CN116896582A - Multi-level organization-oriented real-time message pushing method, device and system - Google Patents

Multi-level organization-oriented real-time message pushing method, device and system Download PDF

Info

Publication number
CN116896582A
CN116896582A CN202311160489.4A CN202311160489A CN116896582A CN 116896582 A CN116896582 A CN 116896582A CN 202311160489 A CN202311160489 A CN 202311160489A CN 116896582 A CN116896582 A CN 116896582A
Authority
CN
China
Prior art keywords
message
pushed
target
person
pushing
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.)
Granted
Application number
CN202311160489.4A
Other languages
Chinese (zh)
Other versions
CN116896582B (en
Inventor
李强
赵峰
宋卫平
魏伟
高胜杰
桂胜
李晓珍
刘丹
白宇冰
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.)
Sichuan Zhongdian Aostar Information Technologies Co ltd
State Grid Information and Telecommunication Co Ltd
Original Assignee
Sichuan Zhongdian Aostar Information Technologies Co ltd
State Grid Information and Telecommunication 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 Sichuan Zhongdian Aostar Information Technologies Co ltd, State Grid Information and Telecommunication Co Ltd filed Critical Sichuan Zhongdian Aostar Information Technologies Co ltd
Priority to CN202311160489.4A priority Critical patent/CN116896582B/en
Publication of CN116896582A publication Critical patent/CN116896582A/en
Application granted granted Critical
Publication of CN116896582B publication Critical patent/CN116896582B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/50—Network services
    • H04L67/55—Push-based network services
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/16—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks using machine learning or artificial intelligence
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/01—Protocols
    • H04L67/1396—Protocols specially adapted for monitoring users' activity
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/50—Network services
    • H04L67/535—Tracking the activity of the user

Abstract

The application provides a method, a device and a system for pushing real-time messages for multi-level organization, wherein the method comprises the following steps: acquiring a first target content characteristic of a message to be pushed; acquiring a second target content characteristic of historical messaging of a person to be pushed; inputting the first target content characteristics and the second target content characteristics into a pushing judgment model to obtain a judgment result of whether the person to be pushed output by the pushing judgment model is the target person who needs to push the message to be pushed; and pushing the message to be pushed to the target personnel. Therefore, when the message pushing is performed for the multi-level organization, the manual intervention can be reduced, the timeliness of the message pushing is improved, and the risk of missed and false sending of manual operation is reduced.

Description

Multi-level organization-oriented real-time message pushing method, device and system
Technical Field
The present application relates to the field of information technologies, and in particular, to a method, an apparatus, and a system for pushing real-time messages for multi-level organization.
Background
In some scenarios, a message needs to be pushed for some people in an organization architecture with multiple levels, but the people are located at different levels, it is often difficult for the message source to determine all the people that need to push the message at one time, so the message needs to be forwarded or delivered layer by layer in the organization architecture with multiple levels. Specifically, in the existing scheme, please refer to fig. 1, the unfilled box indicates personnel not needed to receive the message, the filled box indicates personnel needed to receive the message, generally, each level personnel is needed to receive the message from the previous level, then, manually judges which personnel need to be sent to the next level according to the content of the message, and forwards the message layer by layer until the message transmission is completed. The time required for pushing the whole organization information is long, and the missed or false sending is easy to occur in manual judgment.
Disclosure of Invention
In order to overcome the defects that the pushing time is long and the leakage is easy to occur in the manual judgment in the prior art, the application aims to provide a real-time message pushing method facing to a multi-level organization, which is applied to a message pushing server and comprises the following steps:
acquiring a first target content characteristic of a message to be pushed;
acquiring a second target content characteristic of historical messaging of a person to be pushed;
inputting the first target content characteristics and the second target content characteristics into a pre-trained push judgment model based on machine learning, and obtaining a judgment result of whether the person to be pushed, which is output by the push judgment model, is a target person who needs to push the message to be pushed;
and pushing the message to be pushed to the target personnel.
In some possible implementations, the method further includes:
obtaining a plurality of training samples, the training samples comprising an anchor sample comprising a first sample content feature of a sample to push message, a positive sample comprising a second sample content feature of historical messaging messages of a first sample person who received the sample to push message, and a negative sample comprising a third sample content feature of historical messaging messages of a second sample person who does not need to receive the sample to push message;
And respectively inputting the training samples into the push judgment model to be trained, extracting characteristics of the anchor sample, the positive sample and the negative sample aiming at each training sample, and adjusting model parameters of the push judgment model according to the characteristic similarity of the anchor sample and the positive sample and the characteristic similarity of the anchor sample and the negative sample so that the push judgment model gives a result for the positive sample, and gives a result for the negative sample, wherein the result is not required to carry out message push.
In some possible implementations, the step of obtaining the first target content feature of the message to be pushed includes:
processing text content in the message to be pushed through a pre-trained language model to obtain a first word vector set corresponding to the message to be pushed;
taking the first word vector set as the first target content feature;
the step of obtaining the second target content characteristics of the historical messaging of the person to be pushed comprises the following steps:
acquiring a plurality of historical transceiver messages of the personnel to be pushed, wherein the historical transceiver messages comprise historical received messages and historical sent messages;
Processing text contents in the historical received and transmitted messages respectively through the language model to obtain a plurality of second word vector sets corresponding to the historical received and transmitted messages;
and taking a plurality of second word vector sets as the second target content characteristics.
In some possible implementations, the step of obtaining a plurality of historical messaging messages of the person to be pushed includes:
acquiring historical receiving information and historical sending information of the personnel to be pushed in a preset time period;
for each history received message, inputting a browsing operation record, an attachment processing record and a forwarding record related to the history received message into a pre-trained attention judging model to be processed, and obtaining a judging result of whether the to-be-pushed person who is output by the attention judging model pays attention to the history received message; the browsing operation record comprises the browsing duration of the history received message and whether all the browsing is performed, the attachment processing record comprises whether the person to be pushed opens and previews the attachment carried by the history received message and downloads the attachment carried by the history received message, and the forwarding record comprises whether the person to be pushed forwards the history received message;
And taking the historical received message with the judgment result output by the attention judgment model as attention and the historical transmitted message as the historical transmitted and received message of the person to be pushed.
In some possible implementations, after the step of processing, by the language model, text content in the plurality of historical received messages respectively to obtain a plurality of second word vector sets corresponding to the plurality of historical received messages, the method further includes:
performing similarity comparison on the plurality of second word vector sets, and dividing the plurality of second word vector sets into different message type groups according to comparison results;
one second word vector set is reserved for each message type group, and other second word vector sets in the message type group are deleted.
In some possible implementations, the method further includes:
obtaining a personnel message forwarding relationship topological graph according to the identity of each target personnel and the historical transceiver relationship record;
judging whether a confirmation consulting license sent by the target person at the upper level of the target person for the target person is received or not according to the person message forwarding relation topological graph;
If the confirmation consulting permission sent by the target personnel is not received, the terminal of the target personnel is prevented from displaying the message to be pushed;
and if the confirmation consulting permission sent by the target personnel is received, allowing the terminal of the target personnel to display the message to be pushed, and sending a lower-level consulting request to the target personnel, wherein the lower-level consulting request is used for requesting the target personnel to send the confirmation consulting permission to other target personnel, the lower level of which is allowed to consult the message to be pushed, aiming at the target personnel.
In some possible implementations, the step of pushing the message to be pushed to the target person includes:
pushing the encrypted message to be pushed to the target personnel;
and if the confirmation consulting permission sent by the target person is received, allowing the terminal of the target person to display the message to be pushed, wherein the step comprises the following steps:
and if the confirmation consulting permission sent by the target personnel is received, sending a decryption code to the terminal of the target personnel so as to allow the terminal of the target personnel to decrypt and display the message to be pushed.
In some possible implementations, after the step of preventing the terminal of the target person from displaying the message to be pushed if the confirmation review permission sent for the target person is not received, the method further includes:
And after receiving no confirmation and reference permission sent by the target personnel within a preset time length, notifying the terminal of the target personnel to delete the message to be pushed.
Another object of the present application is to provide a real-time message pushing device for multi-level organization, which is applied to a message pushing server, and the real-time message pushing device for multi-level organization includes:
the first data acquisition module is used for acquiring first target content characteristics of the message to be pushed;
the second data acquisition module is used for acquiring a second target content characteristic of the historical transceiving information of the person to be pushed;
the pushing judgment module is used for inputting the first target content characteristics and the second target content characteristics into a pre-trained pushing judgment model based on machine learning, and obtaining a judgment result of whether the person to be pushed, which is output by the pushing judgment model, is a target person needing to push the message to be pushed;
and the pushing execution module is used for pushing the message to be pushed to the target personnel.
Another object of the present application is to provide a multi-level organization oriented real-time message pushing system, the system comprising a message pushing server, the message pushing server comprising a processor and a machine-readable storage medium storing machine-executable instructions which, when executed by the processor, implement the multi-level organization oriented real-time message pushing method described above.
Compared with the prior art, the application has the following beneficial effects:
according to the real-time message pushing method, device and system for the multi-level organization, the pushing judgment model predicts according to the first target content characteristics of the message to be pushed and the second target content characteristics of the historical receiving and sending messages of the person to be pushed, so that the target person needing to push the message to be pushed can be automatically determined according to the content of the message to be pushed, and further the pushing action of the message to be pushed is automatically completed. Therefore, when the message pushing is performed for the multi-level organization, the manual intervention can be reduced, the timeliness of the message pushing is improved, and the risk of missed and false sending of manual operation is reduced.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are needed in the embodiments will be briefly described below, it being understood that the following drawings only illustrate some embodiments of the present application and therefore should not be considered as limiting the scope, and other related drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a diagram of a conventional message pushing process;
fig. 2 is a schematic flow chart of steps of a method for pushing real-time messages for multi-level organization according to an embodiment of the present application;
fig. 3 is a schematic diagram of a message pushing process of a real-time message pushing method for multi-level organization according to an embodiment of the present application;
FIG. 4 is a second step flow chart of a method for pushing real-time messages to a multi-level organization according to an embodiment of the present application;
fig. 5 is a second schematic diagram of a message pushing process of the multi-level organization-oriented real-time message pushing method according to an embodiment of the present application;
fig. 6 is a schematic diagram of an electronic device according to an embodiment of the present application;
fig. 7 is a schematic functional block diagram of a real-time message pushing device for multi-level organization 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. The components of the embodiments of the present application generally described and illustrated in the figures herein may be arranged and designed in a wide variety of different configurations.
Thus, the following detailed description of the embodiments of the application, as presented in the figures, is not intended to limit the scope of the application, as claimed, but is merely representative of selected embodiments of the 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.
It should be noted that: like reference numerals and letters denote like items in the following figures, and thus once an item is defined in one figure, no further definition or explanation thereof is necessary in the following figures.
In the description of the present application, it should be noted that the terms "first," "second," "third," and the like are used merely to distinguish between descriptions and are not to be construed as indicating or implying relative importance.
Furthermore, the terms "horizontal," "vertical," "overhang," and the like do not denote a requirement that the component be absolutely horizontal or overhang, but rather may be slightly inclined. As "horizontal" merely means that its direction is more horizontal than "vertical", and does not mean that the structure must be perfectly horizontal, but may be slightly inclined.
In the description of the present application, it should also be noted that, unless explicitly specified and limited otherwise, the terms "disposed," "mounted," "connected," and "connected" are to be construed broadly, and may be, for example, fixedly connected, detachably connected, or integrally connected; can be mechanically or electrically connected; can be directly connected or indirectly connected through an intermediate medium, and can be communication between two elements. The specific meaning of the above terms in the present application will be understood in specific cases by those of ordinary skill in the art.
Referring to fig. 2, fig. 2 is a schematic step flow diagram of a method for pushing real-time messages for multi-level organization according to the present embodiment, and each step of the method is described in detail below.
Step S110, obtaining a first target content feature of the message to be pushed.
In this embodiment, the first target content feature may characterize content, such as text content, in the message to be pushed.
Optionally, in some possible implementations, text content in the message to be pushed may be processed through a pre-trained language model to obtain a first set of word vectors corresponding to the message to be pushed. The first set of word vectors is then treated as the first target content feature.
Illustratively, in this example, the language model may perform text cleansing (removing special characters, punctuation, etc.), word segmentation (dividing sentences into word or sub-word units), stop-word removal (common but useless for analysis), etc. on the text content in the input message to be pushed. And then converting the words or characters in the processed text content into word vectors, and taking the word vectors of the word characters as the first target content characteristics.
In one example, the language model may employ Word2Vec, gloVe (Global Vectors for Word Representation), fastText or BERT (Bidirectional Encoder Representations from Transformers), etc.
Step S120, obtaining a second target content feature of the historical messaging of the person to be pushed.
In this embodiment, the second target content feature may represent a content feature of a message received by the person to be pushed in a history and a content of a message sent by the person to be pushed in a history, such as a text content.
Alternatively, a plurality of history send-receive messages of the person to be pushed may be acquired in step S120, where the history send-receive messages include a history receive message and a history send message. And then, respectively processing text contents in the historical received and transmitted messages through the language model to obtain a plurality of second word vector sets corresponding to the historical received and transmitted messages. And then taking the plurality of second word vector sets as the second target content characteristics.
For example, in this embodiment, a plurality of the history reception messages and a plurality of the history transmission messages of the person to be pushed may be acquired. And then respectively processing the plurality of historical received messages and the plurality of historical transmitted messages through the language model, so as to obtain a plurality of second word vector sets respectively corresponding to the plurality of historical received messages and the plurality of historical transmitted messages. A plurality of the second set of word vectors may then be taken together as the second target content feature.
Step S130, inputting the first target content feature and the second target content feature into a pre-trained push judgment model based on machine learning, to obtain a judgment result of whether the person to be pushed output by the push judgment model is the target person who needs to push the message to be pushed.
In this embodiment, the second target content feature may characterize what content the person to be pushed has historically received, or send/forward what content. On the basis, the push judgment model extracts and analyzes the correlation between the first target content characteristic and the second target content characteristic of the message to be pushed, so as to determine whether the message to be pushed needs to be matched with the person to be pushed.
The second word vector set corresponding to the historical received messages of the person to be pushed can represent what content of the message the person to be pushed has received, so that the second word vector set can represent that the person to be pushed has a greater possibility of receiving messages with similar content to a certain extent.
However, the history received message may further include a message that the person to be pushed receives, but does not need to pay attention to, so in this embodiment, the second word vector set corresponding to the history sent messages of the person to be pushed needs to be combined, where the second word vector set corresponding to the history sent message may indicate what content has been sent or forwarded by the person to be pushed, that is, the person to be pushed needs to process the messages, so that the second word vector set may also indicate that the person to be pushed has a greater likelihood of receiving a message with similar content.
Compared with the scheme that only the historical received message is concerned, the scheme provided by the embodiment also needs to pay attention to the historical transmitted message, so that the second target content characteristic can more accurately reflect the tendency of the person to be pushed to pay attention to the message content, and more powerful data support is provided for the follow-up prediction judging action.
And step S140, pushing the message to be pushed to the target personnel.
In this embodiment, after the target person is determined, the message to be pushed may be directly pushed to the target person.
Based on the above design, in the solution provided in this embodiment, the prediction processing is performed by the push judgment model according to the first target content feature of the message to be pushed and the second target content feature of the historical message receiving and sending of the person to be pushed, so that the target person to be pushed can be automatically determined according to the content of the message to be pushed, and further the pushing action of the message to be pushed is automatically completed. That is, referring to fig. 3, the box without filling in fig. 3 indicates a person who does not need to receive a message, and the box with filling indicates a person who needs to receive a message, in this embodiment, the original multi-stage person is manually confirmed and forwarded step by step according to a message to be pushed, and is converted to automatically determine the target person who needs to receive pushing according to the content of the message to be pushed by the message pushing server. Therefore, when the message pushing is performed for the multi-level organization, the manual intervention can be reduced, the timeliness of the message pushing is improved, and the risk of missed and false sending of manual operation is reduced.
In some possible implementations, the method provided in this embodiment may further include a step of training the push judgment model, and specifically may include the following steps.
Step S210, obtaining a plurality of training samples, where the training samples include an anchor sample, a positive sample, and a negative sample, where the anchor sample includes a first sample content feature of a sample to-be-pushed message, the positive sample includes a second sample content feature of a historical messaging message of a first sample person who received the sample to-be-pushed message, and the negative sample includes a third sample content feature of a historical messaging message of a second sample person who does not need to receive the sample to-be-pushed message.
In this embodiment, a message that has been pushed by the history may be used as the sample to-be-pushed message, a person that has received the sample to-be-pushed message may be used as the first sample person, and a person that does not need to receive the sample to-be-pushed message may be used as the second sample person.
A first sample content feature of the sample to-be-pushed message is obtained, for example, in the same manner as in step S110, as the anchor sample.
And acquiring a plurality of historical transceiving messages of the first sample personnel, and extracting features to acquire the second sample content features as the positive samples. For example, feature extraction is performed in a manner similar to step S120.
And acquiring a plurality of historical transceiving messages of the second sample personnel, and extracting features to acquire the third sample content features as the negative samples. For example, feature extraction is performed in a manner similar to step S120.
It should be noted that, in this embodiment, for the same message to be pushed, different combinations of the first sample person and the second sample person may be determined, so as to form different training samples. Different training samples can be formed for different messages to be pushed.
Step S220, respectively inputting the training samples into the push judgment model to be trained, extracting features of the anchor sample, the positive sample and the negative sample for each training sample, and adjusting model parameters of the push judgment model according to feature similarity of the anchor sample and the positive sample and feature similarity of the anchor sample and the negative sample, so that the push judgment model gives a result of message push required to be executed for the positive sample, and gives a result of message push not required to be executed for the negative sample.
In this embodiment, the push judgment model may further extract features of the anchor sample, the positive sample, and the negative sample in the training sample, then calculate feature similarities of the anchor sample and the positive sample, and feature similarities of the anchor sample and the negative sample, and according to a preset loss function (e.g., a triplet loss function), adjust model parameters of the push judgment model with the objective of improving feature similarities of the anchor sample and the positive sample and reducing feature similarities of the anchor sample and the negative sample. And obtaining the trained push judgment model until the feature similarity of the anchor sample and the positive sample extracted by the push judgment model is greater than a preset threshold, or the difference between the feature similarity of the anchor sample and the positive sample and the feature similarity of the anchor sample and the negative sample reaches a preset threshold, or the training times reach a preset threshold.
In this embodiment, the push judgment model may use a general feature extraction and regression network, which is not specifically limited herein.
On this basis, in the step S130, the pushing judgment model may input the first target content feature and the second target content feature, perform further feature extraction, and then calculate the feature similarity, and judge that the person to be pushed is the target person when the feature similarity reaches a preset threshold.
In one possible implementation manner, after the step of processing text content in the historical transceiver messages through the language model to obtain a plurality of second word vector sets corresponding to the historical transceiver messages, similarity comparison may be performed on the second word vector sets, and the second word vector sets are divided into different message type groups according to the comparison result. Then, one of the second word vector sets is reserved for each of the message type groups, and the other second word vector sets in the message type group are deleted.
And performing de-duplication processing on the second word vector sets with higher similarity, so that each reserved second word vector set represents a message with one type of content. In this way, the calculation amount of the subsequent processing can be reduced.
In a possible implementation manner, the accuracy of the second target content feature acquired in step S120 may perform a filtering process on the historical received message after acquiring a plurality of historical received messages of the person to be pushed.
Specifically, in step S120, a history reception message and a history transmission message of the person to be pushed in a preset period of time may be acquired first.
And then, aiming at each historical received message, inputting a browsing operation record, an attachment processing record and a forwarding record related to the historical received message into a pre-trained attention judging model to process, and obtaining a judging result of whether the to-be-pushed person outputting by the attention judging model pays attention to the historical received message.
The browsing operation record comprises browsing duration of the historical received message and whether all browsing is performed. For example, if the browsing time corresponding to a certain historical received message is short, or the browsing is not completed, it means that the person receiving the person waiting for pushing does not pay attention to the historical received message. If the browsing time corresponding to a certain historical received message is longer, the fact that the to-be-pushed person receiving the message carefully reads the historical received message is indicated, and further the to-be-pushed person receiving the message is characterized to pay attention to the historical received message.
The accessory processing record comprises whether the person to be pushed opens and previews the accessory carried by the history receiving message or downloads the accessory carried by the history receiving message. For example, if an attachment in a certain historical received message is not opened for preview or downloaded by the person to be pushed, the person to be pushed who receives the message is characterized to not pay attention to the historical received message; if the attachment in a certain historical received message is opened for preview or downloaded by the person to be pushed, the person to be pushed who receives the message is characterized to pay attention to the historical received message.
The forwarding record comprises whether the person to be pushed forwards the historical received message or not. For example, if a certain history reception is forwarded by the person to be pushed, the person to be pushed who characterizes the received message pays attention to the history reception message, or the lower level thereof needs to pay attention to the history reception message.
After the judging result output by the attention judging model is obtained, the historical received message with the judging result output by the attention judging model as attention and the historical transmitted message can be used as the historical transmitted and received message of the person to be pushed.
And in the final historical received messages, only the historical received messages which are judged to be focused by the people to be pushed by the focusing judgment model do not include the historical received messages which are judged to be not focused by the people to be pushed by the focusing judgment model.
Accordingly, the attention judgment model may be trained prior to step S120.
For example, a plurality of historical received messages of a fourth sample person are obtained, and labels of the fourth sample person for manual labeling of the historical received messages are obtained, wherein the labels comprise attention and non-attention, browsing operation records, attachment processing records and forwarding records corresponding to the historical received messages are also respectively obtained as training samples, and the training samples and the manually labeled labels are input into the attention judging model for training, so that the attention model can judge whether the historical received messages are concerned or not according to the browsing operation records, the attachment processing records and the forwarding records.
In this embodiment, the attention judging model may employ a general feature extraction and classification network, which is not particularly limited herein.
In a possible implementation manner, the steps shown in fig. 2 may be used to push the message to be pushed to the terminal of the target person in advance, and then determine whether the target person can use the message to be pushed by a manual step-by-step confirmation manner later. The message to be pushed can be a file with larger data quantity, the target personnel can be determined through the steps shown in fig. 2, then the file with larger data quantity is pushed to each target personnel in advance, and then whether the target personnel can open the file is determined through a manual step-by-step confirmation mode. Thus, the pre-pushing of the larger file is finished in advance through the steps shown in fig. 2, and the message transmission with smaller data volume needs to be executed in the subsequent manual step-by-step confirmation, so that the efficiency of file pushing confirmation can be improved.
Specifically, in this embodiment, referring to fig. 4, the method further includes the following steps.
Step S310, a personnel message forwarding relationship topological graph is obtained according to the identity of each target personnel and the history transceiver relationship record.
Different target persons may be located at different levels in the multi-level organization, and the person message forwarding relationship topology map related to each target person may be obtained according to the identity or the historical sender-receiver relationship record.
Step S320, according to the person message forwarding relationship topology diagram, for each target person, determining whether a confirmation review permission sent by the target person for the target person at the previous stage of the target person is received.
In this embodiment, except for the person from the source of the message to be pushed, judgment may be performed step by step for each target person, and the previous stage of the target person is checked to check the confirmation reference permission sent by the target person for the target person.
If the confirmation review permission sent for the target person is not received, step S330 is performed.
If the confirmation review permission sent for the target person has been received, step S340 is performed.
And step S330, preventing the terminal of the target person from displaying the message to be pushed.
Specifically, if the confirmation review permission sent by the target person is not received, the upper level of the target person determines that the target person does not need or does not pay attention to or should not review the message to be pushed, so that the terminal of the target person can be prevented from displaying the message to be pushed.
Step S340, allowing the terminal of the target person to display the message to be pushed, and sending a lower-level review request to the target person, where the lower-level review request is used to request the target person to send a confirmation review permission to other target persons for which lower-level review is allowed.
Specifically, if the confirmation and reference permission sent by the target person is received, the upper level of the target person determines that the target person needs or should refer to the message to be pushed, so that the terminal of the target person can be allowed to display the message to be pushed. And, the target person is required to confirm whether other target persons who have received the message to be pushed in the lower stage can review the message to be pushed, so that a lower stage review request can be sent to the target person.
Further, in this embodiment, in step S140, the encrypted message to be pushed may be pushed to the target person.
Then in step S340, if the confirmation review permission sent for the target person has been received, a decryption code is sent to the terminal of the target person, so as to allow the terminal of the target person to decrypt and display the message to be pushed.
That is, referring to fig. 5, the box without filling in fig. 5 indicates a person who does not need to receive a message, and the box with filling indicates a person who needs to receive a message, in this embodiment, the pre-pushing of the encrypted message to be pushed with a larger data size can be automatically completed through the pushing judgment model, and then the pushing of the decryption code is completed through a manual step-by-step confirmation manner. Therefore, as the pre-pushing of the file with larger data volume is executed first, only the password-decoding interaction with smaller data volume is needed to be executed in the subsequent step-by-step manual confirmation process, the pushing efficiency of the message file can be effectively improved, and the safety is ensured.
Further, in this embodiment, after step S330, after receiving no confirmation review permission sent by the target person within a preset period of time, the terminal of the target person is notified to delete the message to be pushed. Therefore, the occupation of the storage space of the terminal of the target person needing no consulting the message to be pushed is avoided, and the data security is improved.
The embodiment also provides a real-time message pushing system facing to the multi-level organization, the system comprises a message pushing server, the message pushing server comprises a processor and a machine-readable storage medium, the machine-readable storage medium stores machine-executable instructions, and the machine-executable instructions realize the real-time message pushing method facing to the multi-level organization provided by the embodiment when being executed by the processor.
In some examples, the message push server may be an electronic device with certain data processing capabilities. Referring to fig. 6, a block diagram of the electronic device 100 in fig. 6 is shown. The electronic device 100 comprises a real-time message pushing means 110, a machine readable storage medium 120, a processor 130 for multi-level organization.
The machine-readable storage medium 120 and the processor 130 are directly or indirectly electrically connected to each other to implement data transmission or interaction. For example, the components may be electrically connected to each other via one or more communication buses or signal lines. The multi-level organization oriented real-time message pushing device 110 includes at least one software function module that may be stored in the machine readable storage medium 120 in the form of software or firmware (firmware) or cured in an Operating System (OS) of the electronic device 100. The processor 130 is configured to execute executable modules stored in the machine-readable storage medium 120, such as software functional modules and computer programs included in the multi-level organization-oriented real-time message pushing device 110.
The machine-readable storage medium 120 may be, but is not limited to, random access Memory (Random Access Memory, RAM), read Only Memory (ROM), programmable read Only Memory (Programmable Read-Only Memory, PROM), erasable read Only Memory (Erasable Programmable Read-Only Memory, EPROM), electrically erasable read Only Memory (ElectricErasable Programmable Read-Only Memory, EEPROM), etc. The machine-readable storage medium 120 is configured to store a program, and the processor 130 executes the program/executable the multi-level organization-oriented real-time message pushing method provided in the present embodiment after receiving an execution instruction.
The processor 130 may be an integrated circuit chip with signal processing capabilities. The processor may be a general-purpose processor, including a central processing unit (Central Processing Unit, CPU for short), a network processor (Network Processor, NP for short), etc.; but also Digital Signal Processors (DSPs), application Specific Integrated Circuits (ASICs), field Programmable Gate Arrays (FPGAs) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components. The disclosed methods, steps, and logic blocks in the embodiments of the present application may be implemented or performed. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
Referring to fig. 7, the present embodiment further provides a real-time message pushing device 110 for multi-level organization, where the real-time message pushing device 110 for multi-level organization includes at least one functional module that can be stored in a machine-readable storage medium 120 in a software form. Functionally, the real-time message pushing device 110 for multi-level organization may include a first data acquisition module 111, a second data acquisition module 112, a pushing judgment module 113, and a pushing execution module 114.
The first data obtaining module 111 is configured to obtain a first target content feature of a message to be pushed.
In this embodiment, the first data acquisition module 111 may be used to perform step S110 shown in fig. 2, and a specific description of the first data acquisition module 111 may refer to a description of step S110.
The second data obtaining module 112 is configured to obtain a second target content characteristic of the historical messaging of the person to be pushed.
In this embodiment, the second data acquisition module 112 may be used to perform step S120 shown in fig. 2, and a specific description of the second data acquisition module 112 may refer to a description of step S120.
The pushing judgment module 113 is configured to input the first target content feature and the second target content feature into a pre-trained pushing judgment model based on machine learning, and obtain a judgment result output by the pushing judgment model, whether the person to be pushed is a target person who needs to push the message to be pushed.
In this embodiment, the push determining module 113 may be configured to execute step S130 shown in fig. 2, and a specific description of the push determining module 113 may refer to a description of step S130.
The push execution module 114 is configured to push the message to be pushed to the target person
In this embodiment, the push execution module 114 may be used to execute step S140 shown in fig. 2, and a specific description of the push execution module 114 may refer to a description of step S140.
In summary, according to the method, the device and the system for pushing real-time messages for multi-level organization provided by the embodiment of the application, the pushing judgment model predicts according to the first target content characteristics of the messages to be pushed and the second target content characteristics of the historical transceiving messages of the people to be pushed, so that the target people needing to push the messages to be pushed can be automatically determined according to the content of the messages to be pushed, and further the pushing action of the messages to be pushed is automatically completed. Therefore, when the message pushing is performed for the multi-level organization, the manual intervention can be reduced, the timeliness of the message pushing is improved, and the risk of missed and false sending of manual operation is reduced.
In the embodiments provided in the present application, it should be understood that the disclosed apparatus and method may be implemented in other manners. The apparatus embodiments described above are merely illustrative, for example, of the flowcharts and block diagrams in the figures that illustrate the architecture, functionality, and operation of possible implementations of apparatus, methods and computer program products according to various embodiments of the present application. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
In addition, functional modules in the embodiments of the present application may be integrated together to form a single part, or each module may exist alone, or two or more modules may be integrated to form a single part.
The functions, if implemented in the form of software functional modules and sold or used as a stand-alone product, may be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution, in the form of a software product stored in a storage medium, comprising several instructions for causing a computer device (which may be a personal computer, a server, a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present application. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
It is noted that 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 above description is merely illustrative of various embodiments of the present application, but the scope of the present application is not limited thereto, and any person skilled in the art can easily think about variations or substitutions within the scope of the present application, and the application is intended to be covered by the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.

Claims (10)

1. A real-time message pushing method for a multi-level organization, which is characterized by being applied to a message pushing server, the method comprising:
acquiring a first target content characteristic of a message to be pushed;
acquiring a second target content characteristic of historical messaging of a person to be pushed;
inputting the first target content characteristics and the second target content characteristics into a pre-trained push judgment model based on machine learning, and obtaining a judgment result of whether the person to be pushed, which is output by the push judgment model, is a target person who needs to push the message to be pushed;
and pushing the message to be pushed to the target personnel.
2. The multi-level organization oriented real-time message pushing method of claim 1, further comprising:
Obtaining a plurality of training samples, the training samples comprising an anchor sample comprising a first sample content feature of a sample to push message, a positive sample comprising a second sample content feature of historical messaging messages of a first sample person who received the sample to push message, and a negative sample comprising a third sample content feature of historical messaging messages of a second sample person who does not need to receive the sample to push message;
and respectively inputting the training samples into the push judgment model to be trained, extracting characteristics of the anchor sample, the positive sample and the negative sample aiming at each training sample, and adjusting model parameters of the push judgment model according to the characteristic similarity of the anchor sample and the positive sample and the characteristic similarity of the anchor sample and the negative sample so that the push judgment model gives a result for the positive sample, and gives a result for the negative sample, wherein the result is not required to carry out message push.
3. The method for pushing real-time messages to a multi-level organization according to claim 1, wherein said step of obtaining a first target content feature of a message to be pushed comprises:
Processing text content in the message to be pushed through a pre-trained language model to obtain a first word vector set corresponding to the message to be pushed;
taking the first word vector set as the first target content feature;
the step of obtaining the second target content characteristics of the historical messaging of the person to be pushed comprises the following steps:
acquiring a plurality of historical transceiver messages of the personnel to be pushed, wherein the historical transceiver messages comprise historical received messages and historical sent messages;
processing text contents in the historical received and transmitted messages respectively through the language model to obtain a plurality of second word vector sets corresponding to the historical received and transmitted messages;
and taking a plurality of second word vector sets as the second target content characteristics.
4. The multi-level organization oriented real-time message pushing method according to claim 3, wherein the step of obtaining a plurality of historic transceiving messages of the person to be pushed comprises:
acquiring historical receiving information and historical sending information of the personnel to be pushed in a preset time period;
for each history received message, inputting a browsing operation record, an attachment processing record and a forwarding record related to the history received message into a pre-trained attention judging model to be processed, and obtaining a judging result of whether the to-be-pushed person who is output by the attention judging model pays attention to the history received message; the browsing operation record comprises the browsing duration of the history received message and whether all the browsing is performed, the attachment processing record comprises whether the person to be pushed opens and previews the attachment carried by the history received message and downloads the attachment carried by the history received message, and the forwarding record comprises whether the person to be pushed forwards the history received message;
And taking the historical received message with the judgment result output by the attention judgment model as attention and the historical transmitted message as the historical transmitted and received message of the person to be pushed.
5. The method for pushing real-time messages to a multi-level organization according to claim 3, wherein after said step of processing text contents in a plurality of said historical received and sent messages respectively through said language model to obtain a plurality of second word vector sets corresponding to a plurality of said historical received and sent messages, said method further comprises:
performing similarity comparison on the plurality of second word vector sets, and dividing the plurality of second word vector sets into different message type groups according to comparison results;
one second word vector set is reserved for each message type group, and other second word vector sets in the message type group are deleted.
6. The multi-level organization oriented real-time message pushing method of claim 1, further comprising:
obtaining a personnel message forwarding relationship topological graph according to the identity of each target personnel and the historical transceiver relationship record;
Judging whether a confirmation consulting license sent by the target person at the upper level of the target person for the target person is received or not according to the person message forwarding relation topological graph;
if the confirmation consulting permission sent by the target personnel is not received, the terminal of the target personnel is prevented from displaying the message to be pushed;
and if the confirmation consulting permission sent by the target personnel is received, allowing the terminal of the target personnel to display the message to be pushed, and sending a lower-level consulting request to the target personnel, wherein the lower-level consulting request is used for requesting the target personnel to send the confirmation consulting permission to other target personnel, the lower level of which is allowed to consult the message to be pushed, aiming at the target personnel.
7. The multi-level organization oriented real-time message pushing method of claim 6, wherein the step of pushing the message to be pushed to the target person comprises:
pushing the encrypted message to be pushed to the target personnel;
and if the confirmation consulting permission sent by the target person is received, allowing the terminal of the target person to display the message to be pushed, wherein the step comprises the following steps:
And if the confirmation consulting permission sent by the target personnel is received, sending a decryption code to the terminal of the target personnel so as to allow the terminal of the target personnel to decrypt and display the message to be pushed.
8. The method for pushing real-time messages to a multi-level organization according to claim 7, wherein after said step of preventing a terminal of the target person from displaying said message to be pushed if a confirmation review permission sent for the target person is not received, said method further comprises:
and after receiving no confirmation and reference permission sent by the target personnel within a preset time length, notifying the terminal of the target personnel to delete the message to be pushed.
9. A real-time message pushing device for a multi-level organization, which is characterized by being applied to a message pushing server, and comprising:
the first data acquisition module is used for acquiring first target content characteristics of the message to be pushed;
the second data acquisition module is used for acquiring a second target content characteristic of the historical transceiving information of the person to be pushed;
the pushing judgment module is used for inputting the first target content characteristics and the second target content characteristics into a pre-trained pushing judgment model based on machine learning, and obtaining a judgment result of whether the person to be pushed, which is output by the pushing judgment model, is a target person needing to push the message to be pushed;
And the pushing execution module is used for pushing the message to be pushed to the target personnel.
10. A multi-level organization oriented real-time message pushing system, characterized in that the system comprises a message pushing server comprising a processor and a machine-readable storage medium storing machine-executable instructions which, when executed by the processor, implement the method of any of claims 1-8.
CN202311160489.4A 2023-09-11 2023-09-11 Multi-level organization-oriented real-time message pushing method, device and system Active CN116896582B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202311160489.4A CN116896582B (en) 2023-09-11 2023-09-11 Multi-level organization-oriented real-time message pushing method, device and system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202311160489.4A CN116896582B (en) 2023-09-11 2023-09-11 Multi-level organization-oriented real-time message pushing method, device and system

Publications (2)

Publication Number Publication Date
CN116896582A true CN116896582A (en) 2023-10-17
CN116896582B CN116896582B (en) 2023-12-15

Family

ID=88311110

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202311160489.4A Active CN116896582B (en) 2023-09-11 2023-09-11 Multi-level organization-oriented real-time message pushing method, device and system

Country Status (1)

Country Link
CN (1) CN116896582B (en)

Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20150089007A1 (en) * 2008-12-12 2015-03-26 At&T Intellectual Property I, L.P. E-mail handling based on a behavioral history
CN107222537A (en) * 2017-05-31 2017-09-29 努比亚技术有限公司 Method, mobile terminal and the storage medium of PUSH message
US20190057164A1 (en) * 2017-08-16 2019-02-21 Beijing Baidu Netcom Science And Technology Co., Ltd. Search method and apparatus based on artificial intelligence
CN110321481A (en) * 2019-06-05 2019-10-11 平安科技（深圳）有限公司 Information push method, system, computer equipment and storage medium
US20210157817A1 (en) * 2019-11-21 2021-05-27 International Business Machines Corporation Context-aware dynamic content assist
CN113132429A (en) * 2019-12-31 2021-07-16 云丁网络技术(北京)有限公司 Method and device for pushing messages
CN114416924A (en) * 2022-01-19 2022-04-29 平安国际智慧城市科技股份有限公司 Message pushing method, device, equipment and medium based on natural language processing
US20230123874A1 (en) * 2020-11-04 2023-04-20 FiscalNote, Inc. Use of Machine-Learning Models in Creating Messages for Advocacy Campaigns
CN116521822A (en) * 2023-03-15 2023-08-01 上海帜讯信息技术股份有限公司 User intention recognition method and device based on 5G message multi-round session mechanism
CN116578777A (en) * 2023-04-18 2023-08-11 平安科技（深圳）有限公司 Message pushing method and device, storage medium and computer equipment

Patent Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20150089007A1 (en) * 2008-12-12 2015-03-26 At&T Intellectual Property I, L.P. E-mail handling based on a behavioral history
CN107222537A (en) * 2017-05-31 2017-09-29 努比亚技术有限公司 Method, mobile terminal and the storage medium of PUSH message
US20190057164A1 (en) * 2017-08-16 2019-02-21 Beijing Baidu Netcom Science And Technology Co., Ltd. Search method and apparatus based on artificial intelligence
CN110321481A (en) * 2019-06-05 2019-10-11 平安科技（深圳）有限公司 Information push method, system, computer equipment and storage medium
US20210157817A1 (en) * 2019-11-21 2021-05-27 International Business Machines Corporation Context-aware dynamic content assist
CN113132429A (en) * 2019-12-31 2021-07-16 云丁网络技术(北京)有限公司 Method and device for pushing messages
US20230123874A1 (en) * 2020-11-04 2023-04-20 FiscalNote, Inc. Use of Machine-Learning Models in Creating Messages for Advocacy Campaigns
CN114416924A (en) * 2022-01-19 2022-04-29 平安国际智慧城市科技股份有限公司 Message pushing method, device, equipment and medium based on natural language processing
CN116521822A (en) * 2023-03-15 2023-08-01 上海帜讯信息技术股份有限公司 User intention recognition method and device based on 5G message multi-round session mechanism
CN116578777A (en) * 2023-04-18 2023-08-11 平安科技（深圳）有限公司 Message pushing method and device, storage medium and computer equipment

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
崔艳萍;阎知知;王小巍;彭媛;: "互联网信息资源用户获取优化推送仿真研究", 计算机仿真, no. 07 *

Also Published As

Publication number Publication date
CN116896582B (en) 2023-12-15

Similar Documents

Publication Publication Date Title
US10659412B2 (en) Method and device for saving chat record of instant messaging
CN110826006B (en) Abnormal collection behavior identification method and device based on privacy data protection
CN110489415B (en) Data updating method and related equipment
US20140324999A1 (en) Apparatus and method for communication
WO2015085917A1 (en) Method, apparatus and system for processing information
CN103916513A (en) Method and device for recording communication message at communication terminal
CN110427453B (en) Data similarity calculation method, device, computer equipment and storage medium
WO2019200763A1 (en) Method for processing question answering request from agent, and electronic apparatus and computer-readable storage medium
CN110597965B (en) Emotion polarity analysis method and device for article, electronic equipment and storage medium
WO2014176939A1 (en) Apparatus and method for communication
CN111756644A (en) Hot spot current limiting method, system, equipment and storage medium
CN110969018A (en) Case description element extraction method, machine learning model acquisition method and device
CN110750626B (en) Scene-based task-driven multi-turn dialogue method and system
CN107465797B (en) Incoming call information display method and device for terminal equipment
CN116896582B (en) Multi-level organization-oriented real-time message pushing method, device and system
CN104765784A (en) Key words list maintenance method and system
US20130230248A1 (en) Ensuring validity of the bookmark reference in a collaborative bookmarking system
KR20180021152A (en) Information push method and apparatus
CN113051381B (en) Information quality inspection method, information quality inspection device, computer system and computer readable storage medium
CN110971759A (en) Processing method and device for unsubscribed short message and server
CN111639280A (en) Information sharing method and device
CN108182191B (en) Hotspot data processing method and device
CN113259214A (en) Driving data monitoring system
CN110209934A (en) Information-pushing method and relevant apparatus based on micro- Expression Recognition
CN111160989B (en) Information processing method and device

Legal Events

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