noted that the various features described in the above embodiments may be combined in any suitable manner without departing from the scope of the invention. In order to avoid unnecessary repetition, various possible combinations will not be separately described in this disclosure.
In addition, any combination of various embodiments of the present disclosure may be made, and the same should be considered as the disclosure of the present disclosure, as long as it does not depart from the spirit of the present disclosure.
Claims (12)
1. A value combination determination method, the method comprising:
generating a plurality of first value combinations corresponding to target display information, wherein the value combinations corresponding to the display information comprise the proposed value of the display information and the expected value of the display information;
inputting the first value combination into a target prediction model to obtain a prediction return value corresponding to the first value combination;
and determining the drawn value in the first value combination with the maximum predicted return value as the target drawn value of the target display information, and determining the expected value in the first value combination with the maximum predicted return value as the target expected value of the target display information.
2. The method of claim 1, wherein the target prediction model is obtained by:
acquiring a training sample corresponding to the target display information, wherein the training sample at least comprises a second value combination corresponding to the target display information and an actual return value corresponding to the second value combination, and the second value combination is a value combination corresponding to the target display information in historical actual display;
and training a first prediction model according to the training samples to obtain the target prediction model.
3. The method according to claim 2, wherein the training sample further includes a third value combination corresponding to historical presentation information and an actual return value corresponding to the third value combination, and the historical presentation information is presentation information corresponding to the target presentation information.
4. The method according to any one of claims 1 to 3, wherein the generating a plurality of first value combinations corresponding to the target presentation information comprises:
acquiring an expected value range and a proposed value range corresponding to the target display information;
generating the plurality of first value combinations according to the expected value range and the proposed value range.
5. The method according to any one of claims 1 to 3, wherein the determining the proposed value in the first value combination with the largest predicted return value as the target proposed value of the target presentation information and the expected value in the first value combination with the largest predicted return value as the target expected value of the target presentation information comprises:
if the first value combination with the largest predicted return value is multiple, determining the first value combination with the lowest expected value as a target first value combination;
determining a proposed value in the target first value combination as a target proposed value of the target presentation information, and determining an expected value in the target first value combination as the target expected value of the target presentation information.
6. A value combination determination apparatus, the apparatus comprising:
the generating module is used for generating a plurality of first value combinations corresponding to the target display information, wherein the value combinations corresponding to the display information comprise the proposed value of the display information and the expected value of the display information;
the processing module is used for inputting the first value combination into a target prediction model to obtain a prediction return value corresponding to the first value combination;
and the determining module is used for determining the proposed value in the first value combination with the maximum predicted return value as the target proposed value of the target display information, and determining the expected value in the first value combination with the maximum predicted return value as the target expected value of the target display information.
7. The apparatus of claim 6, wherein the target prediction model is obtained by:
acquiring a training sample corresponding to the target display information, wherein the training sample at least comprises a second value combination corresponding to the target display information and an actual return value corresponding to the second value combination, and the second value combination is a value combination corresponding to the target display information in historical actual display;
and training a first prediction model according to the training samples to obtain the target prediction model.
8. The apparatus of claim 7, wherein the training sample further comprises a third value combination corresponding to historical presentation information and an actual return value corresponding to the third value combination, and the historical presentation information is presentation information corresponding to the target presentation information.
9. The apparatus according to any one of claims 6-8, wherein the generating means comprises:
the acquisition submodule is used for acquiring an expected value range and a proposed value range corresponding to the target display information;
a generating sub-module for generating the plurality of first value combinations according to the expected value range and the proposed value range.
10. The apparatus of any of claims 6-8, wherein the means for determining comprises:
a first determining sub-module, configured to determine, if the first value combination with the largest predicted return value is multiple, the first value combination with the lowest expected value as a target first value combination;
a second determining sub-module, configured to determine the proposed value in the target first value combination as a target proposed value of the target presentation information, and determine an expected value in the target first value combination as the target expected value of the target presentation information.
11. A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the method according to any one of claims 1 to 5.
12. An electronic device, comprising:
a memory having a computer program stored thereon;
a processor for executing the computer program in the memory to carry out the steps of the method of any one of claims 1 to 5.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811644344.0A CN111401593A (en) | 2018-12-29 | 2018-12-29 | Value combination determination method and device, readable storage medium and electronic equipment |
| PCT/CN2019/094747 WO2020134012A1 (en) | 2018-12-29 | 2019-07-04 | Value combination determination method and apparatus, storage medium and electronic device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811644344.0A CN111401593A (en) | 2018-12-29 | 2018-12-29 | Value combination determination method and device, readable storage medium and electronic equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111401593A true CN111401593A (en) | 2020-07-10 |
Family
ID=71128255
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811644344.0A Pending CN111401593A (en) | 2018-12-29 | 2018-12-29 | Value combination determination method and device, readable storage medium and electronic equipment |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN111401593A (en) |
| WO (1) | WO2020134012A1 (en) |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7725502B1 (en) * | 2005-06-15 | 2010-05-25 | Google Inc. | Time-multiplexing documents based on preferences or relatedness |
| CN103136311B (en) * | 2011-12-05 | 2017-12-19 | 微软技术许可有限责任公司 | The combined optimization of bid and budget allocation in paid search |
| CN103164805A (en) * | 2011-12-19 | 2013-06-19 | 阿里巴巴集团控股有限公司 | Keyword putting price optimizing process method and system |
| CN102592235A (en) * | 2011-12-28 | 2012-07-18 | 北京品友互动信息技术有限公司 | Internet advertisement serving system |
| US9911140B2 (en) * | 2013-08-05 | 2018-03-06 | Excalibur Ip, Llc | Keyword price recommendation |
| CN107194729A (en) * | 2017-05-22 | 2017-09-22 | 北京京东尚科信息技术有限公司 | Advertisement price competing method, device, electronic installation and computer-readable medium |
-
2018
- 2018-12-29 CN CN201811644344.0A patent/CN111401593A/en active Pending
-
2019
- 2019-07-04 WO PCT/CN2019/094747 patent/WO2020134012A1/en active Application Filing
Also Published As
| Publication number | Publication date |
|---|---|
| WO2020134012A1 (en) | 2020-07-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106202453B (en) | Multimedia resource recommendation method and device | |
| CN103383655B (en) | For the performance interference model for the workload that merging is managed in perceiving cloud in QOS | |
| CN106557576B (en) | Prompt message recommendation method and device based on artificial intelligence | |
| CN109299344A (en) | The generation method of order models, the sort method of search result, device and equipment | |
| CN109684947A (en) | Mark quality control method, device, computer equipment and storage medium | |
| CN106302579A (en) | The distribution method of a kind of task and server | |
| CN101867637A (en) | Method and terminal for dynamically generating standby interface | |
| CN109547808A (en) | Data processing method, device, server and storage medium | |
| CN109587530B (en) | Data processing method and device, terminal equipment and storage medium | |
| CN111241381A (en) | Information recommendation method and device, electronic equipment and computer-readable storage medium | |
| CN105912549A (en) | Content recommendation method and device thereof | |
| CN111199540A (en) | Image quality evaluation method, image quality evaluation device, electronic device, and storage medium | |
| CN112465565B (en) | User portrait prediction method and device based on machine learning | |
| CN111400616B (en) | Account recommendation method and account recommendation device | |
| CN111401593A (en) | Value combination determination method and device, readable storage medium and electronic equipment | |
| CN109948050B (en) | Information processing method and apparatus, storage medium, and electronic device | |
| CN111460301A (en) | Object pushing method and device, electronic equipment and storage medium | |
| CN111143660A (en) | Intelligent recommendation method and device for knowledge platform and computer readable storage medium | |
| CN109886738A (en) | A kind of intelligence exhibition user in predicting method and apparatus | |
| CN115185606A (en) | Method, device, equipment and storage medium for obtaining service configuration parameters | |
| CN113420165B (en) | Training of classification model and classification method and device of multimedia data | |
| CN112565797B (en) | Page request processing method and device, electronic equipment and medium | |
| CN111598390B (en) | Method, device, equipment and readable storage medium for evaluating high availability of server | |
| CN105142165B (en) | Search the method and device of terminal device | |
| CN112905896A (en) | Training method of recommended number model, and mixed content recommendation 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 |