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

Abstract

The embodiment of the present application discloses an information push method, a push robot, a computer device and a storage medium, which are used to push a specified first target information to a target user. The method of the embodiment of the present application includes: pushing the first target information to the target user through a conversation interface; receiving a user reply information sent by the target user through the conversation interface; judging whether the user reply information matches the target knowledge base; if the user reply information matches the target knowledge base, replying to the user reply information in the conversation interface according to the target knowledge base.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of information processing, and specifically to an information push method, a push robot, a computer device, and a storage medium. Background Art

[0002] Information push is an important means of information dissemination in the information society. In various current software, you can receive push information in the form of text, pictures, links or other forms. These push information are relatively brief to facilitate user reading. If users are interested in the push information and want to know more relevant information, they need to search for it themselves or read through the links attached to the push information. The steps to obtain relevant information are not simple enough. Summary of the invention

[0003] The embodiments of the present application provide an information push method, a push robot, a computer device and a storage medium for pushing information to a target user.

[0004] A first aspect of the present application provides an information push method, which is applied to a push robot, wherein the push robot is configured with a target knowledge base, and the method comprises:

[0005] Pushing the first target information to the target user through the conversation interface;

[0006] Receiving user reply information sent by the target user through the conversation interface;

[0007] Determining whether the user reply information matches the target knowledge base;

[0008] If the user reply information matches the target knowledge base, the user reply information is replied in the conversation interface according to the target knowledge base.

[0009] Optionally, after determining whether the user reply information matches the target knowledge base, the method further includes:

[0010] If the user reply information does not match the target knowledge base, the user reply information is forwarded to the candidate robot, so that the candidate robot configured with the candidate knowledge base replies to the target user in the conversation interface.

[0011] Optionally, the matching requirement of the target knowledge base is lower than that of the candidate knowledge base.

[0012] Optionally, the determining whether the user reply information matches the target knowledge base includes:

[0013] Evaluating the similarity score of the user's reply information according to the knowledge points in the target knowledge base;

[0014] If the similarity score meets the matching requirement of the target knowledge base, confirming that the user reply information matches the target knowledge base;

[0015] If the similarity score does not meet the matching requirement of the target knowledge base, it is confirmed that the user reply information does not match the target knowledge base.

[0016] Optionally, the target knowledge base includes a first target knowledge base and a second target knowledge base;

[0017] The pushing unit is further configured to push the first target information to the target user through the conversation interface, and push the second target information to the target user through the conversation interface;

[0018] The judging unit is specifically used for:

[0019] Determining whether the user reply information matches the first target knowledge base;

[0020] When the user reply information does not match the first target knowledge base, determining whether the user reply information matches the second target knowledge base;

[0021] The reply unit is specifically used for:

[0022] When the user reply information matches the first target knowledge base, replying to the user reply information in the conversation interface according to the first target knowledge base;

[0023] When the user reply information matches the second target knowledge base, the user reply information is replied in the conversation interface according to the second target knowledge base.

[0024] Optionally, the target knowledge base includes a first target knowledge base and a second target knowledge base;

[0025] The push unit is also used for:

[0026] Before pushing the first target information to the target user through the conversation interface, pushing the second target information to the target user through the conversation interface;

[0027] The judging unit is specifically used to judge whether the user reply information matches the first target knowledge base or the second target knowledge base;

[0028] The reply unit is specifically configured to reply to the user reply information in the conversation interface according to the target knowledge base with the highest matching degree when the user reply information matches the first target knowledge base or the second target knowledge base.

[0029] Optionally, the matching requirement of the second target knowledge base is higher than that of the target knowledge base.

[0030] Optionally, the matching requirement of the second target knowledge base is the same as the matching requirement of the target knowledge base.

[0031] Optionally, the matching requirement of the second target knowledge base is positively correlated with the sending time of the second target information.

[0032] Optionally, the method further includes:

[0033] If the user reply information does not match the target knowledge base, a guide statement is sent to the target user, and the content of the guide statement is associated with the first target information.

[0034] Optionally, the sending a guide message to the target user includes:

[0035] Calculate the number of times the introductory message has been sent;

[0036] According to the sending times, a guiding sentence is selected from a plurality of preset guiding sentences and sent to the target user.

[0037] Optionally, the method further includes:

[0038] If the user reply information does not match the target knowledge base, the number of times the guide sentence has been sent is calculated;

[0039] When the number of times the guide message has been sent reaches a preset upper limit, the conversation with the target user is handed over to the candidate robot.

[0040] Corresponding to the above-mentioned information push method, the second aspect of the present application provides a push robot, wherein the push robot is configured with a target knowledge base, and the push robot comprises:

[0041] A push unit, configured to push the first target information to the target user through the conversation interface;

[0042] A receiving unit, configured to receive user reply information sent by a target user through the conversation interface;

[0043] A judging unit, used to judge whether the user reply information matches the target knowledge base;

[0044] A reply unit is used to reply the user reply information in the conversation interface according to the target knowledge base when the user reply information matches the target knowledge base.

[0045] Optionally, the push robot further includes:

[0046] A forwarding unit is used to forward the user reply information to the candidate robot when the user reply information does not match the target knowledge base, so that the candidate robot configured with the candidate knowledge base replies to the target user in the conversation interface.

[0047] Optionally, the matching requirement of the target knowledge base is lower than that of the candidate knowledge base.

[0048] Optionally, the judging unit is specifically configured to:

[0049] Evaluating the similarity score of the user's reply information according to the knowledge points in the target knowledge base;

[0050] If the similarity score meets the matching requirement of the target knowledge base, confirming that the user reply information matches the target knowledge base;

[0051] If the similarity score does not meet the matching requirement of the target knowledge base, it is confirmed that the user reply information does not match the target knowledge base.

[0052] Optionally, the target knowledge base includes a first target knowledge base and a second target knowledge base;

[0053] The pushing unit is further configured to push the first target information to the target user through the conversation interface, and push the second target information to the target user through the conversation interface;

[0054] The judging unit is specifically used for:

[0055] Determining whether the user reply information matches the first target knowledge base;

[0056] The reply unit is specifically used for:

[0057] When the user reply information matches the first target knowledge base, replying to the user reply information in the conversation interface according to the first target knowledge base;

[0058] When the user reply information does not match the first target knowledge base, determining whether the user reply information matches the second target knowledge base;

[0059] When the user reply information matches the second target knowledge base, the user reply information is replied in the conversation interface according to the second target knowledge base.

[0060] Optionally, the target knowledge base includes a first target knowledge base and a second target knowledge base;

[0061] The pushing unit is further used for: before pushing the first target information to the target user through the conversation interface, pushing the second target information to the target user through the conversation interface;

[0062] The determining whether the user reply information matches the target knowledge base includes:

[0063] Determining whether the user reply information matches the first target knowledge base or the second target knowledge base;

[0064] The reply unit is specifically configured to reply to the user reply information in the conversation interface according to the target knowledge base with the highest matching degree when the user reply information matches the first target knowledge base or the second target knowledge base.

[0065] Optionally, the matching requirement of the second target knowledge base is higher than that of the target knowledge base.

[0066] Optionally, the matching requirement of the second target knowledge base is the same as the matching requirement of the target knowledge base.

[0067] Optionally, the matching requirement of the second target knowledge base is positively correlated with the sending time of the second target information.

[0068] Optionally, the push robot further includes:

[0069] A guiding unit is used to send guiding words to the target user when the user reply information does not match the target knowledge base, and the content of the guiding words is associated with the first target information.

[0070] Optionally, the guiding unit is specifically used for:

[0071] Calculate the number of times the introductory message has been sent;

[0072] According to the sending times, a guiding sentence is selected from a plurality of preset guiding sentences and sent to the target user.

[0073] Optionally, the push robot further includes:

[0074] A calculation unit, used for calculating the number of times the guide words have been sent when the user reply information does not match the target knowledge base;

[0075] The transfer unit is used to transfer the conversation with the target user to a candidate robot when the number of times the guide message has been sent reaches a preset upper limit.

[0076] A third aspect of the present application provides a computer device, the computer device comprising:

[0077] A memory; one or more processors coupled to the memory; one or more applications, wherein the one or more applications are stored in the memory and configured to be executed by the one or more processors, and the one or more applications are configured to execute the method described in any one of the first aspects.

[0078] A fourth aspect of the present application provides a computer storage medium, wherein the computer storage medium stores instructions, and when the instructions are executed on a computer, the computer executes any one of the methods of the first aspect.

[0079] It can be seen from the above technical scheme that the embodiments of the present application have the following advantages: the present invention directly pushes information to users in the form of a conversation, and configures a push robot related to the pushed content in the conversation. If the user wants to know more relevant information after receiving the push information, he can directly learn about the relevant information of the push content in the form of a chat in the conversation, which simplifies the steps for the user to understand the push information. BRIEF DESCRIPTION OF THE DRAWINGS

[0080] Figure 1A This is a schematic diagram of an application scenario of the information push method provided by this application;

[0081] Figure 1B This is a schematic diagram of an application scenario of the information push method provided by this application;

[0082] Figure 2 It is a schematic diagram of an embodiment of the information push method provided by the present application;

[0083] Figure 3 It is a schematic diagram of an embodiment of the information push method provided by the present application;

[0084] Figure 4 It is a schematic diagram of an embodiment of the information push method provided by the present application;

[0085] Figure 5 It is a schematic diagram of an embodiment of the information push method provided by the present application;

[0086] Figure 6 It is a schematic diagram of an embodiment of the information push method provided by the present application;

[0087] Figure 7 It is a schematic diagram of an embodiment of the information push method provided by the present application;

[0088] Figure 8 This is a schematic diagram of an embodiment of a push robot provided by the present application;

[0089] Fig. 9 It is a schematic diagram of an embodiment of a computer device provided by the present application. DETAILED DESCRIPTION

[0090] The embodiments of the present application provide an information push method, a push robot, a computer device and a storage medium for guiding users to understand the target information being pushed.

[0091] Figure 1A and Figure 1B Two possible application scenarios of this embodiment are shown. In this application, the push robot can be as follows: Figure 1A A part of the instant messaging system in the target user to access the communication system through mobile phones, computers, or other terminal devices. Push robots can also Figure 1B As shown, as a separate device or program connected to the instant messaging system, it interacts with the target user who is connected to the communication system through a mobile phone, computer, or other terminal device. In this information push method, the main task of the push robot is to push the target information to the target user, and the push robot is configured with at least a target knowledge base and a candidate knowledge base. The target knowledge base is a special knowledge base configured for pushing target information, in which knowledge points related to the target information are stored. Compared with the target knowledge base, the candidate knowledge base contains more knowledge points that are not related to the target information, but in some embodiments, the candidate knowledge base may also contain some or all of the knowledge points of the target knowledge base.

[0092] Figure 2 An embodiment of the information push method of the present application is presented, and the embodiment mainly includes the following steps:

[0093] 201. Pushing first target information to a target user through a conversation interface;

[0094] The push robot sends a first target message to the terminal device of the target user through the conversation interface of the instant messaging system for push. Taking the instant messaging platform WeChat as an example, the WeChat public account can be used as a push robot to send a link with a picture as the first target message to the WeChat account of the target user through the WeChat conversation interface. The conversation interface may have existed before, for example, the target user and the push robot have had a conversation communication before, and the push robot can continue to send the first target message to the target user through the last conversation interface. Alternatively, the conversation interface can also be a new conversation interface specially created by the push robot to push the first target message to the target user.

[0095] It should be noted that in the present application, the push robot can be a part of the instant messaging system, running on the same server or server group as the instant messaging system; or the push robot can be a separate computer program different from the instant messaging system, running on a server or server group outside the communication system.

[0096] 202. Receive user reply information sent by the target user through the conversation interface;

[0097] After receiving the first target information, the target user can browse the general content of the first target information from the conversation interface, or click on a link therein for more detailed browsing. In the present application, the target user receives the push information through the conversation interface, and the conversation interface itself provides a special input box or input bar for the target user to enter the user reply information. After the user enters the user reply information in the input bar or input box and sends it, the push robot receives the user reply information.

[0098] 203. Determine whether the user reply information matches the target knowledge base; if so, execute step 204;

[0099] The push robot is configured with a target knowledge base, which stores semantic information or keywords and key sentences related to the first target information. When receiving a user reply message, the push robot will first try to compare the user reply message with the information recorded in the target knowledge base, and use the target knowledge base to reply and answer. If the user reply message matches the target knowledge base, the push robot will answer the user reply message according to the guidance of the target knowledge base.

[0100] Preferably, the match between the user reply information and the target knowledge base can be determined by the push robot scoring the user reply information. For example, the target knowledge base can store multiple dialogue sentences and corresponding reply contents as knowledge points. When the push robot receives the user reply information, it performs a similarity assessment and scores each dialogue in the first knowledge base. The higher the score, the more similar the user reply information is to the dialogue sentence. If the scoring result exceeds a preset value, the user reply information is deemed to match the target knowledge base, otherwise it is deemed not to match. It is understandable that the push robot can also be combined with one or more other methods to evaluate whether the matching of the user reply information and the target knowledge base meets the matching requirements, such as keyword screening, semantic category matching, etc., which are not specifically limited here.

[0101] 204. Reply to the user response information in the conversation interface according to the target knowledge base.

[0102] When the target knowledge base matches the previous user reply information, the push robot replies to the target user with content related to the first target information in the conversation interface based on the information recorded in the target knowledge base.

[0103] In this embodiment, the communication between the push robot and the user is carried out in the same conversation interface. After roughly browsing the pushed first target information, the user can directly inquire about the relevant content of the first target information in the conversation interface. The push robot quickly answers the question, making it convenient for the user to obtain relevant content of the first target information, thereby improving communication efficiency.

[0104] based on Figure 2 The present application also provides another embodiment of the information push method. Figure 3 , the steps of this implementation include:

[0105] 301. Pushing first target information to a target user through a conversation interface;

[0106] The push robot sends a first target message to the terminal device of the target user through the conversation interface of the instant messaging system for push. Taking the instant messaging platform WeChat as an example, the WeChat public account can be used as a push robot to send a link with a picture as the first target message to the WeChat account of the target user through the WeChat conversation interface. The conversation interface may have existed before, for example, the target user and the push robot have had a conversation communication before, and the push robot can continue to send the first target message to the target user through the last conversation interface. Alternatively, the conversation interface can also be a new conversation interface specially created by the push robot to push the first target message to the target user.

[0107] It should be noted that in the present application, the push robot can be a part of the instant messaging system, running on the same server or server group as the instant messaging system; or the push robot can be a separate computer program different from the instant messaging system, running on a server or server group outside the communication system.

[0108] 302. Receive user reply information sent by the target user through the conversation interface;

[0109] After receiving the first target information, the target user can browse the general content of the first target information from the conversation interface, or click on a link therein for more detailed browsing. In the present application, the target user receives the push information through the conversation interface, and the conversation interface itself provides a special input box or input bar for the target user to enter the user reply information. After the user enters the user reply information in the input bar or input box and sends it, the push robot receives the user reply information.

[0110] 303. Determine whether the user reply information matches the target knowledge base; if it matches, execute step 304; otherwise, execute step 305;

[0111] The push robot is configured with a target knowledge base, which stores semantic information or keywords and key sentences related to the first target information. When receiving the user reply information, the push robot will first try to compare the user reply information with the information recorded in the target knowledge base, and use the target knowledge base to reply and answer. If the user reply information matches the target knowledge base, the push robot will answer the user reply information based on the knowledge points recorded in the target knowledge base. If it does not match, execute step 305.

[0112] In this embodiment, there are one or more candidate robots different from the push robot, and the candidate robots are configured with a candidate knowledge base. If there are multiple candidate robots, they can each correspond to a different candidate knowledge base. The candidate knowledge base stores information that has a low correlation with the first target information or has no correlation at all. The candidate robot can reply to the user's reply information based on the information in the candidate knowledge base.

[0113] Preferably, the match between the user reply information and the target knowledge base can be determined by the push robot scoring the user reply information. For example, the target knowledge base can store multiple dialogue sentences and corresponding reply contents as knowledge points. When the push robot receives the user reply information, it scores the similarity assessment with each dialogue in the first knowledge base. The higher the score, the more similar the user reply information is to the dialogue sentence. If the scoring result exceeds a preset value, the user reply information is deemed to match the target knowledge base, otherwise it is not matched. It is understandable that the push robot can also comprehensively evaluate whether the match between the user reply information and the target knowledge base meets the matching requirements through a combination of one or more other methods, such as keyword screening, semantic category matching, etc., which are not specifically limited here.

[0114] 304. Replying to the user reply information in the conversation interface according to the target knowledge base;

[0115] When the target knowledge base matches the previous user reply information, the push robot replies to the target user with content related to the first target information in the conversation interface based on the information recorded in the target knowledge base.

[0116] 305. Forward the user reply information to the candidate robot, so that the candidate robot configured with the candidate knowledge base replies to the target user in the conversation interface.

[0117] After the push robot determines that the user's reply information does not match the target knowledge base, it forwards the user's reply information to the candidate robot so that the candidate robot can use its own candidate knowledge base to reply to the target user.

[0118] Preferably, the matching requirement of the candidate knowledge base should be lower than that of the target knowledge base. For example, the push robot scores the user's reply information according to the similarity between it and the target knowledge base. The higher the similarity between the user's reply information and the knowledge points stored in the target knowledge base, the higher the score. If the matching requirement of the target knowledge base is a score higher than 6.0, then the score requirement of the candidate knowledge base should be a value higher than 6.0. In this way, the user's reply information will always be matched with the target knowledge base with a lower matching requirement first, avoiding the situation where the user's reply information has a higher match with the candidate knowledge base of a candidate robot, but the push robot replies according to the target knowledge base.

[0119] based on Figure 2 or Figure 3 In the embodiment shown, the present application also provides another embodiment for describing the reply mechanism of the push robot when multiple messages are pushed. Figure 4 , this embodiment includes:

[0120] 401. Pushing second target information to the target user through the conversation interface;

[0121] Before pushing the first target information to the target user, the push robot has already pushed the second target information to the target user, and the target knowledge base configured by the push robot includes the first target knowledge base and the second target knowledge base, wherein the knowledge points in the first target knowledge base are mainly associated with the first target information, and the knowledge points in the second target knowledge base are mainly associated with the second target information. It can be understood that in this embodiment, there may be multiple second target knowledge points, that is, the push robot may have pushed multiple times to the target user in the same conversation interface, in which case the push robot can be configured with multiple second target knowledge bases.

[0122] 402. Pushing first target information to a target user through a conversation interface;

[0123] After receiving the first target information, the target user can browse the general content of the first target information from the conversation interface, or click on a link therein for more detailed browsing. In the present application, the target user receives the push information through the conversation interface, and the conversation interface itself provides a special input box or input bar for the target user to enter the user reply information. After the user enters the user reply information in the input bar or input box and sends it, the push robot receives the user reply information.

[0124] 403. Receive user reply information sent by the target user through the conversation interface;

[0125] After receiving the first target information and the second target information, the target user can browse the approximate content of the first target information or the second target information from the conversation interface, or click on a link therein for a more detailed browsing. In the present application, the target user receives the push information through the conversation interface, and the conversation interface itself provides a special input box or input bar for the target user to enter the user reply information. After the user enters the user reply information in the input bar or input box and sends it, the push robot receives the user reply information.

[0126] 404. Determine whether the user reply information matches the first target knowledge base; if so, execute step 405; if not, execute step 406;

[0127] The push robot first compares the user reply information with the first target knowledge base to determine whether the user reply information meets the matching requirements of the first target knowledge base. If the user reply information meets the matching requirements of the first target knowledge base, step 405 is executed. If the user reply information does not meet the matching requirements of the first target knowledge base, step 406 is executed;

[0128] 405. Reply to the user reply information in the conversation interface according to the first target knowledge base;

[0129] When the previous user reply information matches the first target knowledge base, the push robot replies to the target user with content related to the first target information in the conversation interface based on the information recorded in the first target knowledge base, and no longer attempts to match the user reply information with the second target knowledge base.

[0130] 406. Determine whether the user reply information matches the second target knowledge base; if the user reply information matches the second target knowledge base, execute step 407;

[0131] After the push robot fails to match the user reply information with the first target knowledge base, it will also try to match the user reply information with the second target knowledge base. If the matching requirement of the second target knowledge base is met, step 407 is executed. If the user reply information does not match the first target knowledge base and the second target knowledge base, the push robot can confirm that the user reply information does not match the target knowledge base, and there is no need to reply to the user with the knowledge points in the target knowledge base.

[0132] 407. Reply to the user reply information in the conversation interface according to the second target knowledge base.

[0133] When the previous user reply message does not match the first target knowledge base, but matches the second target knowledge base, the push robot replies to the target user with content related to the second target information in the conversation interface based on the information recorded in the second target knowledge base.

[0134] In this embodiment, the push robot will always try to associate the user with the first target knowledge base first. If the user's reply information can match the first target knowledge base, it will no longer consume memory to calculate whether the user's reply information matches the second target knowledge base. Preferably, the matching requirement of the second target knowledge base should be higher than that of the first target knowledge base, so as to avoid the situation where the user's reply information has a higher matching degree with the second target knowledge base, but the push robot replies according to the first target knowledge base. For specific knowledge base matching determination methods, please refer to Figure 2 The embodiments shown are not described in detail here.

[0135] In another embodiment, another reply mechanism is provided when the push robot pushes multiple messages. Figure 5 , this embodiment includes:

[0136] 501. Pushing second target information to the target user through the conversation interface;

[0137] Before pushing the first target information to the target user, the push robot has already pushed the second target information to the target user, and the target knowledge base configured by the push robot includes the first target knowledge base and the second target knowledge base, wherein the knowledge points in the first target knowledge base are mainly associated with the first target information, and the knowledge points in the second target knowledge base are mainly associated with the second target information. It can be understood that in this embodiment, there may be multiple second target knowledge points, that is, the push robot may have pushed multiple times to the target user in the same conversation interface, in which case the push robot can be configured with multiple second target knowledge bases.

[0138] 502. Pushing first target information to a target user through a conversation interface;

[0139] After receiving the first target information, the target user can browse the general content of the first target information from the conversation interface, or click on a link therein for more detailed browsing. In the present application, the target user receives the push information through the conversation interface, and the conversation interface itself provides a special input box or input bar for the target user to enter the user reply information. After the user enters the user reply information in the input bar or input box and sends it, the push robot receives the user reply information.

[0140] 503. Receive user reply information sent by the target user through the conversation interface;

[0141] After receiving the first target information and the second target information, the target user can browse the approximate content of the first target information or the second target information from the conversation interface, or click on a link therein for a more detailed browsing. In the present application, the target user receives the push information through the conversation interface, and the conversation interface itself provides a special input box or input bar for the target user to enter the user reply information. After the user enters the user reply information in the input bar or input box and sends it, the push robot receives the user reply information.

[0142] 504. Determine whether the user reply information matches the first target knowledge base or the second target knowledge base; if so, execute step 505;

[0143] The push robot matches the user reply information with the first target knowledge base and the second target knowledge base respectively, and calculates the similarity between the user reply information and the first target knowledge base, and the similarity between the user reply information and the second target knowledge base. If the user reply information matches with any one of the two target knowledge bases, it can be determined that the user reply information matches with the target knowledge base, and step 405 is executed. If the user reply information does not match with any target knowledge base, it is not necessary to use the target knowledge base to reply to the target user.

[0144] 505. Reply to the user reply information in the conversation interface according to the target knowledge base with the highest matching degree.

[0145] After the push robot determines that the user's reply information matches the target knowledge base, it will reply to the target user in the conversation interface based on the knowledge base in the target knowledge base that has the highest match with the user's reply information.

[0146] In this embodiment, the push robot will match the user's reply information with each target knowledge base, and use the target knowledge base with a higher matching degree and meeting the matching requirements to reply to the target user. At this time, the second target knowledge base and the first target knowledge base can be matched with equal difficulty. For specific knowledge base matching determination methods, please refer to Figure 2 The embodiments shown are not described in detail here.

[0147] Figure 4 and Figure 5In the illustrated embodiment, the second target information is sent earlier than the first target. Considering that users are more likely to reply based on the first target information that is closer in time, the matching requirement of the second target knowledge base can be set higher than that of the first target knowledge base, so that the push robot is more inclined to reply to the user based on the first target knowledge base. For example, the similarity score requirement of the second target knowledge base is set to increase with the sent time of the second target information, thereby increasing the probability of the push robot matching the user's reply information with the first target knowledge base. For example, when the push robot uses a similarity scoring mechanism to judge the similarity between the user's reply information and the knowledge base, it can be set that the first target knowledge base requires the user's reply information score to reach 6.0 or above to meet the matching requirement of the first target knowledge base, while the second target knowledge base requires the similarity score to reach 7.0 or above to meet the matching requirement of the second target knowledge base. Taking the marketing scenario as an example, the push robot is configured with a first target database of mobile phone promotion knowledge points, and also with a second target database of TV promotion knowledge points. The push robot pushed a TV promotion poster to the target user as the second target information three days ago, and pushed a mobile phone promotion to the target user as the first target information one hour ago. If the user sends "how long will the activity last" as the user reply information at this time, then the user is likely to be asking about the mobile phone promotion. Since the matching requirements of the first target knowledge base are lower, the push robot will match "how long will the activity last" to the first target database and inform the user of the duration of the mobile phone promotion.

[0148] In another embodiment of the present application, when the user reply information does not match the target knowledge base, the push robot will also send a guide to the target user to prompt the user to pay attention to the first target information. Figure 6 The embodiment shown, this embodiment includes:

[0149] 601. Pushing first target information to a target user through a conversation interface;

[0150] 602. Receive user reply information sent by the target user through the conversation interface;

[0151] 603. Determine whether the user reply information matches the target knowledge base; if it matches, execute step 604; otherwise, execute steps 605 and 606;

[0152] 604. Reply to the user reply information in the conversation interface according to the target knowledge base;

[0153] 605. Forward the user reply information to the candidate robot, so that the candidate robot configured with the candidate knowledge base replies to the target user in the conversation interface;

[0154] Steps 601 to 605 of this embodiment are similar to Figure 3 The details of steps 301 to 305 of the illustrated embodiment will not be repeated here.

[0155] 606. Send a guide message to the target user, where content of the guide message is associated with the first target information.

[0156] In this embodiment, when the user reply information does not match the target knowledge base, a guiding message such as "Dear, this promotion is really affordable, would you like to consider it again" will be sent to the target user to encourage the user to continue to consult the target user for other information related to the first target information.

[0157] Preferably, the push robot can preset multiple guide words. When sending the guide words, the push robot will count how many times the guide words have been sent before, and select a guide word from the preset multiple guide words to send to the target user, so as to avoid sending the same target guide word every time.

[0158] based on Figure 6 In another embodiment of the present application, a mechanism is added to transfer the conversation of the target user to other candidate robots. Figure 7 , this embodiment includes:

[0159] 701. Pushing first target information to a target user through a conversation interface;

[0160] 702. Receive user reply information sent by the target user through the conversation interface;

[0161] 703. Determine whether the user reply information matches the target knowledge base; if so, execute step 704; otherwise, execute steps 705 and 706;

[0162] 704. Reply to the user reply information in the conversation interface according to the target knowledge base;

[0163] 705. Forward the user reply information to the candidate robot, so that the candidate robot configured with the candidate knowledge base replies to the target user in the conversation interface;

[0164] Steps 701 to 705 of this embodiment are similar to Figure 3 The details of steps 301 to 305 of the illustrated embodiment will not be repeated here.

[0165] 706. Calculate the number of times the guide message has been sent;

[0166] When the target user always sends user reply information that is not related to the first target information, it means that the target user does not intend to learn other information related to the first target information. In this case, the push robot sending the guide words to the target user multiple times may lead to the target user's plan. Because the push robot will count how many times the guide words have been sent, it will judge whether the target user still intends to learn the first target information based on the number of times the guide words have been sent.

[0167] 707. When the number of times the guide message has been sent reaches a preset upper limit, the conversation with the target user is handed over to the candidate robot.

[0168] The push robot will transfer the conversation with the target user to the candidate robot, which will then directly reply to the target user's message without the push robot's participation, reducing the push robot's computing power.

[0169] Corresponding to the above information push method, Figure 8 A push robot is also presented that can perform Figures 2 to 7 The information push method shown. The push robot includes:

[0170] The push unit 801 is used to push the first target information to the target user through the conversation interface;

[0171] A receiving unit 802 is used to receive user reply information sent by the target user through the conversation interface;

[0172] A judging unit 803 is used to judge whether the user reply information matches the target knowledge base;

[0173] The reply unit 804 is used to reply the user reply information in the conversation interface according to the target knowledge base when the user reply information matches the target knowledge base.

[0174] Optionally, the push robot may further include:

[0175] The forwarding unit 805 is used to forward the user reply information to the candidate robot when the user reply information does not match the target knowledge base, so that the candidate robot configured with the candidate knowledge base replies to the target user in the conversation interface.

[0176] Optionally, the matching requirement of the target knowledge base is lower than that of the candidate knowledge base.

[0177] Optionally, the judging unit 803 is specifically configured to:

[0178] Evaluating the similarity score of the user's reply information according to the knowledge points in the target knowledge base;

[0179] If the similarity score meets the matching requirement of the target knowledge base, confirming that the user reply information matches the target knowledge base;

[0180] If the similarity score does not meet the matching requirement of the target knowledge base, it is confirmed that the user reply information does not match the target knowledge base.

[0181] Optionally, the target knowledge base includes a first target knowledge base and a second target knowledge base;

[0182] The pushing unit 801 is further configured to push the first target information to the target user through the conversation interface, and push the second target information to the target user through the conversation interface;

[0183] The determining unit 803 is specifically configured to:

[0184] Determining whether the user reply information matches the first target knowledge base;

[0185] When the user reply information does not match the first target knowledge base, determining whether the user reply information matches the second target knowledge base;

[0186] The recovery unit is specifically used for:

[0187] When the user reply information matches the first target knowledge base, replying to the user reply information in the conversation interface according to the first target knowledge base;

[0188] When the user reply information matches the second target knowledge base, the user reply information is replied in the conversation interface according to the second target knowledge base.

[0189] Optionally, the target knowledge base includes a first target knowledge base and a second target knowledge base;

[0190] The pushing unit 801 is further configured to push the second target information to the target user through the conversation interface before pushing the first target information to the target user through the conversation interface;

[0191] The judging unit 803 is specifically used to judge whether the user reply information matches the first target knowledge base or the second target knowledge base;

[0192] The reply unit 804 is specifically configured to reply to the user reply information in the conversation interface according to the target knowledge base with the highest matching degree when the user reply information matches the first target knowledge base or the second target knowledge base.

[0193] Optionally, the matching requirement of the second target knowledge base is higher than that of the target knowledge base.

[0194] Optionally, the matching requirement of the second target knowledge base is the same as the matching requirement of the target knowledge base.

[0195] Optionally, the matching requirement of the second target knowledge base is positively correlated with the sending time of the second target information.

[0196] Optionally, the push robot may further include:

[0197] The guiding unit 806 is configured to send guiding words to the target user when the user reply information does not match the target knowledge base, wherein the content of the guiding words is associated with the first target information.

[0198] Optionally, the guiding unit 806 is specifically used for:

[0199] Calculate the number of times the introductory message has been sent;

[0200] According to the sending times, a guiding sentence is selected from a plurality of preset guiding sentences and sent to the target user.

[0201] Optionally, the push robot may further include:

[0202] A calculation unit 807, configured to calculate the number of times the guide sentence has been sent when the user reply information does not match the target knowledge base;

[0203] The transfer unit 808 is used to transfer the conversation with the target user to a candidate robot when the number of times the guide message has been sent reaches a preset upper limit.

[0204] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0205] Fig. 9A computer device is shown, the computer device 900 may include one or more central processing units (CPU) 901 and a memory 905, wherein the memory 905 stores one or more application programs or data. The memory 905 may be a volatile storage or a persistent storage. The program stored in the memory 905 may include one or more modules, each of which may include a series of instruction operations in the computer device. Furthermore, the central processing unit 901 may be configured to communicate with the memory 905 to execute a series of instruction operations in the memory 905 on the computer device 900. The computer device 900 may also include one or more power supplies 902, one or more wired or wireless network interfaces 903, one or more input and output interfaces 904, and / or one or more operating systems, such as Windows ServerTM, Mac OS XTM, UnixTM, LinuxTM, FreeBSDTM, etc. The central processing unit 901 may execute the aforementioned Figures 1A to 7 The information push method in any of the illustrated embodiments will not be described in detail here.

[0206] The present application also provides a computer storage medium, wherein the computer storage medium stores instructions, and when the instructions are executed on a computer, the computer executes Figures 1A to 7 Any information push method in the illustrated embodiments.

[0207] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0208] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0209] 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 separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0210] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment 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), disk or optical disk and other media that can store program code.

Claims

1. An information push method, characterized in that: Applied to a push robot, the push robot is configured with a target knowledge base, and the method comprises: Pushing the first target information to the target user through the conversation interface; Receiving user reply information sent by the target user through the conversation interface; Determine whether the user reply information matches the target knowledge base; the target knowledge base stores semantic information or keywords and key sentences related to the first target information; If the user reply information matches the target knowledge base, replying to the user reply information in the conversation interface according to the target knowledge base; If the user reply information does not match the target knowledge base, the user reply information is forwarded to a candidate robot, so that the candidate robot configured with a candidate knowledge base replies to the target user in the conversation interface; wherein the candidate knowledge base stores information that has a low correlation with the first target information, or has no correlation at all; and the matching requirement of the target knowledge base is lower than that of the candidate knowledge base.

2. The information push method according to claim 1, characterized in that: The determining whether the user reply information matches the target knowledge base includes: Evaluating the similarity score of the user's reply information according to the knowledge points in the target knowledge base; If the similarity score meets the matching requirement of the target knowledge base, confirming that the user reply information matches the target knowledge base; If the similarity score does not meet the matching requirement of the target knowledge base, it is confirmed that the user reply information does not match the target knowledge base.

3. The information push method according to claim 1, characterized in that: The target knowledge base includes a first target knowledge base and a second target knowledge base; Before pushing the first target information to the target user through the conversation interface, the method further includes: Pushing second target information to the target user through the conversation interface; The determining whether the user reply information matches the target knowledge base includes: Determining whether the user reply information matches the first target knowledge base; If the user reply information matches the first target knowledge base, replying to the user reply information in the conversation interface according to the first target knowledge base; If the user reply information does not match the first target knowledge base, determining whether the user reply information matches the second target knowledge base; If the user reply information matches the second target knowledge base, the user reply information is replied in the conversation interface according to the second target knowledge base.

4. The information push method according to claim 1, characterized in that: The target knowledge base includes a first target knowledge base and a second target knowledge base; Before pushing the first target information to the target user through the conversation interface, the method further includes: Pushing second target information to the target user through the conversation interface; The determining whether the user reply information matches the target knowledge base includes: Determining whether the user reply information matches the first target knowledge base or the second target knowledge base; If the user reply information matches the first target knowledge base or the second target knowledge base, the user reply information is replied in the conversation interface according to the target knowledge base with the highest matching degree.

5. The information push method according to claim 3 or 4, characterized in that: The matching requirement of the second target knowledge base is higher than that of the first target knowledge base.

6. The information push method according to claim 3 or 4, characterized in that: The matching requirement of the second target knowledge base is the same as the matching requirement of the first target knowledge base.

7. The information push method according to claim 3 or 4, characterized in that: The matching requirement of the second target knowledge base is positively correlated with the sending time of the second target information.

8. The information push method according to claim 1, characterized in that: The method further comprises: If the user reply information does not match the target knowledge base, a guide statement is sent to the target user, and the content of the guide statement is associated with the first target information.

9. The information push method according to claim 8, characterized in that: The sending of the guide words to the target user includes: Calculate the number of times the introductory message has been sent; According to the number of times the messages have been sent, a guide sentence is selected from a plurality of preset guide sentences and sent to the target user.

10. The information push method according to claim 8, characterized in that: The method further comprises: If the user reply information does not match the target knowledge base, the number of times the guide sentence has been sent is calculated; When the number of times the guide message has been sent reaches a preset upper limit, the conversation with the target user is handed over to the candidate robot.

11. A push robot, characterized in that: The push robot is configured with a target knowledge base, and the push robot comprises: A push unit, configured to push the first target information to the target user through the conversation interface; A receiving unit, configured to receive user reply information sent by a target user through the conversation interface; A judging unit, used to judge whether the user reply information matches the target knowledge base; the target knowledge base stores semantic information or keywords and key sentences related to the first target information; A reply unit, configured to reply to the user reply information in the conversation interface according to the target knowledge base when the user reply information matches the target knowledge base; A forwarding unit is used to forward the user reply information to a candidate robot when the user reply information does not match the target knowledge base, so that the candidate robot configured with the candidate knowledge base replies to the target user in the conversation interface; wherein the candidate knowledge base stores information that has a low correlation with the first target information, or has no correlation at all; and the matching requirement of the target knowledge base is lower than that of the candidate knowledge base.

12. The push robot according to claim 11, characterized in that: The judging unit is specifically used for: Evaluating the similarity score of the user's reply information according to the knowledge points in the target knowledge base; If the similarity score meets the matching requirement of the target knowledge base, confirming that the user reply information matches the target knowledge base; If the similarity score does not meet the matching requirement of the target knowledge base, it is confirmed that the user reply information does not match the target knowledge base.

13. The push robot according to claim 11, characterized in that: The target knowledge base includes a first target knowledge base and a second target knowledge base; The pushing unit is further configured to push the second target information to the target user through the conversation interface before pushing the first target information to the target user through the conversation interface; The judging unit is specifically used for: Determining whether the user reply information matches the first target knowledge base; When the user reply information does not match the first target knowledge base, determining whether the user reply information matches the second target knowledge base; The reply unit is specifically used for: When the user reply information matches the first target knowledge base, replying to the user reply information in the conversation interface according to the first target knowledge base; When the user reply information matches the second target knowledge base, the user reply information is replied in the conversation interface according to the second target knowledge base.

14. The push robot according to claim 11, characterized in that: The target knowledge base includes a first target knowledge base and a second target knowledge base; The push unit is also used for: Before pushing the first target information to the target user through the conversation interface, pushing the second target information to the target user through the conversation interface; The judging unit is specifically used to judge whether the user reply information matches the first target knowledge base or the second target knowledge base; The reply unit is specifically configured to reply to the user reply information in the conversation interface according to the target knowledge base with the highest matching degree when the user reply information matches the first target knowledge base or the second target knowledge base.

15. The push robot according to claim 13 or 14, characterized in that: The matching requirement of the second target knowledge base is higher than that of the first target knowledge base.

16. The push robot according to claim 13 or 14, characterized in that: The matching requirement of the second target knowledge base is the same as the matching requirement of the first target knowledge base.

17. The push robot according to claim 13 or 14, characterized in that: The matching requirement of the second target knowledge base is positively correlated with the sending time of the second target information.

18. The push robot according to claim 11, characterized in that: Also includes: A guiding unit is used to send guiding words to the target user when the user reply information does not match the target knowledge base, and the content of the guiding words is associated with the first target information.

19. The push robot according to claim 18, characterized in that: The guiding unit is specifically used for: Calculate the number of times the introductory message has been sent; According to the number of times the messages have been sent, a guide sentence is selected from a plurality of preset guide sentences and sent to the target user.

20. The push robot according to claim 18, characterized in that: Also includes: A calculation unit, used for calculating the number of times the guide words have been sent when the user reply information does not match the target knowledge base; The transfer unit is used to transfer the conversation with the target user to a candidate robot when the number of times the guide message has been sent reaches a preset upper limit.

21. A computer device, characterized in that: include: Memory; One or more processors coupled to the memory; One or more applications, wherein the one or more applications are stored in the memory and configured to cause the one or more processors to execute the method according to any one of claims 1 to 10 when executed by the one or more processors.

22. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores instructions, and when the instructions are executed on a computer, the computer executes the method according to any one of claims 1 to 10.

Images