CN116684481B - Method and device for processing push information homogenization, electronic equipment and storage medium - Google Patents
Method and device for processing push information homogenization, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN116684481B CN116684481B CN202310959221.0A CN202310959221A CN116684481B CN 116684481 B CN116684481 B CN 116684481B CN 202310959221 A CN202310959221 A CN 202310959221A CN 116684481 B CN116684481 B CN 116684481B
- Authority
- CN
- China
- Prior art keywords
- information
- push
- homogenization
- type
- user
- 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
Links
- 238000000265 homogenisation Methods 0.000 title claims abstract description 125
- 238000000034 method Methods 0.000 title claims abstract description 63
- 238000012545 processing Methods 0.000 title claims description 9
- 230000008569 process Effects 0.000 claims description 31
- 238000004891 communication Methods 0.000 claims description 14
- 238000012549 training Methods 0.000 claims description 13
- 238000004590 computer program Methods 0.000 claims description 7
- 238000003672 processing method Methods 0.000 abstract description 7
- 230000000694 effects Effects 0.000 abstract description 6
- 230000006399 behavior Effects 0.000 description 5
- 230000009471 action Effects 0.000 description 4
- 230000003139 buffering effect Effects 0.000 description 4
- 238000004422 calculation algorithm Methods 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 4
- 238000007637 random forest analysis Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 230000004044 response Effects 0.000 description 2
- 238000003066 decision tree Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D30/00—Reducing energy consumption in communication networks
- Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks
Landscapes
- User Interface Of Digital Computer (AREA)
- Information Transfer Between Computers (AREA)
Abstract
The application relates to a processing method, a device, electronic equipment and a storage medium for homogenizing push information, wherein the method judges whether a first push information set generated based on current user information belongs to a homogenization type or not by acquiring the current user information; and determining a pushing cause type of the first pushing information set under the condition that the first pushing information set belongs to the homogeneous type, so as to determine an adjustment strategy based on the pushing cause type, and adjusting pushing information in the first pushing information set based on the adjustment strategy to generate a target pushing information set. Therefore, whether the homogenization problem exists at present is judged through whether the first push information set belongs to the homogenization type, and under the condition that the homogenization problem exists, the target push information set can be determined by utilizing the adjustment strategy to adjust the push information, so that the homogenization problem of the first push information is overcome, and the effect of providing various push contents for users is realized.
Description
Technical Field
The present application relates to the field of computers, and in particular, to a method and apparatus for homogenizing push information, an electronic device, and a storage medium.
Background
Along with the development of the Internet, the personalized pushing mode is the core of the current information stream pushing product, and the personalized pushing mode mainly helps users to screen out interested contents from mass data through user information.
The personalized pushing mode can be used in various scenes, such as a shopping platform, a short video platform and the like. In the process of generating push information in a personalized push mode, a push strategy is mainly determined according to data such as click rate, consumption time length and viewing time length in user information, push information which is more in line with the user information is gradually generated, however, at the moment, the content of the push information is more and more single, and the situation is called that the push information has a homogenization problem.
Disclosure of Invention
In order to solve the technical problems, the application provides a processing method, a device, electronic equipment and a storage medium for homogenizing push information.
In a first aspect, the present application provides a processing method for homogenizing push information, where the method includes:
acquiring current user information;
judging whether a first pushing information set generated based on the current user information belongs to a homogenization type or not; the homogenization type represents that the proportion of push information belonging to a target class in the first push information set is larger than a preset threshold;
determining a pushing cause type of the first pushing information set under the condition that the first pushing information set belongs to the homogenization type;
determining an adjustment strategy based on the push cause type, wherein the adjustment strategy represents a strategy for adjusting push information in the first push information set;
and adjusting the push information in the first push information set based on the adjustment strategy to generate a target push information set.
Optionally, the determining whether the first push information set generated based on the current user information is of a homogenization type includes:
determining a first set of user features based on the current user information, the first set of user features representing a set of features determined based on the current user information;
and inputting the first user characteristic set into a preset prediction model to judge whether the first pushing information set belongs to the homogenization type.
Optionally, the training process of the prediction model is as follows:
acquiring first user information and determining a second pushing information set based on the first user information;
determining second homogenization information of the second push information set, wherein the second homogenization information represents whether the second push information set belongs to the homogenization type;
determining a second set of user features based on the first user information, the second set of user features representing a set of features determined based on the first user information;
and training to obtain the prediction model by taking the second user characteristic set as input data and the second homogeneous information as expected output data.
Optionally, the inputting the first user feature set into a preset prediction model to determine whether the first push information set belongs to the homogenization type includes:
inputting the first user feature set into a preset prediction model;
determining first homogenization information based on an output of the prediction model, the first homogenization information representing whether the first push information set is of the homogenization type;
based on the first homogenization information, whether the first push information set belongs to the homogenization type is judged.
Optionally, the inputting the first set of user features into a preset prediction model further includes:
determining a feature ranking table based on the output of the prediction model, wherein the feature ranking table represents a ranking table of the importance of features in the first user feature set from high to low;
determining the features with importance degrees lower than a preset importance degree in the feature ordering table as first target feature information;
the features represented by the first target feature information are removed from the first set of user features.
Optionally, the determining, in a case that the first push information set belongs to the homogenization type, a push cause type of the first push information set includes:
determining type information of the target category under the condition that the first pushing information set belongs to the homogenization type;
a push cause type of the first set of push information is determined based on the type information.
Optionally, the determining whether the first push information set generated based on the current user information is of a homogenization type includes:
determining target user information after a preset time under the condition that a first pushing information set generated based on the current user information does not belong to the homogenization type;
judging whether the generated third pushing information set belongs to the homogenization type or not based on the target user information;
and outputting early warning information under the condition that the third pushing information set belongs to the homogenization type.
In a third aspect, a processing apparatus for homogenizing push information is provided, where the apparatus includes:
the acquisition module is used for acquiring current user information;
the first judging module is used for judging whether a first pushing information set generated based on the current user information belongs to a homogenization type or not; the homogenization type represents that the proportion of push information belonging to a target class in the first push information set is larger than a preset threshold;
the first determining module is used for determining a pushing cause type of the first pushing information set under the condition that the first pushing information set belongs to the homogenization type;
a second determining module, configured to determine an adjustment policy based on the push cause type, where the adjustment policy represents a policy for adjusting push information in the first set of push information;
and the generation module is used for adjusting the push information in the first push information set based on the adjustment strategy so as to generate a target push information set.
In a third aspect, an electronic device is provided, including a processor, a communication interface, a memory, and a communication bus, where the processor, the communication interface, and the memory complete communication with each other through the communication bus;
a memory for storing a computer program;
a processor, configured to implement the method according to any one of the embodiments of the first aspect when executing a program stored on a memory.
In a fourth aspect, a computer-readable storage medium is provided, on which a computer program is stored which, when being executed by a processor, implements a method according to any one of the embodiments of the first aspect.
Compared with the prior art, the technical scheme provided by the embodiment of the application has the following advantages:
according to the method provided by the embodiment of the application, whether the first pushing information set generated based on the current user information belongs to the homogenization type is judged by acquiring the current user information; the method comprises the steps that the proportion of push information belonging to a target category in a first push information set is larger than a preset threshold, and the push cause type of the first push information set is determined under the condition that the first push information set belongs to the homogeneous type, so that an adjustment strategy is determined based on the push cause type, the adjustment strategy represents a strategy for adjusting the push information in the first push information set, and the push information in the first push information set is adjusted based on the adjustment strategy to generate the target push information set. Therefore, whether the homogenization problem exists at present is judged through whether the first push information set belongs to the homogenization type, and under the condition that the homogenization problem exists, the target push information set can be determined by utilizing the adjustment strategy to adjust the push information, so that the homogenization problem of the first push information is overcome, and the effect of providing various push contents for users is realized.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the application and together with the description, serve to explain the principles of the application.
In order to more clearly illustrate the embodiments of the application or the technical solutions of the prior art, the drawings which are used in the description of the embodiments or the prior art will be briefly described, and it will be obvious to a person skilled in the art that other drawings can be obtained from these drawings without inventive effort.
Fig. 1 is a schematic flow chart of a processing method for homogenizing push information according to an embodiment of the present application;
fig. 2 is a schematic structural diagram of a processing device for homogenizing push information according to an embodiment of the present application;
fig. 3 is a schematic structural diagram of an electronic device according to an embodiment of the present application.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present application more apparent, the technical solutions of the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application, and it is apparent that the described embodiments are some embodiments of the present application, but not all embodiments of the present application. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.
In the existing push information generation process, more emphasis is paid to whether the push information accords with the browsing habit or the using habit of the user, namely, the push information aiming at the user is expected to be generated, but if the push information aiming at the user is generated only according to the browsing habit or the using habit of the user, the content of the push information is more and more single, and the user can only receive the too single push information along with the increase of time, and the situation is called that the push information has the problem of homogeneity; at this time, diversified push information cannot be provided for the user, aesthetic fatigue is likely to occur to the user, and the problem that the use rate of the user to the client is low is further caused. That is, although a single push message can provide targeted content for users, there is a homogenization problem and the users cannot be guided to view various content, resulting in a problem of single experience of the users.
Fig. 1 is a flow chart of a processing method for homogenizing push information according to an embodiment of the present application.
As shown in fig. 1, the present application discloses an embodiment, and provides a method for processing push information homogenization, where the method includes:
s110: and acquiring current user information.
The processing method for homogenizing the push information in the embodiment can be applied to a server, and the current user information can comprise information such as user identification, time stamp and the like; the current user information may also include historical information determined by user identification and time stamping, and the historical information may include browsing records, click records, browsing durations, historical push information, and the like.
In general, the process of generating push information is performed based on user information only, and this method has the problem that the generation of push content is too single.
S120: judging whether a first pushing information set generated based on current user information belongs to a homogenization type or not; the homogeneity type indicates whether the proportion of push information belonging to the target class in the first push information set is greater than a preset threshold.
In this embodiment, after current user information is obtained, whether a first push information set generated based on the current user information belongs to a homogenization type is determined, wherein whether the first push information set belongs to the homogenization type, and indicates whether the proportion of push information belonging to a target category in the first push information set is greater than a preset threshold value, and if so, the first push information set belongs to the homogenization type, and if not, the first push information set does not belong to the homogenization type, and the homogenization type indicates that the push information content in the first push information set is too single and has a homogenization problem, and if not, the first push information set does not have the homogenization problem; the first push information set represents a set of push information generated based on current user information, and the target category represents a category of push information with the largest proportion in the first push information set, and each push information has a predetermined category, such as a fun category, a life category, an entertainment star category and the like. Therefore, when the proportion of the pushing information of the target category is greater than the preset threshold value, the current pushing information in the first pushing information set is single and does not have diversified pushing information, and at the moment, the first pushing information set belongs to a homogeneous type and the pushing information in the first pushing information set needs to be adjusted; and under the condition that the proportion of the push information of the target category is not greater than a preset threshold value, the push information in the current first push information set is not single in homogenization problem, at the moment, the first push information set is not of homogenization type, and the push information in the first push information set can be temporarily not adjusted. It should be noted that the preset threshold may be determined according to requirements, for example, 40%, 50%, etc. By judging whether the proportion of the push information of the target category is larger than a preset threshold value, the method plays a role in judging whether the push information in the first push information set has the homogenization problem that the content is single and the content is not diversified.
In the process of judging whether the first pushing information set generated by the current user information belongs to the homogenization type, a neural network model, a decision tree algorithm, a random forest algorithm and other modes can be adopted.
In an embodiment, the step of S120 of determining whether the first push information set generated based on the current user information is of a homogenization type may include:
determining a first set of user features based on the current user information, the first set of user features representing a set of features determined based on the current user information;
and inputting the first user characteristic set into a preset prediction model to judge whether the first pushing information set belongs to a homogenization type.
In this embodiment, taking a model set up by a random forest algorithm as an example for explanation, after current user information is acquired, a first user feature set is determined firstly based on the current user information, where the first user feature set represents a set of features that can be determined based on the current user information, the current user information may include a user identifier, a timestamp, a day of the week, day of the week information, weekend information, holiday information, a time from last access, and the like, and features in the first user feature set may include history information of the user identifier before the timestamp, such as a browsing record, a click record, a browsing duration, a history push information, an average click time interval, an average browsing time interval, and the like. Then, the first user feature set is input to a preset prediction model to determine whether the first push information set belongs to a homogeneous type, wherein the preset prediction model is a model built by a random forest algorithm which is trained in advance, the output of the prediction model is information indicating whether the first push information set belongs to the homogeneous type, for example, the output is 0 or 1,0 indicates that the first push information set does not belong to the homogeneous type, and 1 indicates that the first push information set belongs to the homogeneous type.
It should be noted that, in this embodiment, the user information may include a time stamp or a time period, where the time stamp indicates a specific time point, and the time period indicates a time period when a preset time length passes by taking the current time point as a starting time point, in addition, the step of determining whether the first push information set generated based on the current user information belongs to a homogeneous type may be performed within a time window range, where the time window range indicates a time segment including a plurality of continuous behaviors, and the time interval between two behaviors is continuously indicated as not exceeding a buffering duration, and when exceeding the buffering duration, the buffering is released, the behavior at this time is determined as a new time window, for example, the buffering duration is 30 minutes, the time window range indicates that the time interval between two behaviors is not exceeding 30 minutes, and the behavior exceeding 30 minutes is determined as a new time window; the buffer time length is a parameter in the buffer mechanism, and the buffer mechanism of the client enables the client to temporarily store the data for the time of the buffer time length, for example, after the client receives the first push information set sent by the server, the first push information set can store the time of the buffer time length in a buffer form in the client. The process of displaying the recommended information in the first push information set in the cache duration includes that when 100 push information is included in the first push information set, the client displays 10 push information each time, the remaining 90 push information is not displayed, and when the client refreshes the push information in the cache duration, 10 push information is reselected from the 90 push information not displayed to be displayed. Thus, an interval duration exceeding the cache duration will release the cached information, considered as the start of a new time window.
In one example, the features in the first push information set may include the following feature contents of table 1:
TABLE 1
Wherein the history timestamp represents a history timestamp before the timestamp of the current user information, the feature names represent names of the respective features, and the feature description represents an explanation of the features.
In one embodiment, the training process of the predictive model is as follows:
acquiring first user information and determining a second pushing information set based on the first user information;
determining second homogenization information of a second push information set, wherein the second homogenization information represents whether the second push information set belongs to the homogenization type or not;
determining a second set of user features based on the first user information, the second set of user features representing a set of features determined based on the first user information;
and taking the second user characteristic set as input data, taking the second homogenization information as expected output data, and training to obtain a prediction model.
In this embodiment, the training of the prediction model includes a training data acquisition process and a training process, where the training data acquisition process may be to acquire first user information, and determine a second push information set based on the first user information, where the first user information represents historical user information, and the second push information set represents a second push information set generated according to the first user information; next, a second set of user features is determined based on the first user information, the second set of user features representing a set of features determined based on the first user information. The training process may be to train to obtain a prediction model with the second user feature set as input data and the second homogenized information as desired output data.
In an embodiment, inputting the first user feature set into a preset prediction model to determine whether the first push information set is of a homogenization type, including:
inputting the first user feature set into a preset prediction model;
determining first homogenization information based on the output of the prediction model, the first homogenization information representing whether the first push information set belongs to a homogenization type;
based on the first homogenization information, whether the first push information set belongs to a homogenization type is judged.
In this embodiment, after the first user feature set is input to the preset prediction model, the prediction model outputs first homogenization information based on the first user feature set, where the first homogenization information indicates whether the first push information set belongs to a homogenization type, and then determines whether the first push information set belongs to a homogenization type based on the first homogenization information, for example, the first homogenization information may be 0 or 1, where the first homogenization information indicates that the first push information set does not belong to a homogenization type when the first homogenization information is 0, and indicates that the first push information set does not belong to a homogenization type when the first homogenization information is 1.
In an embodiment, the first set of user features is input to a preset predictive model, and then further comprising:
determining a feature ranking table based on the output of the prediction model, wherein the feature ranking table represents a ranking table of importance of features in the first user feature set from high to low;
determining the features with importance degrees lower than a preset importance degree in the feature ordering table as first target feature information;
the features represented by the first target feature information are removed from the first set of user features.
In this embodiment, since the input of the prediction model is the first user feature set, and the features in the first user feature set may be features configured in advance, some of the features may not be utilized in the process of performing the calculation by the prediction model. In this embodiment, the prediction model may output, in addition to the first homogeneous information, a feature ranking table, where the feature ranking table indicates a ranking table of importance of features in the first user feature set from high to low, that is, the feature ranking table may play a role in determining importance ranking of each feature in the process of predicting; then, determining the features with importance lower than a preset importance in the feature ranking table as first target feature information, namely the features represented by the first target feature information are features with lower importance in the process of performing prediction model operation calculation; the features represented by the first target feature information are then removed from the first set of user features. The method has the advantages that the characteristics with lower importance in the running calculation process of the prediction model are removed, the process of generating the first user characteristic set is simplified, the input data quantity of the prediction model is reduced, and the effect of improving the response efficiency of the prediction model and the effectiveness of the first user characteristic set is achieved.
It should be noted that the preset importance level may be determined according to requirements, for example, the importance level may include four levels of importance levels 0, 1, 2 and 3, the importance level 0 indicates that the importance level is not used in the process of performing the operation calculation of the prediction model, and the importance levels 1, 2 and 3 indicate importance level levels with sequentially increasing use rates. For example, the first user feature set includes features 1-100, and the importance degree of the features 10-20 is determined to be 0 according to the sorting order of the feature sorting table, at this time, importance degree 1 is preset and is lower than importance degree 0 of importance degree 1, so that features 10-20 are determined to be first target feature information, and features 10-20 represented by the first target feature information are removed from the first user feature set, so that the step of determining the first user feature set based on the current user information is performed again, and the feature 10-20 is not required to be determined, which plays a role of simplifying the first user feature set, thereby improving the response efficiency of the prediction model.
S130: and determining the pushing cause type of the first pushing information set under the condition that the first pushing information set belongs to the homogenization type.
In this embodiment, when the first push information set belongs to the homogenization type, it is described that the first push information set directly generated according to the current user information may have a problem of single content homogenization, at this time, the push information in the first push information set needs to be adjusted, and specific adjustment needs to determine the push cause type of the first push information set first, where the push cause type indicates a main generation type of the push information in the first push information set, and since the push information is generated according to the current user information, the specific generation process may be generated according to a user personality, a current event hot spot is generated, a system active push is generated, and thus the push cause types may be a system active push cause type, a user personality push cause type, a hot spot push cause type, and so on. The push information in the first push information set can be adjusted in a targeted manner according to the push cause type.
In an embodiment, in case the first set of push information belongs to a homogeneity type, determining a push cause type of the first set of push information comprises:
determining type information of a target category under the condition that the first pushing information set belongs to a homogeneous type;
a push cause type of the first set of push information is determined based on the type information.
In this embodiment, under the condition that the first push information set belongs to a homogeneous type, type information of a target type is determined, the target type represents a type of push information with a maximum proportion in the first push information set, the type information may represent a generation type of the target type, and the type information may be a system active push cause type, a user individual push cause type, and a hot spot push cause type. In an example, the first push information set includes 10 push information, wherein 6 types of push information are entertainment circle hot spots, the target type is the entertainment circle hot spot, the type information of the entertainment circle hot spot is a current event hot spot, and at this time, based on the type information of the current event hot spot, the push cause type of the first push information set can be determined to be a hot event push cause type.
S140: an adjustment policy is determined based on the push cause type, the adjustment policy representing a policy to adjust push information in the first set of push information.
In this embodiment, the adjustment policy indicates a policy for adjusting the push information in the first push information set, which may be a manner of changing the push information in the first push information set, for example, when the push cause type is a hotspot push cause type, it is indicated that more push information is currently generated according to a current event hotspot, the determined adjustment policy may be a policy for reducing the proportion of current event hotspots in the process of generating push information, when the push cause type is a user personality push cause type, it is indicated that more push information is currently generated according to user individuation, the determined adjustment policy may be a policy for reducing the proportion of user individuation in the process of generating push information or a policy for increasing the proportion of current event hotspots, etc., so as to determine different adjustment policies according to different push cause types, thereby implementing an effect of determining an adjustment policy suitable for the current first push information set according to the push cause type.
S150: and adjusting the push information in the first push information set based on the adjustment strategy to generate a target first push information set.
In this embodiment, the adjustment of the push information in the first push information set based on the adjustment policy is to adjust the process of generating the first push information set, when the push cause type is a hotspot push cause type, the determined adjustment policy is to reduce the proportion of current event hotspots in the process of generating the push information, and at this time, the adjustment reduces the proportion of push information of generating the current event hotspots in the process of generating the first push information set, so that the proportion of other push information is increased, and the generated target first push information set overcomes the problem of single homogeneity of push content; and finally, outputting the target first pushing information set to the client, so that the effect of generating the target first pushing information set with the diversity of the pushing contents is realized, and the using experience of the user is more diversified.
In an embodiment, determining whether the first set of push information generated based on the current user information is of a homogenization type, then comprises:
determining target user information after preset time under the condition that a first pushing information set generated based on current user information does not belong to a homogenization type;
judging whether the generated third pushing information set belongs to a homogenization type or not based on the target user information;
and outputting early warning information under the condition that the third pushing information set belongs to the homogenization type.
In this embodiment, when the first push information set generated based on the current user information is not of a homogeneous type, it is explained that there is no problem of homogeneous content according to the first push information set generated at present, and since the generated push information has an affinity with the timestamp of the current user information, there is a possibility that there is a problem of homogeneous content of the third push information set generated after the timestamp is changed, that is, in the future after the preset time. In order to predict whether the third push information set generated in the future after the preset time has the problem of content single homogenization, target user information after the preset time can be determined, whether the generated third push information set belongs to a homogenization type based on the target user information is judged, wherein the target user information is user information with a time stamp changed in the current user information, and the steps of S120-S150 can be executed again aiming at the target user information, so that the problem of judging whether the third push information set generated after the preset time has the problem of content single homogenization is solved. Under the condition that the third pushing information set belongs to the homogenization type, the problem that the content of the third pushing information set generated after the preset time is single is solved, the early warning information needs to be output at the moment, and the server side can adjust and optimize the strategy for generating the third pushing information set in advance according to the output early warning information, so that the effect of carrying out homogenization problem early warning on the third pushing information set after the preset time is achieved.
In an example, the prediction model may be based on obtaining a plurality of user information of the platform through windows, so that the level prediction may be performed through sampling at a plurality of time points, and the steps of S120-S150 are performed on the plurality of user information, wherein the plurality of user information may include user information of different users and different time points, so as to raise the prediction result to the homogenization problem prediction of the platform.
Further, since the processing method for homogenizing the push information in this embodiment may be applied to a server, where the server may be connected to a plurality of clients, each client may include different user information, that is, there may be a plurality of user information, at this time, the user information of the client connected to the server may be obtained, and the steps S120-S150 may be performed on each user information, so as to play a role in determining whether the manner in which the server currently generates the push information set has diversity.
As shown in fig. 2, the present application further discloses an embodiment of a processing apparatus for homogenizing push information, where the apparatus includes:
an obtaining module 210, configured to obtain current user information;
a first determining module 220, configured to determine whether a first push information set generated based on the current user information belongs to a homogenization type; the homogenization type represents that the proportion of push information belonging to a target class in the first push information set is larger than a preset threshold;
a first determining module 230, configured to determine a push cause type of the first push information set if the first push information set belongs to the homogenization type;
a second determining module 240, configured to determine an adjustment policy based on the push cause type, where the adjustment policy represents a policy for adjusting push information in the first set of push information;
the generating module 250 is configured to adjust the push information in the first set of push information based on the adjustment policy, so as to generate a target set of push information.
In an embodiment, the first determining module 220 may include:
a first determining unit configured to determine a first user feature set based on current user information, the first user feature set representing a set of features determined based on the current user information;
the prediction unit is used for inputting the first user characteristic set into a preset prediction model so as to judge whether the first pushing information set belongs to the homogenization type.
In one embodiment, the training process of the predictive model is as follows:
acquiring first user information and determining a second pushing information set based on the first user information;
determining second homogenization information of the second push information set, wherein the second homogenization information represents whether the second push information set belongs to the homogenization type;
determining a second set of user features based on the first user information, the second set of user features representing a set of features determined based on the first user information;
and training to obtain the prediction model by taking the second user characteristic set as input data and the second homogeneous information as expected output data.
In an embodiment, the prediction unit may include:
the input subunit is used for inputting the first user characteristic set into a preset prediction model;
a first determining subunit, configured to determine, based on an output of a prediction model, first homogenization information, where the first homogenization information indicates whether the first push information set belongs to the homogenization type;
and the judging subunit is used for judging whether the first pushing information set belongs to the homogenization type or not based on the first homogenization information.
In an embodiment, the prediction unit may further include:
a second determining subunit, configured to determine, based on the output of the prediction model, a feature ranking table, where the feature ranking table represents a ranking table of importance of features in the first user feature set from high to low;
a third determining subunit, configured to determine, as first target feature information, a feature with an importance degree lower than a preset importance degree in the feature ranking table;
a removing subunit for removing the feature represented by the first target feature information from the first user feature set.
In an embodiment, the first determining module 230 may include:
a second determining unit, configured to determine type information of a target category if the first push information set belongs to the homogenization type;
and a third determining unit, configured to determine a push cause type of the first push information set based on the type information.
In an embodiment, the apparatus may further include:
the third determining module is used for determining target user information after preset time under the condition that the first pushing information set generated based on the current user information does not belong to the homogenization type;
the second judging module is used for judging whether the generated third pushing information set belongs to the homogenization type or not based on the target user information;
and the early warning module is used for outputting early warning information under the condition that the third pushing information set belongs to the homogenization type.
The implementation process of the functions and roles of each module in the above device is specifically shown in the implementation process of the corresponding steps in the above method, and will not be described herein again.
As shown in fig. 3, an embodiment of the present application provides an electronic device including a processor 310, a communication interface 320, a memory 330, and a communication bus 340, wherein the processor 310, the communication interface 320, the memory 330 complete communication with each other through the communication bus 340,
a memory 330 for storing a computer program;
in one embodiment of the present application, the processor 310 is configured to implement the method provided in any of the foregoing method embodiments when executing the program stored on the memory 330.
The present application also provides a computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements a method as provided in any of the method embodiments described above.
It should be noted that in this document, relational terms such as "first" and "second" and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.
The foregoing description has been directed to specific embodiments of this specification. Other embodiments are within the scope of the following claims. In some cases, the actions or steps recited in the claims can be performed in a different order than in the embodiments and still achieve desirable results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
The foregoing is only a specific embodiment of the application to enable those skilled in the art to understand or practice the application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the application. Thus, the present application is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (9)
1. A method for processing push information homogenization, the method comprising:
acquiring current user information;
judging whether the first pushing information set generated based on the current user information belongs to a homogenization type or not, wherein the method comprises the following steps: determining a first set of user features based on the current user information, the first set of user features representing a set of features determined based on the current user information; inputting the first user feature set into a preset prediction model to judge whether the first pushing information set belongs to the homogenization type; the homogenization type represents that the proportion of push information belonging to a target class in the first push information set is larger than a preset threshold;
determining a pushing cause type of the first pushing information set under the condition that the first pushing information set belongs to the homogenization type;
determining an adjustment strategy based on the push cause type, wherein the adjustment strategy represents a strategy for adjusting push information in the first push information set;
and adjusting the push information in the first push information set based on the adjustment strategy to generate a target push information set.
2. The method according to claim 1, wherein the training process of the predictive model is as follows:
acquiring first user information and determining a second pushing information set based on the first user information;
determining second homogenization information of the second push information set, wherein the second homogenization information represents whether the second push information set belongs to the homogenization type;
determining a second set of user features based on the first user information, the second set of user features representing a set of features determined based on the first user information;
and training to obtain the prediction model by taking the second user characteristic set as input data and the second homogeneous information as expected output data.
3. The method of claim 1, wherein the inputting the first set of user features into a preset predictive model to determine whether the first set of push information is of the homogenization type comprises:
inputting the first user feature set into a preset prediction model;
determining first homogenization information based on an output of the prediction model, the first homogenization information representing whether the first push information set is of the homogenization type;
based on the first homogenization information, whether the first push information set belongs to the homogenization type is judged.
4. The method of claim 3, wherein the inputting the first set of user features into a preset predictive model, further comprises thereafter:
determining a feature ranking table based on the output of the prediction model, wherein the feature ranking table represents a ranking table of the importance of features in the first user feature set from high to low;
determining the features with importance degrees lower than a preset importance degree in the feature ordering table as first target feature information;
the features represented by the first target feature information are removed from the first set of user features.
5. The method of claim 1, wherein the determining a push cause type for the first set of push information if the first set of push information is of the homogenization type comprises:
determining type information of the target category under the condition that the first pushing information set belongs to the homogenization type;
a push cause type of the first set of push information is determined based on the type information.
6. The method of claim 1, wherein the determining whether the first set of push information generated based on the current user information is of a homogenous type, then comprises:
determining target user information after a preset time under the condition that a first pushing information set generated based on the current user information does not belong to the homogenization type;
judging whether the generated third pushing information set belongs to the homogenization type or not based on the target user information;
and outputting early warning information under the condition that the third pushing information set belongs to the homogenization type.
7. A processing apparatus for homogenizing push information, the apparatus comprising:
the acquisition module is used for acquiring current user information;
the first judging module is configured to judge whether a first push information set generated based on the current user information belongs to a homogenization type, and includes: determining a first set of user features based on the current user information, the first set of user features representing a set of features determined based on the current user information; inputting the first user feature set into a preset prediction model to judge whether the first pushing information set belongs to the homogenization type; the homogenization type represents that the proportion of push information belonging to a target class in the first push information set is larger than a preset threshold;
the first determining module is used for determining a pushing cause type of the first pushing information set under the condition that the first pushing information set belongs to the homogenization type;
a second determining module, configured to determine an adjustment policy based on the push cause type, where the adjustment policy represents a policy for adjusting push information in the first set of push information;
and the generation module is used for adjusting the push information in the first push information set based on the adjustment strategy so as to generate a target push information set.
8. The electronic equipment is characterized by comprising a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory are communicated with each other through the communication bus;
a memory for storing a computer program;
a processor for implementing the method of any of claims 1-6 when executing a program stored on a memory.
9. A computer readable storage medium, on which a computer program is stored, which computer program, when being executed by a processor, implements the method according to any of claims 1-6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310959221.0A CN116684481B (en) | 2023-08-01 | 2023-08-01 | Method and device for processing push information homogenization, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310959221.0A CN116684481B (en) | 2023-08-01 | 2023-08-01 | Method and device for processing push information homogenization, electronic equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116684481A CN116684481A (en) | 2023-09-01 |
| CN116684481B true CN116684481B (en) | 2023-11-21 |
Family
ID=87779516
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310959221.0A Active CN116684481B (en) | 2023-08-01 | 2023-08-01 | Method and device for processing push information homogenization, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116684481B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110020192A (en) * | 2018-07-31 | 2019-07-16 | 北京微播视界科技有限公司 | A kind of information content method for pushing and device, server device |
| CN110677482A (en) * | 2019-09-29 | 2020-01-10 | 上海掌门科技有限公司 | Method, equipment and computer storage medium for pushing information |
| CN112995248A (en) * | 2019-12-12 | 2021-06-18 | 阿里巴巴集团控股有限公司 | Information pushing method, device and equipment |
| CN114218482A (en) * | 2021-12-15 | 2022-03-22 | 上海幻电信息科技有限公司 | Information pushing method and device |
| CN114579861A (en) * | 2022-03-03 | 2022-06-03 | 腾讯科技（深圳）有限公司 | Information pushing method and device, electronic equipment and readable storage medium |
| CN116405453A (en) * | 2023-04-23 | 2023-07-07 | 中航信移动科技有限公司 | Information pushing method based on multiple features, storage medium and electronic equipment |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2015004276A2 (en) * | 2013-07-12 | 2015-01-15 | Canon Kabushiki Kaisha | Adaptive data streaming method with push messages control |
-
2023
- 2023-08-01 CN CN202310959221.0A patent/CN116684481B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110020192A (en) * | 2018-07-31 | 2019-07-16 | 北京微播视界科技有限公司 | A kind of information content method for pushing and device, server device |
| CN110677482A (en) * | 2019-09-29 | 2020-01-10 | 上海掌门科技有限公司 | Method, equipment and computer storage medium for pushing information |
| CN112995248A (en) * | 2019-12-12 | 2021-06-18 | 阿里巴巴集团控股有限公司 | Information pushing method, device and equipment |
| CN114218482A (en) * | 2021-12-15 | 2022-03-22 | 上海幻电信息科技有限公司 | Information pushing method and device |
| CN114579861A (en) * | 2022-03-03 | 2022-06-03 | 腾讯科技（深圳）有限公司 | Information pushing method and device, electronic equipment and readable storage medium |
| CN116405453A (en) * | 2023-04-23 | 2023-07-07 | 中航信移动科技有限公司 | Information pushing method based on multiple features, storage medium and electronic equipment |
Non-Patent Citations (1)
| Title |
|---|
| 云环境下个性化推送搜索引擎的设计;苟廷熹;北京邮电大学工程硕士专业学位论文;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116684481A (en) | 2023-09-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105119809B (en) | dynamic information display method and device | |
| CN108073659B (en) | Wedding and love object recommendation method and device | |
| JP5134091B2 (en) | Method and system for determining user suitability of a target content message using a cache missed state match indicator in a mobile environment | |
| US20090240647A1 (en) | Method and appratus for detecting patterns of behavior | |
| CN109462769A (en) | Direct broadcasting room pendant display methods, device, terminal and computer-readable medium | |
| CN109858962B (en) | Advertisement display method based on electronic book and electronic equipment | |
| WO2022165664A1 (en) | Live broadcast interface display method and apparatus, terminal, server, and storage medium | |
| CN104992348B (en) | A kind of method and apparatus of information displaying | |
| CN113535991B (en) | Multimedia resource recommendation method and device, electronic equipment and storage medium | |
| CN109074359A (en) | Use model optimization content distribution | |
| CN109493138A (en) | Information recommendation method, device, server and storage medium | |
| KR101991609B1 (en) | Filtering content based on user mobile networks and data plans | |
| CN113015010B (en) | Push parameter determination method, device, equipment and computer readable storage medium | |
| CN116684481B (en) | Method and device for processing push information homogenization, electronic equipment and storage medium | |
| CN107547626B (en) | User portrait sharing method and device | |
| US20230364501A1 (en) | Information provision device, information provision method, and information provision program | |
| CN116777518A (en) | Transaction management method, device, storage medium and equipment | |
| CN113553509B (en) | Content recommendation method and device, electronic equipment and storage medium | |
| WO2022150573A1 (en) | Providing ambient information based on learned user context and interaction, and associated systems and devices | |
| KR20230019821A (en) | Editable video search and ranking in multimedia messaging applications | |
| CN111309960A (en) | Singing bill recommendation method and device | |
| CN105554088B (en) | Information-pushing method and device | |
| CN108683926A (en) | Client withdrawal control method, client, server and computer equipment | |
| CN117291670B (en) | Video advertisement playing method and device based on crowd data | |
| CN114374881B (en) | Method and device for distributing user traffic, electronic equipment and storage medium |
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 |