CN114398558B - Information recommendation method, device, electronic equipment and storage medium - Google Patents

Information recommendation method, device, electronic equipment and storage medium Download PDF

Info

Publication number
CN114398558B
CN114398558B CN202210063127.2A CN202210063127A CN114398558B CN 114398558 B CN114398558 B CN 114398558B CN 202210063127 A CN202210063127 A CN 202210063127A CN 114398558 B CN114398558 B CN 114398558B
Authority
CN
China
Prior art keywords
information
matching degree
recommended
determining
data
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN202210063127.2A
Other languages
Chinese (zh)
Other versions
CN114398558A (en
Inventor
周成瑜
曲巍
江兴何
黄庆亚
李晨阳
顾文婷
余飞
周天力
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beijing Baidu Netcom Science and Technology Co Ltd
Original Assignee
Beijing Baidu Netcom Science and 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 Beijing Baidu Netcom Science and Technology Co Ltd filed Critical Beijing Baidu Netcom Science and Technology Co Ltd
Priority to CN202210063127.2A priority Critical patent/CN114398558B/en
Publication of CN114398558A publication Critical patent/CN114398558A/en
Application granted granted Critical
Publication of CN114398558B publication Critical patent/CN114398558B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90—Details of database functions independent of the retrieved data types
    • G06F16/95—Retrieval from the web
    • G06F16/953—Querying, e.g. by the use of web search engines
    • G06F16/9535—Search customisation based on user profiles and personalisation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • 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
    • G06Q30/0241—Advertisements
    • G06Q30/0251—Targeted advertisements
    • G06Q30/0269—Targeted advertisements based on user profile or attribute
    • G06Q30/0271—Personalized advertisement

Abstract

The disclosure provides an information recommendation method, an information recommendation device, electronic equipment and a storage medium, and relates to the technical field of Internet, in particular to the technical field of intelligent recommendation and user understanding. The specific implementation scheme of the information recommendation method is as follows: in response to obtaining object information of a target object, determining a first matching degree between information to be recommended and the target object according to the object information; and determining that the information to be recommended is the candidate information for the target object in response to the first matching degree being greater than or equal to a matching degree threshold value for the information to be recommended, wherein an initial value of the matching degree threshold value is determined according to offline flow data of the information to be recommended, and the matching degree threshold value is adjusted by adopting online flow data of the information to be recommended.

Description

Information recommendation method, device, electronic equipment and storage medium
Technical Field
The disclosure relates to the technical field of internet, in particular to the field of intelligent recommendation and the field of user understanding, and especially relates to an information recommendation method, an information recommendation device, electronic equipment and a storage medium.
Background
With the development of internet technology, the conversion rate requirement on the put information is gradually increased. And whether the information can be accurately put in is a main factor affecting the conversion rate. The conversion process of the information may refer to a process in which a user selects a corresponding product by browsing the information.
Disclosure of Invention
Based on this, the present disclosure provides an information recommendation method, apparatus, device, and storage medium aiming at improving information conversion rate.
According to one aspect of the present disclosure, there is provided an information recommendation method including: in response to obtaining object information of a target object, determining a first matching degree between information to be recommended and the target object according to the object information; and determining the information to be recommended as the alternative information for the target object in response to the first matching degree being greater than or equal to a matching degree threshold for the information to be recommended. The initial value of the matching degree threshold is determined according to the offline flow data of the information to be recommended, and the matching degree threshold is adjusted by adopting the online flow data of the information to be recommended.
According to another aspect of the present disclosure, there is provided an information recommendation apparatus including: the first matching determining module is used for determining a first matching degree between the information to be recommended and the target object according to the object information in response to the acquired object information of the target object; and an alternative determining module, configured to determine, in response to the first matching degree being greater than or equal to a matching degree threshold for the information to be recommended, that the information to be recommended is alternative information for the target object, where an initial value of the matching degree threshold is determined according to offline flow data of the information to be recommended, and the matching degree threshold is adjusted by using online flow data of the information to be recommended.
According to another aspect of the present disclosure, there is provided an electronic device including: at least one processor; and a memory communicatively coupled to the at least one processor; wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the information recommendation method provided by the present disclosure.
According to another aspect of the present disclosure, there is provided a non-transitory computer-readable storage medium storing computer instructions for causing a computer to perform the information recommendation method provided by the present disclosure.
According to another aspect of the present disclosure, there is provided a computer program product comprising a computer program/instruction which, when executed by a processor, implements the information recommendation method provided by the present disclosure.
It should be understood that the description in this section is not intended to identify key or critical features of the embodiments of the disclosure, nor is it intended to be used to limit the scope of the disclosure. Other features of the present disclosure will become apparent from the following specification.
Drawings
The drawings are for a better understanding of the present solution and are not to be construed as limiting the present disclosure. Wherein:
Fig. 1 is an application scenario schematic diagram of an information recommendation method and apparatus according to an embodiment of the present disclosure;
FIG. 2 is a flow diagram of an information recommendation method according to an embodiment of the present disclosure;
FIG. 3 is a schematic diagram of a method for determining a match threshold from offline flow data according to an embodiment of the present disclosure;
FIG. 4 is a schematic diagram of determining a match threshold according to an embodiment of the present disclosure;
FIG. 5 is a schematic diagram of a mechanism for adjusting a match threshold based on-line traffic data in accordance with an embodiment of the present disclosure;
FIG. 6 is a block diagram of an information recommendation device according to an embodiment of the present disclosure; and
fig. 7 is a block diagram of an electronic device for implementing an information recommendation method of an embodiment of the present disclosure.
Detailed Description
Exemplary embodiments of the present disclosure are described below in conjunction with the accompanying drawings, which include various details of the embodiments of the present disclosure to facilitate understanding, and should be considered as merely exemplary. Accordingly, one of ordinary skill in the art will recognize that various changes and modifications of the embodiments described herein can be made without departing from the scope and spirit of the present disclosure. Also, descriptions of well-known functions and constructions are omitted in the following description for clarity and conciseness.
The disclosure provides an information recommendation method, which comprises a matching determination stage and an alternative determination stage. In the matching determination stage, in response to obtaining object information of a target object, a first degree of matching between information to be recommended and the target object is determined according to the object information. In the alternative determining stage, determining that the information to be recommended is the alternative information for the target object in response to the first matching degree being greater than or equal to a matching degree threshold for the information to be recommended. The initial value of the matching degree threshold is determined according to the offline flow data of the information to be recommended, and the matching degree threshold is adjusted by adopting the online flow data of the information to be recommended.
An application scenario of the method and apparatus provided by the present disclosure will be described below with reference to fig. 1.
Fig. 1 is an application scenario schematic diagram of an information recommendation method and apparatus according to an embodiment of the present disclosure.
As shown in fig. 1, the scenario 100 of this embodiment includes a user 110, a terminal device 120, and a server 130.
The terminal device 120 may be communicatively coupled to the server 130 via a network. Wherein the network may comprise a wired or wireless network.
The terminal device 120 may be, for example, any electronic device having processing capabilities including, but not limited to, a smart phone, a tablet computer, a laptop portable computer, a desktop computer, and the like. For example, the terminal device 120 may be installed with various client applications, such as a news browsing-type application, a game-type application, an instant messaging-type application, and the like (by way of example only).
The server 130 may be a server providing various services, such as a background management server that may provide support for applications running on the terminal device 120. The server may be a cloud server, a server of a distributed system, or a server incorporating a blockchain.
In an embodiment, the terminal device 120 may respond to an opening or refreshing operation of the client application by the user 110, and may send the object information 140 of the target object to the server 130 with the user 110 as the target object. The server 130 may, for example, in response to receiving the object information 140, determine recommendation information 150 matching the user 110 according to the preferences of the user 110, and send the recommendation information 150 to the terminal device 120 for presentation.
In an embodiment, the server 130 may maintain a matching degree threshold for each information to be recommended in the database 160, and take the information to be recommended as the candidate recommendation information if it is determined from the user representation of the user 110 that the matching degree of the user and the information to be recommended is greater than the matching degree threshold. It is to be appreciated that the threshold level of matching maintained by the server 130 can be determined by the server 130 or by other servers in communication with the server 130.
It should be noted that, each step in the information recommendation method provided in the embodiments of the present disclosure may be generally performed by the server 130, or may be partially performed by the server 130, and partially performed by another server communicatively connected to the server 130. Accordingly, each module in the information recommendation apparatus provided in the embodiments of the present disclosure may be disposed in the server 130, or may be partially disposed in the server 130, and partially disposed in another server communicatively connected to the server 130.
It should be understood that the number and types of terminal devices, servers, and databases in fig. 1 are merely illustrative. There may be any number and type of terminal devices, servers, and databases as desired for implementation.
The information recommendation method provided by the present disclosure will be described in detail below with reference to fig. 1 through fig. 2 to 5.
Fig. 2 is a flowchart illustrating an information recommendation method according to an embodiment of the present disclosure.
As shown in fig. 2, the information recommendation method 200 of this embodiment may include operations S210 to S220.
In operation S210, in response to acquiring object information of a target object, a first degree of matching between information to be recommended and the target object is determined according to the object information.
According to an embodiment of the disclosure, the target object may be, for example, a user of the terminal device, and the object information may include, for example, identification information of the target object, which may be an account identification of the client application.
After receiving the object information, the server may determine attribute information and/or portrait characteristics of the target object based on the object information. For example, the attribute information and/or portrait characteristics determined from the object information may include the number of information browsed in a predetermined period, the number of times information is clicked in a predetermined period, and the like. It will be appreciated that the attribute information and/or the portrait characteristic of the target object includes information that is authorized and available for the target object.
The embodiment can determine, for each piece of information to be recommended in the database, a first matching degree between each piece of information to be recommended and the target object according to the determined attribute information and/or portrait characteristic of the target object and the characteristic of each piece of information to be recommended.
In one embodiment, a matching degree prediction model may be employed to determine the first matching degree. For example, the determined attribute information and/or portrait features of the target object and the features of each piece of information to be recommended may be spliced to be used as input of a matching degree estimation model, and the matching degree estimation model outputs the first matching degree. The matching degree estimation model can be a conversion rate estimation model or a click rate estimation model, and specifically can comprise a cyclic neural network model, a deep neural network model and the like.
In one embodiment, the matching degree estimation model may include a two-tower deep neural network model composed of a conversion rate estimation model and a click rate estimation model. The embodiment can be used for inputting the conversion rate estimation model and the click rate estimation model simultaneously after the determined attribute information and/or the portrait characteristic of the target object and the characteristic of each piece of information to be recommended are spliced, and taking the product of the conversion rate estimated by the conversion rate estimation model and the click rate estimated by the click rate estimation model as the first matching degree.
For example, in the double-tower deep neural network model, the structural complexity of the click rate estimation model may be higher than that of the conversion rate estimation model, so as to improve the learning capability of the matching rate estimation model. This is because, in a normal case, the positive sample size of the click rate estimation is larger than the positive sample size of the conversion rate, and by designing a more complex structure for the click rate estimation model, the learning ability of the click rate estimation model can be improved on the premise of ensuring the convergence of the model.
In an embodiment, the matching degree estimation model may be adjusted periodically, and when the method for recommending information is executed, the latest matching degree estimation model may be obtained to perform matching degree estimation.
In an embodiment, in determining the first matching degree, in addition to the determined attribute information and/or portrait characteristic of the target object and the characteristic of each piece of information to be recommended, a cross characteristic between the target object and the information to be recommended may be considered. For example, the cross feature may include an association between the type of the target object and the type of the information to be recommended. For example, if the preference of the target object of the type a to the information to be recommended of the type a is high, the cross feature may include a feature pair formed by the type a and the type a.
In operation S220, in response to the first matching degree being greater than or equal to the matching degree threshold for the information to be recommended, the information to be recommended is determined to be the candidate information for the target object.
According to an embodiment of the disclosure, each piece of information to be recommended may have a unique corresponding matching degree threshold value, which is used as a basis for determining whether the piece of information to be recommended is the candidate information. The embodiment may determine that the certain information to be recommended is the candidate information for the target object when the first matching degree between the target object and the certain information to be recommended is greater than or equal to a matching degree threshold value uniquely corresponding to the certain information to be recommended. It can be understood that, for different information to be recommended, the values of the unique corresponding matching degree thresholds may be equal or unequal.
According to an embodiment of the disclosure, a matching degree threshold value uniquely corresponding to a certain information to be recommended may have an initial value, for example, and the matching degree threshold value may be adjusted according to online traffic data. The initial value may be determined according to the offline flow data of the certain information to be recommended. If the unit of day is taken, the offline traffic data may refer to traffic data within a predetermined period of time before the day, and the online traffic data may refer to traffic data before the current time in the day. It is to be understood that the above-described on-line traffic data, off-line traffic data, and units (days) are merely examples to facilitate an understanding of the present disclosure, which is not limited thereto.
For example, the traffic data may include presentation time information of information to be recommended, object information of a presentation object (i.e., a recommended object) of the information to be recommended, operation information of the recommended object on the information to be recommended, and the like. The operation information may include, for example, a click operation, a conversion operation for a recommended object in the information to be recommended, and the like. The transformation operation may be different for different types of recommended objects. For example, if the information to be recommended is an APP promotion advertisement, the recommended object is an APP, and the conversion operation may include a downloading operation and/or a registration operation of the APP; if the information to be recommended is the article promotion advertisement, the recommended object is the article, and the conversion operation can comprise the order placing operation of the article and the like.
For example, the matching degree threshold for the information to be recommended may be determined with the conversion rate of the information to be recommended reaching the target conversion rate as a target. For example, a minimum value of the matching degree required to reach the target conversion rate may be determined according to the offline flow data of the information to be recommended, and the minimum value may be used as an initial value of the matching degree threshold. In this way, the conversion rate of the information to be recommended can be made to reach the target conversion rate by recommending the information to be recommended to the object having the degree of matching with the information to be recommended greater than or equal to the initial value.
For example, the on-line conversion rate of the information to be recommended may be determined according to the on-line traffic data, and when the on-line conversion rate is lower than the target conversion rate, for example, the matching degree threshold may be appropriately increased so that the information to be recommended may be recommended to a more matched object to increase the on-line conversion rate of the object to be recommended.
For example, when the online conversion rate is higher than the target conversion rate, the embodiment may also appropriately reduce the matching degree threshold so that the information to be recommended is recommended to more objects, and the exposure of the information to be recommended is increased.
In summary, it can be known that, according to the information recommendation method disclosed by the embodiment of the disclosure, whether to take the information to be recommended as the candidate information is determined by setting the matching degree threshold for the information to be recommended, so that the conversion rate of the information to be recommended can be improved to a certain extent. Furthermore, the matching degree threshold value not only determines an initial value according to the off-line flow data, but also adjusts the value of the matching degree threshold value according to the on-line flow data, so that the condition that the conversion rate cannot be ensured due to large difference between the off-line flow and the on-line flow can be avoided, and the information to be recommended is further ensured to have a large conversion rate.
The principle of determining the initial value of the matching threshold from the off-line flow data will be described in detail below with reference to fig. 3 to 4.
Fig. 3 is a schematic diagram of a principle of determining a matching degree threshold from offline flow data according to an embodiment of the present disclosure.
As shown in fig. 3, in this embodiment 300, when determining the matching degree threshold according to the offline flow data, a second matching degree 320 between the offline flow data 310 and the recommended object 311 may be determined in response to acquiring the offline flow data 310 of any one piece of information to be recommended. Wherein the recommended object 311 may be determined according to the offline traffic data. The second matching degree 320 is similar to the first matching degree described above, and will not be described herein. A minimum matching degree that enables the actual conversion rate for the information to be recommended to reach the target conversion rate may then be determined as a matching degree threshold 330 based on the second matching degree 320 and the operation information 312. The offline flow data 310 includes object information of the recommended object 311, operation information 312 of the recommended object 311 on the information to be recommended, and the like. The object information of the recommended object 311 is similar to the object information of the target object described above, and the operation information 312 may be a click operation, a conversion operation, or the like described above.
In this embodiment 300, the offline flow data 310 may include m pieces of accumulated flow data, and this embodiment may first determine object information of a plurality of recommended objects included in the m pieces of flow data. And then, determining a second matching degree between the information to be recommended and each recommended object according to each recommended object in the m recommended objects, and obtaining m second matching degrees corresponding to the m flow data. And then ordering the m second matching degrees from high to low to form a matching degree sequence. For the matching degree sequence, the conversion rate of the information to be recommended can be determined according to i pieces of operation information included in the offline flow data corresponding to the first i pieces of second matching degrees from the second matching degrees arranged at the first bit. The conversion is then determined to be greater than the target conversion, and if so, i is set to i+1. If the conversion rate is smaller than the actual conversion rate, the conversion rate determined according to the offline flow data corresponding to the previous (i-1) second matching degree is taken as the actual conversion rate, and the (i-1) th second matching degree is taken as a matching degree threshold 330. Wherein m is an integer greater than 1, and the value range of i is greater than or equal to 1 and less than or equal to m.
In an embodiment, before determining the matching degree threshold according to the offline traffic data, it may further be determined whether the offline accumulated traffic data 340 satisfies a predetermined condition, and if so, the offline accumulated traffic data 340 is taken as the offline traffic data 310, thereby determining that the offline traffic data is acquired. If not, continuing to accumulate the flow data on line. Wherein the predetermined condition may be used to limit the amount of flow data accumulated offline and/or to limit the conversion determined by the flow data accumulated offline, etc. In this way, stability and accuracy of the determined matching degree threshold value can be ensured.
Illustratively, the predetermined condition may include at least one of: the number of the off-line accumulated flow data is equal to or greater than a predetermined number threshold, the number of the target flow data in the off-line accumulated flow data is equal to or greater than a first threshold, the conversion rate determined from the operation information included in the off-line accumulated flow data is equal to or greater than the target conversion rate, and the like. It will be appreciated that this embodiment may accumulate flow data in real time without interruption, and determine the conversion number of the information to be recommended based on the accumulated flow data. The conversion number is the number of the target flow data, that is, the number of the flow data of which the operation information indicates that the information to be recommended is converted. Wherein, in the case where the operation information includes a conversion operation, it may be determined that the operation information indicates that the information to be recommended is converted. Alternatively, an identification bit may be added to the operation information by a developer of the information to be recommended, and the identification bit may be used to indicate whether the information to be recommended is converted. For example, the embodiment may also determine a historical total conversion number of the information to be recommended based on all traffic data of the information to be recommended. The first threshold value may be any value such as 20 or 30, for example, which is not limited in the present disclosure.
Wherein the start time of the offline accumulated flow data may be a start time of day. Alternatively, for the current date, from all the traffic data accumulated offline, traffic data generated before the current date may be acquired one by one from the back to the front according to the generation time until the acquired traffic data satisfies a predetermined condition, thereby obtaining offline traffic data 310. The process of acquiring flow data one by one can be understood as a process of accumulating flow data offline.
In an embodiment, the first matching degree and the second matching degree may be determined by using the matching degree estimation model described above. The foregoing predetermined condition may further include: when the flow data accumulated offline are used as test data of the matching degree prediction model, the sorting index of the matching degree prediction model is larger than or equal to an index threshold value.
For example, the embodiment may obtain sample data according to the accumulated flow data under each line, where the sample data is similar to the data input to the matching degree estimation model described above, and will not be described herein. And inputting the obtained sample data into a matching degree prediction model to obtain the prediction matching degree. If the predicted matching degree is greater than or equal to the predetermined matching degree, determining that the predicted conversion result corresponding to the flow data accumulated under each line is converted, otherwise determining that the predicted conversion result is not converted. If the predicted conversion result and the conversion result indicated by the operation information included in the flow data accumulated under each line are both converted, the flow data accumulated under each line is true positive data. If the predicted conversion result is converted, the conversion result indicated by the operation information included in the flow data accumulated under each line is not converted, and the flow data accumulated under each line is pseudo-cation data. The embodiment can count the proportion of false positive data and the proportion of true positive data in all the flow data accumulated under the line, and the proportion are respectively used as false positive rate and true positive rate. A ranking index is determined based on the false positive rate and the true positive rate. The ranking index may be, for example, a AUC (Area Under Curve) index, i.e., the area under the subject's working characteristics curve (Receiver Operating Characteristic Curve, ROC). The ROC curve is a curve drawn with the true positive rate as the ordinate and the false positive rate as the abscissa. When the ranking index is an AUC index, the index threshold may be any value less than 1, such as 0.6 or 0.65. It is to be understood that the foregoing types of ranking indicators and values of the indicator thresholds are merely examples to facilitate understanding of the present disclosure, and the present disclosure is not limited thereto.
The predetermined conditions in this embodiment may enable the acquired offline flow data to be more reliable by defining the ranking index of the matching degree prediction model, and thus facilitate improving the accuracy and stability of the determined matching degree threshold.
Fig. 4 is a schematic diagram of determining a match threshold according to an embodiment of the present disclosure.
According to embodiments of the present disclosure, a binning principle may be employed to determine a match threshold from offline traffic data. In this way, the efficiency of determining the matching degree threshold value and the robustness of the matching degree threshold value can be improved.
As shown in fig. 4, in determining the matching degree threshold, m pieces of traffic data may be divided into n data groups ordered from large to small according to the aforementioned m pieces of second matching degrees. Wherein, each data group corresponds to a matching degree interval. Wherein n is an arbitrary value of 1 or more and m or less.
Illustratively, the embodiment 400 may set n buckets according to the matching degree, each bucket corresponding to one matching degree interval, and each bucket corresponding to one data set. For example, bucket 401 in the first position corresponds to a matching interval of [0.9,1], bucket 402 in the second position corresponds to a matching interval of [0.8,0.9 ], bucket 403 in the third last position corresponds to a matching interval of [0.2,0.3 ], bucket 404 in the second last position corresponds to a matching interval of [0.1, 0.2), and bucket 405 in the last position corresponds to a matching interval of [0,0.1). The embodiment may divide the m traffic data into n buckets according to a matching degree interval in which the m second matching degrees are located. It is understood that the setting of the matching degree intervals corresponding to the n barrels may be uniformly set, or may be unevenly set, which is not limited in this disclosure. For example, a greater interval of matching may be set for a bucket with a sparse data distribution and a lesser interval of matching may be set for a bucket with a dense data distribution.
Subsequently, a target data set of the n data sets and an ordering position p of the target data set may be determined. The actual conversion rate determined according to the operation information included in the 1 st to p-th data sets in the m data sets is greater than or equal to the target conversion rate, and the actual conversion rate determined according to the operation information included in the 1 st to (p+1) -th data sets in the m data sets is less than the target conversion rate. The embodiment may use the lower limit value of the matching degree interval corresponding to the target data set as the matching degree threshold value. Or, the sum of the lower limit value and the preset value can be used as a matching degree threshold value, so that the conversion rate of the information to be recommended can be further ensured to reach the target conversion rate. Wherein the predetermined value is a value greater than zero.
For example, in embodiment 400, if the optimization objective is to increase the conversion by 10%, and the actual conversion determined from the operation information included in the m flow rate data is set to q=b, the objective conversion may be determined to be 1.1b. If the data set corresponding to bucket 403 in the n buckets is the target data set, the actual conversion rate Qr determined according to the operation information included in the buckets 401 to 403 is greater than or equal to 1.1b, and the actual conversion rate Qr2 determined according to the operation information included in the buckets 401 to 404 is less than 1.1b. This embodiment may determine the matching degree threshold from the lower limit value 0.2 of the matching degree interval [0.2,0.3 ] corresponding to the bucket 403.
According to the embodiment of the disclosure, in the case that the sum of the lower limit value and the predetermined value is used as the matching degree threshold value, the information recommendation method of the embodiment may further perform statistics on the online traffic data to determine the conversion number of the information to be recommended on the online. If the conversion number is greater than the predetermined conversion number, the matching degree threshold may be adjusted to a predetermined threshold. The predetermined threshold is less than the pre-adjustment matching degree threshold. I.e. if the conversion number is greater than the predetermined conversion number, the matching degree threshold is lowered. Therefore, the exposure of the information to be recommended can be increased as much as possible while the conversion rate of the information to be recommended on the line reaches the target conversion rate. Therefore, the higher benefit of the processing platform side of the recommendation information can be ensured to a certain extent.
The predetermined conversion number may be determined according to, for example, an average exposure amount per unit time of the information to be recommended and the target conversion rate. For example, the product of the average exposure amount and the target conversion rate may be taken as the predetermined conversion number. The predetermined threshold may be, for example, a value arbitrarily close to zero, such as zero, which is not limited by the present disclosure.
The principle of adjusting the matching threshold according to the on-line traffic data will be described in detail below with reference to fig. 5.
Fig. 5 is a schematic diagram of adjusting a matching degree threshold according to on-line traffic data according to an embodiment of the present disclosure.
As shown in fig. 5, in this embodiment 500, after the online traffic data 510 of the information to be recommended is obtained, the online conversion 520 of the information to be recommended may be determined according to the online traffic data 510, where the online conversion is real-time. The match threshold is then adjusted based on the difference 540 between the in-line conversion 520 and the target conversion 530. It will be appreciated that the match threshold may be adjusted periodically based on the on-line traffic data. For example, the matching degree threshold is adjusted once every unit period. The length of the unit time period may be, for example, 1h, 0.5h, or the like, which is set according to actual requirements.
For example, the on-line traffic data, similar to the off-line traffic data, may include operational information. The embodiment may determine whether each of the on-line traffic data is converted based on the operation information included in the on-line traffic data, thereby statistically obtaining an on-line conversion 520. This embodiment may decrease the matching degree threshold when the difference between the target conversion 530 and the in-line conversion 520 is greater than the first value. The match threshold may be increased when the difference between the target conversion 530 and the in-line conversion 520 is less than the second value. Wherein the first value is a value greater than zero and the second value is a value less than zero. The adjustment amplitude of the matching degree threshold may be a predetermined amplitude, for example, may be 0.1, which is not limited in the present disclosure.
In one embodiment, the adjustment step of the match threshold may be determined based on the difference 540 between the target conversion 530 and the in-line conversion 520. And then adjusting the matching degree threshold according to the adjustment step length. For example, the absolute value of the difference between the target conversion 530 and the in-line conversion 520 may be positively correlated with the adjustment step size. The larger the absolute value of the difference, the larger the adjustment step.
For example, the absolute value of the difference between the target conversion 530 and the in-line conversion 520 may be proportional to the adjustment step size, and the scaling factor may be any value such as 0.1, which is not limited by the present disclosure.
The adjustment step length is determined according to the difference, so that the adjustment of the matching degree threshold value is more accurate, the on-line conversion rate of the information to be recommended is closer to the target conversion rate, and the conversion rate and the exposure of the information to be recommended can be balanced well.
In an embodiment, when determining the adjustment step according to the difference 540, the base 550 of the adjustment step may be determined according to the difference 540, and then the weight 570 of the adjustment step may be determined according to the real-time exposure 560 of the information to be recommended in unit time, that is, the base is given weight. Finally, an adjustment step 580 is determined based on the weights and cardinality. For example, the product of weight 570 and cardinality 550 may be used as adjustment step 580. In this way, it is possible to sufficiently consider the case where the reference value of the on-line conversion rate determined from the on-line flow rate data is different due to the difference in the exposure amount of the data of different periods, and thus to improve the adjustment accuracy of the matching degree threshold. It is to be understood that the above-mentioned online traffic data may be traffic data generated at least one unit period before the current time, or in the case of information recommendation in units of days, the online traffic data may be all online traffic data that has been generated in the current day.
For example, the base 550 of the adjustment step may be determined using proportional integral derivative (Proportional Integral Derivative, PID algorithm for example, if e is the target conversion minus the on-line conversion, then the base ΔS may be expressed using the following equation:
Figure BDA0003478782840000121
wherein DeltaS j The base of the adjustment step, e, determined for the jth period in the periodic adjustment j The difference between the target conversion determined for this j-th cycle and the on-line conversion. e, e j-1 The difference between the target conversion determined for cycle (j-1) and the on-line conversion. e, e k The difference between the target conversion determined for the kth cycle and the on-line conversion. Alpha 1 、α 2 、α 3 Is an empirical value.
For example, the ratio between the real-time exposure amount and the predetermined exposure amount may be taken as the weight of the adjustment step. Alternatively, the root mean square of the ratio between the real-time exposure amount and the predetermined exposure amount is used as the weight for adjusting the step size. For example, the ratio between the time exposure and the predetermined exposure is positively correlated with the weight 570. It will be appreciated that the above-described method of determining the weight of the adjustment step is merely an example to facilitate an understanding of the present disclosure, which is not limited thereto.
In an embodiment, a larger adjustment step is allocated for the case of increasing the matching degree threshold value, and a smaller adjustment step is allocated for the case of decreasing the matching degree threshold value, so that excessive exposure caused by excessively decreasing the matching degree threshold value is avoided, and real-time adjustment of the matching degree threshold value is facilitated.
For example, in determining the weight of the adjustment step based on the real-time exposure amount and the predetermined exposure amount, it may be determined first whether the cardinality obtained from the difference is smaller than zero. If the cardinality is equal to or greater than zero, the value of the first ratio of the real-time exposure to the predetermined exposure may be used as the weight 570. If the cardinality is less than zero, a value of a second ratio of the real-time exposure to the predetermined exposure may be used as the weight 570. Wherein the second ratio is less than the first ratio. In this way, different adjustment steps can be allocated for the down and up cases. Wherein the first ratio may be 1, the second ratio may be 0.5, etc., which is not limited by the present disclosure.
In an embodiment, as shown in fig. 5, before the matching degree threshold is adjusted according to the online traffic data, it may also be determined whether the number of the online accumulated traffic data 590 is greater than the second threshold, and if so, the online accumulated traffic data 590 is taken as the online traffic data 510, so as to determine that the online traffic data is acquired. If not, continuing to accumulate the flow data on the line. Thus, the accuracy and effectiveness of the adjustment of the matching degree threshold can be improved. The second threshold may be set according to actual requirements, for example, may be any value such as 10, 20, etc., which is not limited in this disclosure.
For example, different second thresholds may also be assigned to different types of information to be recommended. For example, the second threshold may be positively correlated with the conversion complexity of the information to be recommended. The more complex the transformation of the information to be recommended, the higher the second threshold. In this way, the accuracy of the on-line conversion determined from the on-line flow data may be ensured, and thus the accuracy of the matching degree threshold adjustment may be facilitated to be improved. Wherein the conversion complexity may for example be positively correlated with the number of operational steps required for conversion. For example, for the downloading operation and registration operation of the APP, the operation steps are more, and for the ordering operation, the operation steps are fewer.
Based on the information recommendation method provided by the disclosure, the disclosure also provides an information recommendation device. The device will be described in detail below in connection with fig. 6.
Fig. 6 is a block diagram of a structure of an information recommendation apparatus according to an embodiment of the present disclosure.
As shown in fig. 6, the information recommendation apparatus 600 of this embodiment may include a first match determination module 610 and an alternative determination module 620.
The first matching determining module 610 is configured to determine, in response to obtaining object information of the target object, a first matching degree between the information to be recommended and the target object according to the object information. In an embodiment, the first matching determination module 610 may be configured to perform the operation S210 described above, which is not described herein.
The candidate determining module 620 is configured to determine, in response to the first matching degree being greater than or equal to a matching degree threshold for the information to be recommended, that the information to be recommended is candidate information for the target object. The initial value of the matching degree threshold is determined according to the offline flow data of the information to be recommended, and the matching degree threshold is adjusted by adopting the online flow data of the information to be recommended. In an embodiment, the alternative determining module 620 may be configured to perform the operation S220 described above, which is not described herein.
According to an embodiment of the present disclosure, the information recommendation apparatus 600 may further include a conversion rate determination module and a first threshold adjustment module. The conversion rate determining module is used for determining the online conversion rate of the information to be recommended according to the online flow rate data in response to the online flow rate data of the information to be recommended. The first threshold adjustment module is used for adjusting the matching degree threshold according to the difference between the target conversion rate and the on-line conversion rate.
According to an embodiment of the present disclosure, the information recommendation apparatus 600 may further include a second match determination module and a threshold determination module. And the second matching determining module is used for determining a second matching degree between the offline flow data and the recommended object in response to the acquired offline flow data of the information to be recommended. The offline flow data comprises object information of a recommended object and operation information of the recommended object to the information to be recommended. The threshold determining module is used for determining the minimum matching degree which enables the actual conversion rate of the information to be recommended to reach the target conversion rate according to the second matching degree and the operation information, and the minimum matching degree is used as a matching degree threshold.
According to an embodiment of the present disclosure, the first threshold adjustment module may include a step determination sub-module and an adjustment sub-module. The step length determining submodule is used for determining an adjustment step length aiming at the matching degree threshold according to the difference. The adjustment submodule is used for adjusting the matching degree threshold according to the adjustment step length.
According to an embodiment of the present disclosure, the step size determination submodule may include a base determination unit, an exposure amount determination unit, a weight determination unit, and a step size determination unit. The base determination unit is used for determining the base for the adjustment step according to the difference. The exposure determining unit is used for determining the real-time exposure of the information to be recommended in the unit time period according to the online flow data. The weight determining unit is used for determining the weight for adjusting the step length according to the real-time exposure amount and the preset exposure amount. The step length determining unit is used for determining an adjustment step length according to the base number and the weight.
According to an embodiment of the present disclosure, the weight determining unit may include a first determining subunit and a second determining subunit. The first determining subunit is configured to determine, as a weight, a value of a first ratio of a ratio between the real-time exposure amount and the predetermined exposure amount, where the base number is equal to or greater than zero. The second determining subunit is configured to determine, as the weight, a value of a second ratio of the real-time exposure amount to the predetermined exposure amount in a case where the base number is smaller than zero. Wherein the second ratio is less than the first ratio.
According to an embodiment of the present disclosure, the offline traffic data includes a plurality of traffic data. The threshold determination module may include a data dividing sub-module, a target determination sub-module, and a threshold determination sub-module. The data dividing sub-module is used for dividing the plurality of flow data into n data groups which are ordered from large to small according to the matching degree according to the second matching degree between the plurality of flow data and the recommended object; each of the n data sets corresponds to a matching degree interval. The target determination submodule is used for determining the ordering positions p of the target data groups of the n data groups; the actual conversion rate determined according to the operation information included in the 1 st to p-th data sets in the m data sets is greater than or equal to the target conversion rate, and the actual conversion rate determined according to the operation information included in the 1 st to (p+1) -th data sets in the m data sets is less than the target conversion rate. The threshold value determining submodule is used for determining a matching degree threshold value according to the lower limit value of the matching degree interval corresponding to the target data set.
According to an embodiment of the present disclosure, the above-mentioned threshold determining submodule may be configured to determine that a sum of the lower limit value and the predetermined value is a matching degree threshold. The information recommendation device 600 may further include a second threshold adjustment module configured to adjust the matching degree threshold to a predetermined threshold in response to determining that the conversion number for the information to be recommended is greater than a predetermined conversion number according to the acquired online traffic data. Wherein the predetermined value is a value greater than zero and the predetermined threshold is less than the pre-adjustment matching degree threshold.
According to an embodiment of the present disclosure, the first degree of matching and the second degree of matching are determined using a degree of matching estimation model. The information recommendation apparatus 600 may further include a first acquisition determining module configured to determine that the offline traffic data is acquired in response to at least one of the following conditions being satisfied: the quantity of target flow data in the flow data accumulated under the line of the information to be recommended is larger than or equal to a first threshold value, and the operation information included in the target flow data indicates that the information to be recommended is converted; when the flow data accumulated offline are used as test data of the matching degree prediction model, the sorting index of the matching degree prediction model is larger than or equal to an index threshold value.
According to an embodiment of the present disclosure, the information recommendation apparatus 600 may further include a second acquisition determining module configured to determine that the online traffic data is acquired in response to the number of the online accumulated traffic data of the information to be recommended being greater than a second threshold.
According to an embodiment of the present disclosure, the second threshold is positively related to the conversion complexity of the information to be recommended.
It should be noted that, in the technical solution of the present disclosure, the related processes of collecting, storing, using, processing, transmitting, providing, disclosing, etc. of the personal information of the user all conform to the rules of the related laws and regulations, and do not violate the popular regulations.
According to embodiments of the present disclosure, the present disclosure also provides an electronic device, a readable storage medium and a computer program product.
FIG. 7 illustrates a schematic block diagram of an example electronic device 700 that may be used to implement the information recommendation methods of embodiments of the present disclosure. Electronic devices are intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The electronic device may also represent various forms of mobile devices, such as personal digital processing, cellular telephones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions, are meant to be exemplary only, and are not meant to limit implementations of the disclosure described and/or claimed herein.
As shown in fig. 7, the apparatus 700 includes a computing unit 701 that can perform various appropriate actions and processes according to a computer program stored in a Read Only Memory (ROM) 702 or a computer program loaded from a storage unit 708 into a Random Access Memory (RAM) 703. In the RAM 703, various programs and data required for the operation of the device 700 may also be stored. The computing unit 701, the ROM 702, and the RAM 703 are connected to each other through a bus 704. An input/output (I/O) interface 705 is also connected to bus 704.
Various components in device 700 are connected to I/O interface 705, including: an input unit 706 such as a keyboard, a mouse, etc.; an output unit 707 such as various types of displays, speakers, and the like; a storage unit 708 such as a magnetic disk, an optical disk, or the like; and a communication unit 709 such as a network card, modem, wireless communication transceiver, etc. The communication unit 709 allows the device 700 to exchange information/data with other devices via a computer network, such as the internet, and/or various telecommunication networks.
The computing unit 701 may be a variety of general and/or special purpose processing components having processing and computing capabilities. Some examples of computing unit 701 include, but are not limited to, a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), various specialized Artificial Intelligence (AI) computing chips, various computing units running machine learning model algorithms, a Digital Signal Processor (DSP), and any suitable processor, controller, microcontroller, etc. The calculation unit 701 performs the respective methods and processes described above, such as an information recommendation method. For example, in some embodiments, the information recommendation method may be implemented as a computer software program tangibly embodied on a machine-readable medium, such as storage unit 708. In some embodiments, part or all of the computer program may be loaded and/or installed onto device 700 via ROM 702 and/or communication unit 709. When the computer program is loaded into the RAM 703 and executed by the computing unit 701, one or more steps of the information recommendation method described above may be performed. Alternatively, in other embodiments, the computing unit 701 may be configured to perform the information recommendation method by any other suitable means (e.g. by means of firmware).
Various implementations of the systems and techniques described here above may be implemented in digital electronic circuitry, integrated circuit systems, field Programmable Gate Arrays (FPGAs), application Specific Integrated Circuits (ASICs), application Specific Standard Products (ASSPs), systems On Chip (SOCs), complex Programmable Logic Devices (CPLDs), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include: implemented in one or more computer programs, the one or more computer programs may be executed and/or interpreted on a programmable system including at least one programmable processor, which may be a special purpose or general-purpose programmable processor, that may receive data and instructions from, and transmit data and instructions to, a storage system, at least one input device, and at least one output device.
Program code for carrying out methods of the present disclosure may be written in any combination of one or more programming languages. These program code may be provided to a processor or controller of a general purpose computer, special purpose computer, or other programmable data processing apparatus such that the program code, when executed by the processor or controller, causes the functions/operations specified in the flowchart and/or block diagram to be implemented. The program code may execute entirely on the machine, partly on the machine, as a stand-alone software package, partly on the machine and partly on a remote machine or entirely on the remote machine or server.
In the context of this disclosure, a machine-readable medium may be a tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. The machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
To provide for interaction with a user, the systems and techniques described here can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to a user; and a keyboard and pointing device (e.g., a mouse or trackball) by which a user can provide input to the computer. Other kinds of devices may also be used to provide for interaction with a user; for example, feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form, including acoustic input, speech input, or tactile input.
The systems and techniques described here can be implemented in a computing system that includes a background component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front-end component (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such background, middleware, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: local Area Networks (LANs), wide Area Networks (WANs), and the internet.
The computer system may include a client and a server. The client and server are typically remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. The server can be a cloud server, also called a cloud computing server or a cloud host, and is a host product in a cloud computing service system, so as to solve the defects of high management difficulty and weak service expansibility in the traditional physical hosts and VPS service ("Virtual Private Server" or simply "VPS"). The server may also be a server of a distributed system or a server that incorporates a blockchain.
It should be appreciated that various forms of the flows shown above may be used to reorder, add, or delete steps. For example, the steps recited in the present disclosure may be performed in parallel or sequentially or in a different order, provided that the desired results of the technical solutions of the present disclosure are achieved, and are not limited herein.
The above detailed description should not be taken as limiting the scope of the present disclosure. It will be apparent to those skilled in the art that various modifications, combinations, sub-combinations and alternatives are possible, depending on design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present disclosure are intended to be included within the scope of the present disclosure.

Claims (18)

1. An information recommendation method, comprising:
in response to obtaining object information of a target object, determining a first matching degree between information to be recommended and the target object according to the object information; and
determining that the information to be recommended is the candidate information for the target object in response to the first matching degree being greater than or equal to a matching degree threshold for the information to be recommended,
the initial value of the matching degree threshold is determined according to the offline flow data of the information to be recommended, and the matching degree threshold is adjusted by adopting the online flow data of the information to be recommended; the method further comprises the steps of:
In response to acquiring the online flow data of the information to be recommended, determining the online conversion rate of the information to be recommended according to the online flow data; and
adjusting the matching degree threshold according to the difference between the target conversion rate and the on-line conversion rate;
wherein said adjusting said match threshold according to the difference between the target conversion and said on-line conversion comprises:
according to the difference, determining an adjustment step length for the matching degree threshold; and
adjusting the matching degree threshold according to the adjustment step length;
wherein, according to the difference, determining the adjustment step length for the matching degree threshold value includes:
determining a base for the adjustment step according to the difference;
determining the real-time exposure of the information to be recommended in a unit time period according to the online flow data;
determining the weight for the adjustment step according to the real-time exposure and the preset exposure; and
and determining the adjustment step length according to the base number and the weight.
2. The method of claim 1, further comprising:
determining a second matching degree between the offline flow data and the recommended object in response to acquiring the offline flow data of the information to be recommended; the offline flow data comprise object information of the recommended object and operation information of the recommended object on the information to be recommended; and
And determining the minimum matching degree which enables the actual conversion rate of the information to be recommended to reach the target conversion rate according to the second matching degree and the operation information, and taking the minimum matching degree as the matching degree threshold.
3. The method of claim 1, wherein the determining weights for the adjustment step from the real-time exposure and a predetermined exposure comprises:
determining a value of a first ratio of the real-time exposure to the predetermined exposure as the weight when the base number is greater than or equal to zero; and
determining a value of a second ratio of the real-time exposure to the predetermined exposure as the weight in the case that the cardinality is smaller than zero,
wherein the second ratio is less than the first ratio.
4. The method of claim 2, wherein the offline traffic data comprises a plurality of traffic data; the determining, according to the second matching degree and the operation information, a minimum matching degree that enables the actual conversion rate for the information to be recommended to reach a target conversion rate includes:
dividing the flow data into n data groups which are ordered from big to small according to the matching degree according to the second matching degree between the flow data and the recommended object; each of the n data sets corresponds to a matching degree interval;
Determining target data sets of the n data sets and sorting positions p of the target data sets; the actual conversion rate determined according to the operation information included in the 1 st data group to the p-th data group in the n data groups is larger than or equal to the target conversion rate, and the actual conversion rate determined according to the operation information included in the 1 st data group to the (p+1) th data group in the n data groups is smaller than the target conversion rate; and
and determining the matching degree threshold according to the lower limit value of the matching degree interval corresponding to the target data set.
5. The method according to claim 4, wherein:
determining the matching degree threshold according to the lower limit value of the matching degree interval corresponding to the target data set comprises the following steps: determining the sum of the lower limit value and a preset value as the matching degree threshold value;
the method further comprises the steps of: in response to determining that the conversion number for the information to be recommended is greater than a predetermined conversion number based on the acquired online traffic data, adjusting the matching degree threshold to a predetermined threshold,
wherein the predetermined value is a value greater than zero and the predetermined threshold is less than the pre-adjustment matching degree threshold.
6. The method of claim 2, wherein the first and second matches are determined using a match prediction model; the method further comprises the steps of: determining that the offline flow data is acquired in response to at least one of the following conditions being met:
The quantity of target flow data in the flow data accumulated offline of the information to be recommended is greater than or equal to a first threshold value, and the operation information included in the target flow data indicates that the information to be recommended is converted;
when the flow data accumulated offline are used as the test data of the matching degree estimation model, the sorting index of the matching degree estimation model is larger than or equal to an index threshold.
7. The method of claim 1, further comprising:
and determining that the online flow data is acquired in response to the quantity of the online accumulated flow data of the information to be recommended being greater than a second threshold.
8. The method of claim 7, wherein the second threshold is positively correlated with a conversion complexity of the information to be recommended.
9. An information recommendation apparatus, comprising:
the first matching determining module is used for determining a first matching degree between information to be recommended and the target object according to the object information in response to the acquired object information of the target object; and
an alternative determining module, configured to determine, in response to the first matching degree being equal to or greater than a matching degree threshold for the information to be recommended, that the information to be recommended is alternative information for the target object,
The initial value of the matching degree threshold is determined according to the offline flow data of the information to be recommended, and the matching degree threshold is adjusted by adopting the online flow data of the information to be recommended; the apparatus further comprises:
the conversion rate determining module is used for determining the online conversion rate of the information to be recommended according to the online flow rate data in response to the online flow rate data of the information to be recommended; and
the first threshold adjustment module is used for adjusting the matching degree threshold according to the difference between the target conversion rate and the online conversion rate;
wherein the first threshold adjustment module includes:
a step length determining submodule, configured to determine an adjustment step length for the matching degree threshold according to the difference; and
the adjustment sub-module is used for adjusting the matching degree threshold according to the adjustment step length;
wherein the step determination submodule includes:
a base determining unit configured to determine a base for the adjustment step according to the difference;
the exposure determining unit is used for determining the real-time exposure of the information to be recommended in a unit period according to the online flow data;
A weight determining unit for determining a weight for the adjustment step according to the real-time exposure amount and a predetermined exposure amount; and
and the step length determining unit is used for determining the adjustment step length according to the base number and the weight.
10. The apparatus of claim 9, further comprising:
the second matching determining module is used for determining a second matching degree between the offline flow data and the recommended object in response to the offline flow data of the information to be recommended; the offline flow data comprise object information of the recommended object and operation information of the recommended object on the information to be recommended; and
and the threshold value determining module is used for determining the minimum matching degree which enables the actual conversion rate of the information to be recommended to reach the target conversion rate according to the second matching degree and the operation information, and taking the minimum matching degree as the matching degree threshold value.
11. The apparatus of claim 9, wherein the weight determination unit comprises:
a first determining subunit, configured to determine, as the weight, a value of a first ratio of the real-time exposure to the predetermined exposure when the base number is greater than or equal to zero; and
A second determination subunit configured to determine, as the weight, a value of a second ratio of the real-time exposure amount to the predetermined exposure amount in a case where the base number is smaller than zero,
wherein the second ratio is less than the first ratio.
12. The apparatus of claim 10, wherein the offline traffic data comprises a plurality of traffic data; the threshold determination module includes:
the data dividing sub-module is used for dividing the plurality of flow data into n data groups which are ordered from large to small according to the matching degree according to the second matching degree between the plurality of flow data and the recommended object; each of the n data sets corresponds to a matching degree interval;
the target determining submodule is used for determining target data sets of the n data sets and sorting positions p of the target data sets; the actual conversion rate determined according to the operation information included in the 1 st data group to the p-th data group in the n data groups is larger than or equal to the target conversion rate, and the actual conversion rate determined according to the operation information included in the 1 st data group to the (p+1) th data group in the n data groups is smaller than the target conversion rate; and
And the threshold value determining submodule is used for determining the matching degree threshold value according to the lower limit value of the matching degree interval corresponding to the target data set.
13. The apparatus of claim 12, wherein:
the threshold determination submodule is used for: determining the sum of the lower limit value and a preset value as the matching degree threshold value;
the apparatus further includes a second threshold adjustment module for adjusting the matching degree threshold to a predetermined threshold in response to determining from the acquired on-line traffic data that the conversion number for the information to be recommended is greater than a predetermined conversion number,
wherein the predetermined value is a value greater than zero and the predetermined threshold is less than the pre-adjustment matching degree threshold.
14. The apparatus of claim 10, wherein the first and second matches are determined using a match prediction model; the apparatus further includes a first acquisition determination module to determine that the offline flow data is acquired in response to at least one of the following conditions being met:
the quantity of target flow data in the flow data accumulated offline of the information to be recommended is greater than or equal to a first threshold value, and the operation information included in the target flow data indicates that the information to be recommended is converted;
When the flow data accumulated offline are used as the test data of the matching degree estimation model, the sorting index of the matching degree estimation model is larger than or equal to an index threshold.
15. The apparatus of claim 9, further comprising:
and the second acquisition determining module is used for determining that the online flow data is acquired in response to the quantity of the online accumulated flow data of the information to be recommended being greater than a second threshold value.
16. The apparatus of claim 15, wherein the second threshold is positively correlated with a conversion complexity of the information to be recommended.
17. An electronic device, comprising:
at least one processor; and
a memory communicatively coupled to the at least one processor; wherein, the liquid crystal display device comprises a liquid crystal display device,
the memory stores instructions for execution by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-8.
18. A non-transitory computer readable storage medium storing computer instructions for causing a computer to perform the method of any one of claims 1-8.
CN202210063127.2A 2022-01-19 2022-01-19 Information recommendation method, device, electronic equipment and storage medium Active CN114398558B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210063127.2A CN114398558B (en) 2022-01-19 2022-01-19 Information recommendation method, device, electronic equipment and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210063127.2A CN114398558B (en) 2022-01-19 2022-01-19 Information recommendation method, device, electronic equipment and storage medium

Publications (2)

Publication Number Publication Date
CN114398558A CN114398558A (en) 2022-04-26
CN114398558B true CN114398558B (en) 2023-05-23

Family

ID=81231781

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210063127.2A Active CN114398558B (en) 2022-01-19 2022-01-19 Information recommendation method, device, electronic equipment and storage medium

Country Status (1)

Country Link
CN (1) CN114398558B (en)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2017071251A1 (en) * 2015-10-28 2017-05-04 百度在线网络技术（北京）有限公司 Information pushing method and device
CN108197982A (en) * 2017-12-28 2018-06-22 北京奇艺世纪科技有限公司 The correlation analysis and device of flow and multimedia object
CN109871491A (en) * 2019-03-20 2019-06-11 江苏满运软件科技有限公司 Forum postings recommended method, system, equipment and storage medium
CN111611470A (en) * 2019-02-22 2020-09-01 北京搜狗科技发展有限公司 Data processing method and device and electronic equipment
CN112287657A (en) * 2020-11-19 2021-01-29 每日互动股份有限公司 Information matching system based on text similarity

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9785883B2 (en) * 2012-04-27 2017-10-10 Excalibur Ip, Llc Avatars for use with personalized generalized content recommendations
CN106658434A (en) * 2015-10-30 2017-05-10 中兴通讯股份有限公司 Business package recommending method and device
CN111737586B (en) * 2020-08-19 2020-12-04 腾讯科技（深圳）有限公司 Information recommendation method, device, equipment and computer readable storage medium
CN113220866B (en) * 2021-04-28 2023-01-06 西安电子科技大学 Recommendation method based on user social network and product similarity
CN113360761A (en) * 2021-06-11 2021-09-07 北京百度网讯科技有限公司 Information flow recommendation method and device, electronic equipment and computer-readable storage medium

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2017071251A1 (en) * 2015-10-28 2017-05-04 百度在线网络技术（北京）有限公司 Information pushing method and device
CN108197982A (en) * 2017-12-28 2018-06-22 北京奇艺世纪科技有限公司 The correlation analysis and device of flow and multimedia object
CN111611470A (en) * 2019-02-22 2020-09-01 北京搜狗科技发展有限公司 Data processing method and device and electronic equipment
CN109871491A (en) * 2019-03-20 2019-06-11 江苏满运软件科技有限公司 Forum postings recommended method, system, equipment and storage medium
CN112287657A (en) * 2020-11-19 2021-01-29 每日互动股份有限公司 Information matching system based on text similarity

Also Published As

Publication number Publication date
CN114398558A (en) 2022-04-26

Similar Documents

Publication Publication Date Title
CN113065614B (en) Training method of classification model and method for classifying target object
CN112527598B (en) Method, apparatus, device, storage medium and program product for monitoring data
CN114500339B (en) Node bandwidth monitoring method and device, electronic equipment and storage medium
CN114398558B (en) Information recommendation method, device, electronic equipment and storage medium
CN113900731B (en) Request processing method, device, equipment and storage medium
CN113326436B (en) Method, device, electronic equipment and storage medium for determining recommended resources
CN116012859A (en) Text image rejection judgment method, device and equipment based on definition index
CN112861034B (en) Method, device, equipment and storage medium for detecting information
CN113761379B (en) Commodity recommendation method and device, electronic equipment and medium
CN114066278B (en) Method, apparatus, medium, and program product for evaluating article recall
CN112800315B (en) Data processing method, device, equipment and storage medium
CN117611324A (en) Credit rating method, apparatus, electronic device and storage medium
CN115017145A (en) Data expansion method, device and storage medium
CN117608944A (en) Method and device for calculating weight migration volume ratio, electronic equipment and storage medium
CN116228301A (en) Method, device, equipment and medium for determining target user
CN116992150A (en) Research and development component recommendation method, device, equipment and storage medium
CN117574186A (en) Session feature recognition method, device, equipment and storage medium
CN116486144A (en) Method for generating noise label detector, method and device for detecting noise label
CN115952874A (en) Federal learning method, device, server, electronic equipment and readable storage medium
CN117195104A (en) Resource classification method, device, electronic equipment and storage medium
CN115758142A (en) Deep learning model training method, data processing method and device
CN117271882A (en) Multi-target fusion processing method and device, electronic equipment and storage medium
CN116342099A (en) Operation and maintenance work assisting method, device, equipment, medium and program product
CN117851599A (en) Method, device, equipment and medium for extracting text of other elements of investment supervision
CN117495525A (en) Data processing method, apparatus, device, storage medium, and program product

Legal Events

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