WO2022179118A1

WO2022179118A1 - Information push method, push robot, computer device, and storage medium

Info

Publication number: WO2022179118A1
Application number: PCT/CN2021/123851
Authority: WO
Inventors: 邓锐涛; 常向月; 刘云峰
Original assignee: 深圳追一科技有限公司
Priority date: 2021-02-26
Filing date: 2021-10-14
Publication date: 2022-09-01
Also published as: CN112565471A; CN112565471B

Abstract

An information push method, comprising: pushing target information to a target user, the target information being associated with the push knowledge base; receiving user reply information sent by the target user; determining whether the user reply information matches the push knowledge base; if yes, replying to the target user according to the push knowledge base; and if not, replying to the target user according to the candidate knowledge base, and sending a guide word related to the target information.

Description

Information push method, push robot, computer equipment and storage medium

This application claims the priority of the Chinese patent application filed on February 26, 2021 with the application number 2021102179406 and the application title is "information push method, push robot, computer equipment and storage medium", the entire contents of which are by reference Incorporated in this application.

technical field

The embodiments of the present application relate to the field of information processing, and in particular, to an information push method, a push robot, a computer device, and a storage medium.

Background technique

Information push is an important means of information dissemination in the information society. A more common push method is to directly push target information to target users one-way through software. This push method is relatively blunt and rude, and can only be pushed unilaterally by the robot.

SUMMARY OF THE INVENTION

The embodiments of the present application provide an information push method, a push robot, a computer device, and a storage medium.

A first aspect of the present application provides an information push method, which is applied to a push robot, where the push robot is configured with a push knowledge base and a candidate knowledge base, and the method includes:

Pushing target information to target users, where the target information is associated with the pushing knowledge base;

Receive the user reply information sent by the target user;

If the user reply information matches the push knowledge base, reply to the target user according to the push knowledge base;

If the user reply information does not match the push knowledge base, reply to the target user according to the candidate knowledge base, and send a guide language related to the target information.

A second aspect of the present application provides a push robot, including:

a push unit, configured to push target information to a target user, where the target information is associated with the push knowledge base;

a receiving unit, configured to receive the user reply information sent by the target user;

a first reply unit, configured to reply to the target user according to the push knowledge base when the user information matches the push knowledge base;

A second replying unit, configured to reply to the target user according to the candidate knowledge base when the user information does not match the push knowledge base, and send a guide language related to the target information.

A third aspect of the present application provides a computer device, comprising:

memory; one or more processors coupled to the memory; one or more application programs, wherein the one or more application programs are stored in the memory and configured to be executed by the one or more applications The processor executes the one or more application programs configured to perform the method of any one of the foregoing first aspects.

A fourth aspect of the present application provides one or more computer-readable storage media, where computer-readable instructions are stored in the computer storage media, and when the computer-readable instructions are executed on a computer, cause the computer to perform the aforementioned first aspect The method of any one.

The details of one or more embodiments of the application are set forth in the accompanying drawings and the description below. Other features, objects and advantages of the present application will become more apparent from the description, drawings and claims of the present application.

Description of drawings

In order to illustrate the technical solutions in the embodiments of the present application more clearly, the following briefly introduces the drawings that are used in the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, other drawings can also be obtained from these drawings without creative effort.

1 is a schematic diagram of a network architecture provided by an embodiment of the present application;

FIG. 2 is a schematic diagram of a network architecture provided by an embodiment of the present application

3 is a schematic diagram of an embodiment of an information push method provided by an embodiment of the present application;

4 is a schematic diagram of an embodiment of an information push method provided by an embodiment of the present application;

5 is a schematic diagram of an embodiment of an information push method provided by an embodiment of the present application;

6 is a schematic diagram of an embodiment of an information push method provided by an embodiment of the present application;

7 is a schematic diagram of an embodiment of an information push method provided by an embodiment of the present application;

8 is a schematic diagram of an embodiment of an information push method provided by an embodiment of the present application;

FIG. 9 is a schematic diagram of an embodiment of an information push method provided by an embodiment of the present application;

10 is a schematic diagram of an embodiment of an information push method provided by an embodiment of the present application;

11 is a schematic diagram of an embodiment of a push robot provided by an embodiment of the present application;

FIG. 12 is a schematic diagram of an embodiment of a computer device provided by an embodiment of the present application.

Detailed ways

The embodiments of the present application provide an information push method, a push robot, a computer device, and a storage medium, which are used to push target information to a target user.

The push robot mentioned in this application may be a part of the instant messaging system, or may be a separate device or program that is different from the instant messaging system. FIG. 1 and FIG. 2 are schematic diagrams of the network architecture of the information push method of the present application. In this application, the push robot may be a part of the instant messaging system as shown in FIG. 1 , and the target user accesses the communication system through a mobile phone, computer, or other terminal device. The user push robot can also be connected to the instant messaging system as a separate device or program, as shown in Figure 2, to interact with target users who access the communication system through mobile phones, computers, or other terminal devices. In the information push method of the method, the main task of the push robot is to push target information to the target user, and the push robot is at least configured with a push knowledge base and a candidate knowledge base. The push knowledge base is a specialized knowledge base configured for pushing target information, in which knowledge points related to the target information are stored. Compared with the push knowledge base, the candidate knowledge base contains more knowledge points unrelated to the target information, but in some embodiments, the candidate knowledge base may also contain part or all of the knowledge points of the push knowledge base.

FIG. 3 shows an embodiment of the information push method of the present application, and the information push method of the present application mainly includes the following steps:

301. Push target information to a target user, where the target information is associated with the push knowledge base;

The push robot sends the target information that needs to be pushed to the target users of the instant messaging platform, and the push information may include but not limited to text, pictures, links or other various forms of information and combinations thereof. Taking the instant messaging platform WeChat as an example, the WeChat official account can be used as a push robot to send a link with a picture to the target user's WeChat account as the target information.

302. Receive user reply information sent by the target user;

The target user sends user reply information to the push robot through the instant messaging system.

In one embodiment, the user reply information is usually text information, or if the user inputs voice information, the voice information needs to be converted, and the user reply information in text format is output by using a voice recognition technology.

In other embodiments, the user reply information may also be information in formats such as voice, pictures, etc., and is not limited to text formats.

303. Determine whether the user reply information matches the push knowledge base; if so, go to step 304; if not, go to step 305;

The push robot is configured with a push knowledge base, and the push knowledge base stores semantic information or keywords and key sentences related to the first target information. When receiving the user's reply information, the push robot will first try to compare the user's reply information with the information recorded in the push knowledge base, and use the push knowledge base to reply and answer. If the user's reply information matches the push knowledge base, the push robot will answer the user's reply information according to the guidance of the push knowledge base.

304. Reply to the target user according to the push knowledge base;

The push knowledge base is a knowledge base that matches the target information, and can accurately reply to the content related to the target information. When the push knowledge base matches the user's reply information, the push robot replies to the target user with content related to the target information according to the knowledge points recorded in the push knowledge base.

305. Reply to the target user according to the candidate knowledge base, and send a guide language related to the target information.

It can be understood that the introductory language is used to point to the target information, so as to guide attention to the target information.

Compared with the push knowledge base, the knowledge points in the candidate knowledge base are less relevant to the target information. When the push knowledge base does not match the user's reply information, the push robot replies to the target user through the candidate knowledge base, and sends a guide to the user to guide the target user to pay attention to the target information. For example, the target information pushed by the push robot is a mobile phone promotion on New Year's Day, and the user's reply information is a TV after-sales problem. If it does not match the target database, the user can reply to the TV after-sales problem according to the candidate knowledge base. After the push robot responds according to the candidate knowledge base, it will also send messages to the target users such as "The mobile phone promotion is limited to 3 days, do you want to think about it again", "This event is really affordable, I really don't want to try it out. What is the guide language to guide users to pay attention to the target information and consult the push robot about the content related to the target information.

It should be noted that the push robot of the present application can be configured with multiple candidate knowledge bases, and when the push robot replies to the target user according to the candidate knowledge bases, it can further select a suitable candidate knowledge base from the multiple candidate knowledge bases as a reply. The basis of the target user. The information push method of the present application considers that after the user receives the pushed target information, the content of the reply is likely to be content related to the target information, so the push knowledge base is first matched, which reduces the data processing amount of the push robot. At the same time, when the user's reply information is not related to the target information, a guiding language is sent to guide the user to pay attention to the target information, and the effect of pushing the target information is better.

FIG. 4 shows an improved solution of the embodiment shown in FIG. 3 , by adjusting the matching requirements of the push knowledge base to reduce the calculation amount of the push robot and improve the accuracy of the reply. Please refer to FIG. 4 for details. This embodiment includes:

401. Push target information to a target user, where the target information is associated with the push knowledge base;

402. Receive user reply information sent by the target user;

403. Determine whether the user reply information matches the push knowledge base; if so, go to step 404; if not, go to step 405;

404. Reply to the target user according to the push knowledge base;

405. Reply to the target user according to the candidate knowledge base, and send a guide language related to the target information.

Steps 401 to 405 in this embodiment are similar to steps 301 and 305 in the embodiment shown in FIG. 3 , and details are not repeated here.

What is different from the embodiment shown in FIG. 3 is that in this embodiment, different matching requirements are set for the push knowledge base and the candidate knowledge base configured by the push robot, and when the push starts, the matching requirement of the push knowledge base is low. For the candidate knowledge base, the user reply information is easier to match with the push knowledge base. The specific matching requirements of the push knowledge base and the candidate knowledge base can be set by those skilled in the art according to the target user's reply information and the knowledge base matching mechanism. For example, when pushing the target information to the target user, the matching requirement of the push knowledge base is set as the similarity score between the user's reply information and the push knowledge base reaches 6.0 points, and the matching requirement of the candidate knowledge base is that the user's reply information is similar to the push knowledge base. The degree score reached 7.0. In this way, the user's reply information will always be matched with the push knowledge base with lower matching requirements first, so as to avoid the situation where the user's reply information has a higher match with a candidate knowledge base, but the push robot uses the push knowledge base to reply. situation happens.

406. If it does not match, increase the matching requirement of the push knowledge base.

When the user's reply information does not match the push knowledge base, it means that the target user has sent reply content other than the target information, and the target user may not be interested in the target information. Therefore, the matching requirement of the pushed knowledge base can be appropriately increased, so that the matching probability between the user's reply information and the pushed knowledge base can be reduced. For example, the matching requirement of the push knowledge base is that the similarity score between the user's reply information and the push knowledge base reaches 6.0 points, and the similarity score between the user's reply information and the push knowledge base reaches 6.5 points, so that the target user can push the robot to the next time. When sending user reply information, the probability of matching to the push knowledge base becomes smaller.

On the basis of the embodiment shown in FIG. 4 , the present application also provides a further improvement scheme. For details, please refer to FIG. 5 . The embodiment of this improvement scheme includes:

501. Push target information to a target user, where the target information is associated with the push knowledge base;

502. Receive user reply information sent by the target user;

503. Determine whether the user reply information matches the push knowledge base; if it matches, go to step 504; if it does not match, go to step 505;

504. Reply to the target user according to the push knowledge base;

505. Reply to the target user according to the candidate knowledge base, and send a guide language related to the target information.

506. Improve the matching requirement of the pushed knowledge base.

Steps 501 to 506 in this embodiment are similar to steps 401 to 406 in the embodiment shown in FIG. 4 ,

507. When the matching requirement of the pushed knowledge base is higher than a preset value, deactivate the pushed knowledge base in the session with the target user.

For example, if the matching requirement of the push knowledge base is the requirement for the similarity score between the user's reply information and the push knowledge base, then, assuming the default value is 8, if the matching requirement of the push knowledge base is "the user's reply information and the push knowledge base" The similarity score of the library is required to reach 8.5". Since 8.5 is greater than 8, it is determined that the matching requirements of the push knowledge base meet the preset requirements. It can be understood that the matching requirement of the push knowledge base may also be a requirement of the number of information matched in the push knowledge base by the user's reply information, which is not limited thereto.

Due to the existence of step 506, the push robot will continuously improve the matching requirements of the push knowledge base after receiving the user's reply information that does not match the target knowledge base for many times, which may eventually make the matching requirements of the push database higher than the matching requirements of the candidate knowledge base. higher. The reason for this is that the target user is really not interested in the target information, and it is meaningless to continue to try to guide the target user to topics related to the target information when talking with the push robot about topics other than the target information. Therefore, the push robot can deactivate the push database related to the target information, directly use the candidate knowledge base to communicate with the target users, and at the same time no longer send introductory words. The reply efficiency of the robot reduces the calculation amount of the push robot.

FIG. 6 shows an improved solution of the embodiment shown in FIG. 3 , in which another modified way of pushing the knowledge base matching requirement is provided. Please refer to Fig. 6 for details. The embodiment of the improved scheme includes:

601. Push target information to a target user, where the target information is associated with the push knowledge base;

602. Receive user reply information sent by the target user;

603. Determine whether the user reply information matches the push knowledge base; if so, go to step 604; if not, go to step 605;

604. Reply to the target user according to the push knowledge base;

605. Reply to the target user according to the candidate knowledge base, and send a guide language related to the target information.

Steps 601 to 605 in this embodiment are similar to steps 301 to 305 in the embodiment shown in FIG. 3 , and details are not repeated here.

The difference from the embodiment shown in FIG. 3 is that in this embodiment, different matching requirements are set for the push knowledge base and the candidate knowledge base configured for the push robot, and when the push starts, the push knowledge base requirement is lower than that of the push knowledge base. Candidate knowledge base, user reply information is easier to match with the push knowledge base. The specific matching requirements of the push knowledge base and the candidate knowledge base can be set by those skilled in the art according to the target user's reply information and the knowledge base matching mechanism. For example, when pushing the target information to the target user, the matching requirement of the push knowledge base is set as the similarity score between the user's reply information and the push knowledge base reaches 6.0 points, and the matching requirement of the candidate knowledge base is that the user's reply information is similar to the push knowledge base. The degree score reached 7.0. In this way, the user's reply information will always be matched with the push knowledge base with lower matching requirements first, so as to avoid the situation where the user's reply information has a higher match with a candidate knowledge base, but the push robot uses the push knowledge base to reply. situation happens.

606. Calculate the sent time of the target information; improve the matching requirement of the push knowledge base according to the sent time of the target information;

When the push robot has just pushed the target information to the target user, the user reply information sent by the user is likely to be related to the target information. At this time, the matching requirements of the push knowledge base are still relatively low, and the user reply is easy to match with the push knowledge base. The robot uses the push knowledge base to send a reply to the target user, which can answer the target user's consulting questions about the target information. For example, after the target user receives the target information of the pushed mobile phone promotion, and is not interested, the target user will put the target information on hold without replying. After three days, the target user wants to inquire about the after-sales service of the push robot TV. The user reply information sent at this time is likely to be irrelevant to the target message. It should be avoided to match the user reply information with the push knowledge base. Therefore, to add a mechanism for the push knowledge base to grow over time, the matching requirements also increase. The longer the user replies from the target information, the higher the matching requirements of the push knowledge base, and the more difficult it is for the user to reply to the push. Knowledge base matching. And if the target information is news, limited-time promotions or other time-sensitive information, its timeliness will decrease as time increases.

On the basis of the embodiment shown in FIG. 6 , the present application also provides a further improvement scheme. For details, please refer to FIG. 7 . The embodiment of this improved scheme includes:

701. Push target information to a target user, where the target information is associated with the push knowledge base;

702. Receive user reply information sent by the target user;

703. Determine whether the user reply information matches the push knowledge base; if it matches, go to step 704;

704. Reply to the target user according to the push knowledge base;

705. Reply to the target user according to the candidate knowledge base, and send a guide language related to the target information.

706. Calculate the sent time of the target information; improve the matching requirement of the push knowledge base according to the sent time of the target information;

Steps 701 to 706 in this embodiment are similar to steps 601 to 606 in the embodiment shown in FIG. 6 , and details are not repeated here.

707. When the matching requirement of the pushed knowledge base is higher than a preset value, deactivate the pushed knowledge base in the session with the target user.

Due to the existence of step 706, after the push robot pushes the target information, if the target user does not reply immediately, as time increases, the probability of the target user replying to the relevant content of the target information will also decrease. As time grows, the matching requirements of the push database will eventually be higher than those of the candidate knowledge base. It is meaningless to continue to try to guide target users to topics related to the target information, and if the target news is news or limited-time promotions Or other time-sensitive information, and the target information also loses its timeliness after a long period of time. Therefore, when the matching requirement of the push knowledge base is higher than a certain preset value, the push robot can deactivate the push database related to the target information, directly use the candidate knowledge base to communicate with the target user, and no longer send the guide words. It is convenient for the push robot to reply to the information of other users unrelated to the target information, which improves the reply efficiency of the push robot and reduces the calculation amount of the push robot.

Based on the embodiment shown in FIG. 3 , the present application provides another improved solution. Please refer to FIG. 8 for details. Examples of the improved solution include:

801. Push target information to a target user, where the target information is associated with the push knowledge base;

802. Receive user reply information sent by the target user;

803. Determine whether the user reply information matches the push knowledge base; if it matches, go to step 804; if it does not match, go to steps 805 and 806;

804. Reply to the target user according to the push knowledge base;

805. Reply to the target user according to the candidate knowledge base, and send a guide language related to the target information.

806. Count the number of irrelevant information in the user reply information that does not match the push knowledge base;

When the push robot confirms that the user's reply information does not match the push knowledge base, it will calculate the number of mismatched push knowledge bases in the user's reply information replied by the target user in the recent period. For example, after receiving a target message of a mobile phone promotion, the target user is not interested in the mobile phone promotion, but wants to consult the after-sales problem of the TV. In this case, there will be user reply information unrelated to the target message in the chat content between the push bot and the target user. Count the number of words or pieces of replies from users who are not related to the target message.

807. When the quantity of irrelevant information exceeds a preset quantity upper limit, deactivate the pushed knowledge base in the session with the target user.

When the push robot counts that there are multiple user reply information unrelated to the target message in the reply information of the target user, it means that the main intention of the target user is not to understand the content related to the target message, and the push knowledge base cannot accurately reply to the target user. At this time, you can disable the push knowledge base, and stop sending the guide language, and directly use the candidate knowledge base to communicate with the target user, which is convenient for the push robot to reply to the user's reply information unrelated to the target information, which improves the reply efficiency of the push robot and reduces the The calculation amount of the push robot.

Similarly, in this embodiment, the matching requirement of the push knowledge base can also be set lower than that of the candidate knowledge base, so as to avoid that the user reply information has a higher matching degree with a certain candidate knowledge base, but the push robot uses the push knowledge base Reply happens.

Based on the embodiment shown in FIG. 3 , as a preferred implementation manner, in the embodiment shown in FIG. 9 , the present application specifically shows a knowledge base matching method when a push robot replies to a user's reply information. Please refer to FIG. 9 for details. This embodiment includes:

901. Push target information to a target user, where the target information is associated with the push knowledge base;

902. Receive user reply information sent by the target user;

903. Evaluate the similarity score of the user's reply information according to the knowledge points in the pushed knowledge base, and determine whether the similarity score meets the matching requirements of the pushed knowledge base;

The push robot compares the user's reply information with the knowledge points in the target knowledge base, scores according to the similarity between the user's reply information and the knowledge points in the target knowledge base, and evaluates the similarity between the user's reply information and the target knowledge base, and then evaluates the similarity between the user's reply information and the target knowledge base. When the information score is higher than the matching requirements of the push knowledge base, confirm that the user's reply information matches the target knowledge base; when the user's reply information score is lower than the matching requirements of the push knowledge base, confirm that the user's reply information matches the target knowledge base. For example, the target knowledge base mainly stores the relevant knowledge points of a certain mobile phone promotion. When the user sends "how long is the mobile phone battery warranty", "what gift to buy now" and other sentences that are strongly related to mobile phone promotion, you can push the knowledge base If matching knowledge points are found, the similarity score will be higher. If the reply is "What's the weather like today" or "Is there a traffic jam on Renmin Road?" that has little to do with mobile phone promotions, and there is no relevant knowledge point in the push knowledge base, the similarity score obtained will be lower.

The push knowledge base has a fixed or variable matching requirement. When the push robot evaluates the matching degree between the user's reply information and the push knowledge base through similarity score, the matching requirement of the push knowledge base can be a specific score. If the similarity score of the user's reply information meets the score requirement of the pushed knowledge base, it is determined that the user's reply information matches the pushed knowledge base, and step 904 is executed. If the similarity score of the user's reply information does not meet the score requirement of the push knowledge base, it is determined that the user's reply information does not match the push knowledge base, and step 905 is executed.

904. Reply to the target user according to the push knowledge base;

905. Reply to the target user according to the candidate knowledge base, and send a guide language related to the target information.

It can be understood that, in the embodiment shown in FIG. 9 , the similarity score for the user reply information is only a preferred implementation for judging the matching between the user reply information and the knowledge base. The knowledge base matching method when the push robot replies to the user's reply information in this embodiment can be combined with any other embodiment without conflict. Those skilled in the art can set specific matching methods and matching requirements, such as keyword screening, clustering algorithm and other methods, by referring to the existing matching principles between text and knowledge bases, which are not specifically limited here.

Fig. 10 shows another improved solution of the embodiment shown in Fig. 3, which specifically defines the way of pushing target information to target users. Please refer to Fig. 10 for details. The embodiment of this improved solution includes:

1001. Push target information to a target user through a conversational interface, where the conversational interface includes an input box for the target user to input text or voice messages;

When the push robot pushes the target information to the target user, it will send the target information to the target user through a designated session interface. The content of the target information may include but not limited to text, pictures, links or other forms of information and combinations thereof. . The conversation interface also includes an input field. After receiving the target information in the conversation window, the target user can directly input the user reply information in the form of text or voice from the input field to reply to the push robot, so as to consult the push robot about the target information. content. The subsequent interaction process between the target user and the push robot can also be completed in this conversational interface.

1002. Receive user reply information sent by the target user;

1003. Determine whether the user reply information matches the push knowledge base; if so, go to step 1004; if not, go to step 1005;

1004. Reply to the target user according to the push knowledge base;

1005. Reply to the target user according to the candidate knowledge base, and send a guide language related to the target information.

Steps 1002 to 1005 in this embodiment are similar to steps 302 to 305 in the embodiment shown in FIG. 3 , and details are not repeated here.

The process of pushing target information on the session interface in the embodiment shown in FIG. 10 can be combined with any other embodiment without conflict, thereby forming more new embodiments, which also fall within the protection scope of the present invention.

Corresponding to the above information push method, in one embodiment, a push robot is also presented, which can execute any information push method shown in FIG. 3 to FIG. 10 . The push robot includes:

Pushing unit 1101, configured to push target information to target users, where the target information is associated with the push knowledge base;

A receiving unit 1102, configured to receive user reply information sent by the target user;

a first replying unit 1104, configured to reply to the target user according to the push knowledge base when the user information matches the push knowledge base;

The second replying unit 1105 is configured to reply to the target user according to the candidate knowledge base when the user information does not match the push knowledge base, and send a guide language related to the target information.

In some embodiments, before the target information is pushed to the target user, the matching requirement of the pushed knowledge base is lower than that of the candidate knowledge base.

In some embodiments, the push robot may further include:

An improving unit 1106 is configured to improve the matching requirement of the push knowledge base when the user reply information does not match the push knowledge base.

In some embodiments, the improving unit 1106 is configured to: calculate the sent time of the target information; and improve the matching requirement of the push knowledge base according to the sent time of the target information.

In some embodiments, the push robot further includes:

The deactivation unit 1107 is configured to deactivate the pushed knowledge base in the session with the target user when the matching requirement of the pushed knowledge base is higher than a preset value.

In some embodiments, the disabling unit 1107 is used to:

When the user reply information does not match the push knowledge base, count the number of irrelevant information in the user reply information that does not match the push knowledge base; when the quantity of the irrelevant information exceeds a preset upper limit, Deactivate the push knowledge base in the session with the target user.

In some embodiments, as shown in FIG. 11, it also includes:

Judging unit 1103, configured to evaluate the similarity score of the user's reply information according to the knowledge points in the push knowledge base, and if the similarity score meets the matching requirements of the push knowledge base, confirm that the user's reply information matches the the push knowledge base; if the similarity score does not meet the matching requirements of the push knowledge base, confirm that the user reply information does not match the push knowledge base.

In some embodiments, the pushing unit 1101 is specifically configured to: push the target information to the target user through a conversation interface, where the conversation interface includes an input box for the target user to input text or voice messages.

FIG. 12 also shows a computer device, the computer device 1200 may include one or more central processing units (CPU) 1201 and a memory 1205 in which one or more application programs or data are stored . Among them, the memory 1205 may be volatile storage or persistent storage. A program stored in memory 1205 may include one or more modules, each of which may include a series of computer-readable instructions that operate on a computer device. Still further, the central processing unit 1201 may be configured to communicate with the memory 1205 to execute on the computer device 1200 a series of computer readable instructions in the memory 1205. Computer device 1200 may also include one or more power supplies 1202, one or more wired or wireless network interfaces 1203, one or more input and output interfaces 1204, and/or, one or more operating systems, such as Windows Server™, Mac OS XTM, UnixTM, LinuxTM, FreeBSDTM, etc. The central processing unit 1201 can execute the information pushing method in any of the foregoing embodiments shown in FIG. 3 to FIG. 10 , and details are not repeated here.

The present application also provides one or more computer-readable storage media, where computer-readable instructions are stored in the computer storage media, and when the computer-readable instructions are executed on a computer, cause the computer to perform the operations shown in FIGS. 1 to 6 . Any information push method in the example embodiment.

It should be noted that those skilled in the art can define the matching requirements of the push knowledge base and the candidate knowledge base by themselves according to the specific matching mechanism of the push robot to the user's reply information and the knowledge base, which is not specifically limited here.

Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the system, device and unit described above may refer to the corresponding process in the foregoing method embodiments, which will not be repeated here.

In the several embodiments provided in this application, it should be understood that the disclosed system, apparatus and method may be implemented in other manners. For example, the apparatus embodiments described above are only illustrative. For example, the division of the units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components may be combined or Can be integrated into another system, or some features can be ignored, or not implemented. On the other hand, the shown or discussed mutual coupling or direct coupling or communication connection may be through some interfaces, indirect coupling or communication connection of devices or units, and may be in electrical, mechanical or other forms.

The units described as separate components may or may not be physically separated, and components displayed as units may or may not be physical units, that is, may be located in one place, or may be distributed to multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution in this embodiment.

In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically alone, or two or more units may be integrated into one unit. The above-mentioned integrated units may be implemented in the form of hardware, or may be implemented in the form of software functional units.

The integrated unit, if implemented in the form of a software functional unit and sold or used as an independent product, may be stored in a computer-readable storage medium. Based on this understanding, the technical solutions of the present application can be embodied in the form of software products in essence, or the parts that contribute to the prior art, or all or part of the technical solutions, and the computer software products are stored in a storage medium , including several computer-readable instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, read-only memory), random access memory (RAM, random access memory), magnetic disk or optical disk and other media that can store program codes .

Claims

An information push method, characterized in that, it is applied to a push robot, and the push robot is configured with a push knowledge base and a candidate knowledge base, and the method includes:

Pushing target information to target users, where the target information is associated with the pushing knowledge base;

Receive the user reply information sent by the target user;

If the user reply information matches the push knowledge base, reply to the target user according to the push knowledge base;

If the user reply information does not match the push knowledge base, reply to the target user according to the candidate knowledge base, and send a guide language for guiding attention to the target information.
The information pushing method according to claim 1, wherein before the target information is pushed to the target user, the matching requirement of the pushed knowledge base is lower than that of the candidate knowledge base.
The information push method according to claim 2, wherein the method further comprises:

If the user reply information does not match the push knowledge base, the matching requirement of the push knowledge base is increased.
The method for pushing information according to claim 2, wherein after the target information is pushed to the target user, the method further comprises:

calculating the sent time of the target information;

According to the sent time of the target information, the matching requirement of the push knowledge base is improved.
The information push method according to claim 3 or 4, wherein the method further comprises:

When the matching requirement of the push knowledge base is higher than the preset value, the push knowledge base is deactivated in the session with the target user.
The information push method according to claim 1, wherein the method further comprises:

If the user reply information does not match the push knowledge base, count the number of irrelevant information in the user reply information that does not match the push knowledge base;

When the amount of irrelevant information exceeds a preset upper limit, the push knowledge base is deactivated in the session with the target user.
The information push method according to claim 5 or 6, wherein the method further comprises:

With respect to the user reply information received after the push knowledge base is deactivated, reply to the target user according to the candidate knowledge base, and do not send the introductory phrase.
The information push method according to claim 1, wherein the method further comprises:

Evaluate the similarity score of the user's reply information according to the knowledge points in the pushed knowledge base;

If the similarity score meets the matching requirement of the push knowledge base, confirming that the user reply information matches the push knowledge base;

If the similarity score does not meet the matching requirement of the push knowledge base, it is confirmed that the user reply information does not match the push knowledge base.
The information pushing method according to any one of claims 1 to 8, wherein pushing the target information to the target user comprises:

The target information is pushed to the target user through a conversation interface, and the conversation interface includes an input box for the target user to input the user reply information.
A push robot, characterized in that it includes:

a push unit, configured to push target information to a target user, where the target information is associated with the push knowledge base;

a receiving unit, configured to receive the user reply information sent by the target user;

a first reply unit, configured to reply to the target user according to the push knowledge base when the user information matches the push knowledge base;

A second replying unit, configured to reply to the target user according to the candidate knowledge base when the user information does not match the push knowledge base, and send a guide language for guiding attention to the target information.
The push robot according to claim 10, wherein before the target information is pushed to the target user, the matching requirement of the push knowledge base is lower than that of the candidate knowledge base.
The push robot according to claim 11, further comprising:

An improving unit, configured to improve the matching requirement of the push knowledge base when the user reply information does not match the push knowledge base.
The push robot according to claim 11, further comprising:

An improving unit, configured to calculate the sent time of the target information; improve the matching requirement of the push knowledge base according to the sent time of the target information.
The push robot according to claim 12 or 13, further comprising:

The deactivation unit is configured to deactivate the pushed knowledge base in the session with the target user when the matching requirement of the pushed knowledge base reaches a preset requirement restriction condition.
The push robot according to claim 10, further comprising:

A deactivation unit, configured to count the number of irrelevant information in the user reply information that does not match the push knowledge base when the user reply information does not match the push knowledge base; when the amount of irrelevant information exceeds a predetermined amount When the upper limit is set, the push knowledge base is deactivated in the session with the target user.
The push robot according to claim 14 or 15, wherein the second reply unit is further configured to reply to the user reply information received after the push knowledge base is deactivated, according to the candidate knowledge base. describe the target user and do not send the introductory phrase.
The push robot according to claim 10, further comprising:

A judgment unit, configured to evaluate the similarity score of the user's reply information according to the knowledge points in the push knowledge base; if the similarity score meets the matching requirements of the push knowledge base, confirm that the user's reply information matches all the push knowledge base; if the similarity score does not meet the matching requirements of the push knowledge base, confirm that the user reply information does not match the push knowledge base.
The push robot according to any one of claims 10 to 17, wherein the push unit is specifically used for:

The target information is pushed to the target user through a conversation interface, and the conversation interface includes an input box for the target user to input text or voice messages.
A computer equipment, characterized in that, comprising:

memory; one or more processors coupled to the memory; one or more application programs, wherein the one or more application programs are stored in the memory and configured to be executed by the one or more applications The processor executes the one or more application programs configured to perform the method of any of claims 1-9.
One or more computer-readable storage media, characterized in that, computer-readable instructions are stored in the computer-readable storage media, and when the computer-readable instructions are executed on a computer, the computer is made to perform the execution of claims 1-9. The method of any one.