CN113505293A

CN113505293A - Information pushing method and device, electronic equipment and storage medium

Info

Publication number: CN113505293A
Application number: CN202110662590.4A
Authority: CN
Inventors: 邓锐涛; 常向月; 刘云峰
Original assignee: Shenzhen Zhuiyi Technology Co Ltd
Current assignee: Shenzhen Zhuiyi Technology Co Ltd
Priority date: 2021-06-15
Filing date: 2021-06-15
Publication date: 2021-10-15
Anticipated expiration: 2041-06-15
Also published as: CN113505293B

Abstract

The embodiment of the application provides an information pushing method and device, electronic equipment and a storage medium, and relates to the technical field of information. The method comprises the following steps: obtaining a conversation log corresponding to an entrance identifier, wherein the entrance identifier is used for identifying a user interface for realizing conversation with a user; clustering a plurality of user problems in the dialog log to obtain a plurality of problem clusters; and determining at least one push question based on the plurality of question clusters, the at least one push question being for being pushed and displayed on the user interface. The pushing problem can be determined according to the user problem really input by the user interface, so that the obtained pushing problem is more in line with the question-asking intention of the user, and the accuracy of the pushing problem is improved.

Description

Information pushing method and device, electronic equipment and storage medium

Technical Field

The present application relates to the field of information technologies, and in particular, to an information pushing method and apparatus, an electronic device, and a storage medium.

Background

With the development of internet technology, intelligent customer service technology establishes a fast and effective communication mode based on natural language for enterprises and mass users, and is widely applied in various industries at present. The intelligent customer service technology can push related questions to the user before the user asks the questions to guide the user to input the questions, so that the user can quickly respond to the questions. However, the pushing problem of the existing intelligent customer service technology is fixed, the pushing is difficult to meet the conversation intention of the user, the pushing problem is not accurate enough, and the user experience is reduced.

Disclosure of Invention

In view of the foregoing problems, the present application provides an information pushing method, an information pushing apparatus, an electronic device, and a storage medium to improve the above drawbacks.

In a first aspect, an embodiment of the present application provides an information pushing method, including: obtaining a conversation log corresponding to an entrance identifier, wherein the entrance identifier is used for identifying a user interface for realizing conversation with a user; clustering a plurality of user problems in the dialog log to obtain a plurality of problem clusters; and determining at least one push question based on the plurality of question clusters, the at least one push question being for being pushed and displayed on the user interface.

Further, the clustering a plurality of user questions in the dialog log to obtain a plurality of question clusters includes: extracting each of the user questions input by the user in the dialog log; semantic feature extraction is carried out on each user question to obtain a feature vector of each user question; and clustering the characteristic vectors of the user problems to obtain the plurality of problem clusters.

Further, the determining at least one push question based on the plurality of question clusters includes: determining the user question corresponding to the central point of each question cluster as the candidate question; and determining the at least one push question in the candidate questions corresponding to each question cluster.

Further, the determining at least one push question based on the plurality of question clusters includes: acquiring the distance between the feature vector of each user question and the central point of the question cluster; determining the user problem represented by the feature vector smaller than the specified distance as a candidate problem; and determining the at least one push question in the candidate questions corresponding to each question cluster.

Further, the determining the at least one push question in the candidate questions corresponding to each question cluster includes: acquiring the number of the user questions in each question cluster; sorting the candidate problems in a descending order according to the number; and determining a prescribed number of the candidate questions ranked top as the at least one push question.

Further, after the determining the top-ranked specified number of the candidate questions as the at least one push question, the method further comprises: pushing the at least one push question and the ranking order of the at least one push question to the user interface to display the at least one push question on the user interface in the ranking order.

Further, after the determining at least one push question based on the plurality of question clusters, the method further comprises: acquiring at least one preset specified problem and the priority of the specified problem; arranging and combining the at least one designated question and the at least one pushing question according to the priority to obtain a question combination; and pushing the question combination to the user interface.

Further, after the determining at least one push question based on the plurality of question clusters, the method further comprises: acquiring at least one preset standard problem; obtaining semantic similarity between each standard question and each pushed question; and if the semantic similarity is larger than a specified threshold, updating the push question corresponding to the semantic similarity into the standard question.

Further, the obtaining a dialog log corresponding to the entry identifier includes: updating the conversation log corresponding to the entrance identifier into a conversation log at a specified time at intervals of preset time; the clustering processing of the multiple user questions in the dialog log to obtain multiple question clusters includes: clustering a plurality of user problems in the dialog logs of the specified time to obtain a plurality of updated problem clusters; the determining at least one push question based on the plurality of question clusters comprises: updating the at least one push question based on the updated plurality of question clusters.

In a second aspect, an embodiment of the present application provides an information pushing method, including: if a conversation request of a user is detected, acquiring an entrance identifier of a user interface for realizing the conversation with the user; obtaining at least one pushing question corresponding to the entrance identifier, wherein the at least one pushing question is a question determined based on the obtained plurality of question clusters after clustering a plurality of user questions, and the plurality of user questions are questions input by users in a dialog log corresponding to the entrance identifier; and displaying the push question on the user interface.

Further, after the obtaining of the entry identifier of the user interface that implements the session with the user if the session request of the user is detected, the method further includes: and acquiring a dialog log of the user corresponding to the entrance identifier, and sending the dialog log to a server.

In a third aspect, an embodiment of the present application provides an information pushing apparatus, including: the log acquisition module is used for acquiring a conversation log corresponding to an entrance identifier, wherein the entrance identifier is used for identifying a user interface for realizing conversation with a user; the cluster processing module is used for carrying out cluster processing on a plurality of user problems in the conversation log to obtain a plurality of problem clusters; and a question determining module for determining at least one push question based on the plurality of question clusters, the at least one push question being for being pushed and displayed on the user interface.

Further, the cluster processing module comprises: the question extraction submodule is used for extracting each user question input by the user in the dialog log; the feature extraction submodule is used for extracting semantic features of each user question to obtain a feature vector of each user question; and the characteristic clustering submodule is used for clustering the characteristic vectors of the user problems to obtain the plurality of problem clusters.

Further, the problem determination module includes: a central problem determining submodule, configured to determine a user problem corresponding to the central point of each problem cluster as the candidate problem; and a push question determining sub-module for determining the at least one push question in the candidate questions corresponding to each question cluster.

Further, the problem determination module includes: the distance acquisition submodule is used for acquiring the distance between the feature vector of each user question and the central point of the question cluster; the candidate problem determining submodule is used for determining the user problem represented by the feature vector smaller than the specified distance as a candidate problem; and a push question determining sub-module for determining the at least one push question in the candidate questions corresponding to each question cluster.

Further, the push problem determination sub-module includes: a quantity obtaining unit, configured to obtain the quantity of the user questions in each question cluster; the sorting unit is used for sorting the candidate problems in a descending order according to the number; and a push question determining subunit, configured to determine a specified number of the top-ranked candidate questions as the at least one push question.

Further, the information pushing device further comprises: the pushing module is used for pushing the at least one pushing question and the arrangement sequence of the at least one pushing question to the user interface so as to display the at least one pushing question on the user interface in the arrangement sequence.

Further, the information pushing device further comprises: the system comprises a specified problem acquisition module, a priority acquisition module and a priority management module, wherein the specified problem acquisition module is used for acquiring at least one preset specified problem and the priority of the specified problem; the problem combination obtaining module is used for carrying out permutation and combination on the at least one appointed problem and the at least one pushing problem according to the priority to obtain a problem combination; and the question combination pushing module is used for pushing the question combination to the user interface.

Further, the information pushing device further comprises: the standard problem acquisition module is used for acquiring at least one preset standard problem; the similarity obtaining module is used for obtaining the semantic similarity between each standard problem and each push problem; and the standard problem updating module is used for updating the push problem corresponding to the semantic similarity into the standard problem if the semantic similarity is greater than a specified threshold.

Further, the log obtaining module further comprises: the log updating module is used for updating the conversation log corresponding to the entrance identifier into a conversation log at a specified time every preset time; the cluster processing module further comprises: the problem cluster updating module is used for clustering a plurality of user problems in the dialog logs of the specified time to obtain a plurality of updated problem clusters; a push question update module to update the at least one push question based on the updated plurality of question clusters.

In a fourth aspect, an embodiment of the present application provides an information pushing apparatus, including: the entrance identification acquisition module is used for acquiring the entrance identification of a user interface realizing the conversation with the user if the conversation request of the user is detected; a problem obtaining module, configured to obtain at least one push problem corresponding to the entry identifier, where the at least one push problem is a problem determined based on a plurality of obtained problem clusters after clustering a plurality of user problems, and the plurality of user problems are problems input by users in a dialog log corresponding to the entry identifier; and the question display module is used for displaying the push question on the user interface.

Further, the portal identification obtaining module further includes: and the log sending module is used for acquiring the conversation log of the user corresponding to the entrance identifier and sending the conversation log to a server.

In a fifth aspect, an embodiment of the present application provides an electronic device, including: one or more processors; a memory; one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the one or more programs configured to perform the methods of the first or second aspects described above.

In a sixth aspect, embodiments of the present application provide a computer-readable storage medium, in which program code is stored, and the program code can be called by a processor to execute the method according to the first aspect or the second aspect.

Drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the description of the embodiments are briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.

FIG. 1 is a schematic diagram of an application environment suitable for use in embodiments of the present application;

FIG. 2 illustrates a client application interface provided by an embodiment of the present application;

FIG. 3 illustrates a user interface for interacting with a customer provided by an embodiment of the present application;

fig. 4 is a flowchart illustrating an information pushing method provided by an embodiment of the present application;

fig. 5 is a processing diagram illustrating an information pushing method according to an embodiment of the present application;

fig. 6 is a flowchart illustrating an information pushing method according to another embodiment of the present application;

fig. 7 is a flowchart illustrating an information pushing method according to another embodiment of the present application;

fig. 8 shows a schematic flowchart of step S407 in fig. 7;

fig. 9 is a flowchart illustrating an information pushing method according to still another embodiment of the present application;

fig. 10 is a flowchart illustrating an information pushing method according to another embodiment of the present application;

fig. 11 is a flowchart illustrating an information pushing method according to yet another embodiment of the present application;

fig. 12 is a flowchart illustrating an information pushing method according to still another embodiment of the present application;

fig. 13 is a flowchart illustrating an information pushing method according to still another embodiment of the present application;

fig. 14 is a block diagram illustrating a structure of an information pushing apparatus according to an embodiment of the present application;

fig. 15 shows a block diagram of an information pushing apparatus according to another embodiment of the present application;

fig. 16 is a block diagram illustrating a structure of an electronic device according to an embodiment of the present application, the electronic device being configured to execute an information push method according to an embodiment of the present application;

fig. 17 illustrates a storage unit for storing or carrying program codes for implementing an information pushing method according to an embodiment of the present application.

Detailed Description

The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.

With the development of science and technology, artificial intelligence technology is more and more popular, and services such as appointments, consultations and the like in daily life are changed from artificial services to machine services. Compared with manual service, the service efficiency is greatly improved through machine service, and great convenience is brought to daily life of people.

Generally, intelligent customer service can provide services to users by means of dialogue with the users. Specifically, the user may input a question on a user interface that is in conversation with the smart customer service, and the smart customer service pushes an answer to the user by identifying the question input by the user. In order to enable a user to conveniently obtain a desired answer and guide the user to input a question, the method is very critical, and specifically, some questions can be pushed on a user interface to be referred by the user so as to guide the user to input the question, so that the time for inputting the question by the user is shortened, and the question clarification is performed. However, the problem of pushing to users in current intelligent customer service is usually fixed and needs manual addition by service personnel to realize the pushing. Because there may be many problems actually for users served by the intelligent customer service, a service staff is required to add preset problems as many as possible, a large amount of labor cost is consumed, accuracy and efficiency cannot be guaranteed, it is difficult to push the problems meeting the user conversation intention, the user usually needs to input the problems many times to communicate with the customer service staff to obtain answers, user experience is poor, and loss of the user in the intelligent customer service conversation process is caused.

The inventor of the application researches the user behavior when the user interacts with the intelligent customer service, finds that a plurality of interactive interfaces usually exist in a client in practical application, and the interactive interfaces are respectively provided with a conversation inlet of the intelligent customer service, so that the user can enter a user interface which is in conversation with the intelligent customer service by triggering the conversation inlet. By considering more user behaviors, the inventor finds that generally when a user has a problem in the process of interacting with the current interactive interface, the user can trigger the conversation entrance of the current interactive interface so as to have a conversation with the intelligent customer service. That is, the question input by the user is usually related to the business data of the interactive interface corresponding to the current dialog entry. For example, when the interaction page is a credit card application page, user questions to enter into a smart customer service session are typically related to credit card information; when the interactive interface is a shopping page, user questions that enter into an intelligent customer service session are typically related to payment information; when the interactive interface is a personal information interface, user questions that enter into an intelligent customer service session are typically related to personal information; when the interactive interface is a financial interface, the user's question of entering into an intelligent customer service session is typically related to investing in financing. That is, the user questions input by the user who performs the intelligent customer service session through the same interactive interface may be relatively close, if the user questions input by the user can be analyzed according to the interactive interface, the user intention of the interactive interface can be accurately identified, that is, the user inputs questions that the user wants to answer, the questions related to the user intention are pushed to the user during the session, the accuracy of question pushing can be improved, and the user experience of the user can be improved.

In order to improve the above problem, the inventor proposes an information pushing method, an information pushing apparatus, an electronic device, and a storage medium in the embodiments of the present application. The pushing problem can be determined by analyzing the user problem on the user interface, the pushing problem accords with the intention of the user, and the pushing accuracy can be improved.

Referring to fig. 1, fig. 1 is a schematic diagram illustrating an application environment suitable for the embodiment of the present application. The information push method provided by the embodiment of the application can be applied to the information push system 10 shown in fig. 1. The information push system 10 comprises a plurality of or at least one terminal device 100 and a server 200, the terminal device 100 and the server 200 are located in a wireless network or a wired network, and the terminal device 100 and the server 200 perform data interaction, wherein the interaction data includes, but is not limited to, audio, video, text, images, and the like.

The terminal device 100 may be a mobile terminal device, and may include, for example, a smart phone, a tablet computer, an e-book reader, a laptop portable computer, a vehicle-mounted computer, a wearable mobile terminal, and so on. The server 200 may be an individual server, a server cluster, a server center formed by a plurality of servers, a local server, or a cloud server. Server 200 may be configured to provide a background service for a user, which may include, but is not limited to, information push, and the like.

In some embodiments, the terminal device 100 may have a client application installed thereon, and the user may communicate with the server 200 based on the client application (e.g., APP, etc.). Specifically, the terminal device 100 may obtain input information of a user, and based on a client application program on the terminal device 100 communicating with the server 200, the server 200 may process the received input information of the user, and the server 200 may further return corresponding output information to the terminal device 100 according to the information, and the terminal device 100 may perform an operation corresponding to the output information. The input information of the user may be voice information, touch operation information based on a screen, gesture information, action information, and the like, and the output information may be an image, a video, a text, an audio, and the like, which is not limited herein.

Specifically, the client application on the terminal device includes multiple interactive interfaces, different content is displayed on different interactive interfaces, an icon or a key of a dialog entry may be set on the interactive interface, and the current interactive interface may be switched to a user interface for dialog with a user in response to a trigger operation of the user on the dialog entry. That is, the user interface for interacting with the user is the next level interface of the current interactive interface. For example, fig. 2 illustrates a client application interface provided by an embodiment of the present application. The interactive interface in fig. 2 is a credit card application interface, and a "customer service" button is displayed at the upper end of the credit card application interface and is used for entering a user interface for interacting with the customer service in response to the operation of the button by the user, namely, a conversation request of the user. As shown in fig. 3, fig. 3 illustrates a user interface for interacting with a client provided by an embodiment of the present application. The intelligent customer service interface in fig. 3 is provided with a preset display area for guessing a question, the preset display area displays a push question, and the bottom of the intelligent customer service interface is provided with a text or voice input area. Optionally, the user can quickly input the push question by clicking on the push question or speaking the push question by voice.

The server 200 is capable of collecting user data of a user at a client application, and in particular, the user data may include a portal identifier corresponding to a user session request, and a session log of the user. Alternatively, the server 200 may push information for the user of the terminal device 100 according to the collected dialog logs; the server 200 may also transmit the collected conversation log to the terminal device 100 so that the terminal device 100 determines push information from the conversation log.

The above application environments are only examples for facilitating understanding, and it is to be understood that the embodiments of the present application are not limited to the above application environments.

The information pushing method, the information pushing device, the electronic device and the medium provided by the embodiments of the present application will be described in detail below with specific embodiments.

Referring to fig. 3, fig. 3 is a schematic flow chart of an information pushing method according to an embodiment of the present application, and is applied to the server. The information pushing method comprises S201 to S203.

S201, obtaining a dialog log corresponding to the entrance identifier.

The portal identification is used to identify a user interface for implementing a dialog with a user.

The user interface is an interface for implementing a dialog function. Specifically, a dialog entry is arranged on some interactive interfaces, wherein each user interface for implementing dialog corresponds to one dialog entry, the dialog entry is arranged on the upper-level interactive interface of the user interface, each dialog entry corresponds to one entry identifier, and the entry identifiers can identify the user interfaces for implementing dialog with the user.

As an embodiment, each dialog entry has an entry identifier, that is, the user interface and the entry identifier are in a one-to-one correspondence. As another example, multiple conversation portals may correspond to the same portal identification. Wherein the plurality of conversation portals may be interactive interfaces having similar content. For example, the similar contents may be similar service types, and on an interactive interface for purchasing electronic products, such as a notebook computer, a mobile phone, and the like, a plurality of conversation entries have the same identifier; on an interactive interface for purchasing books, such as a children's book, a history book, a textbook, etc., a plurality of conversation entries have the same identification.

The dialogue log which is acquired by the server and corresponds to the entrance identifier comprises the following steps: the entry identifies a plurality of historical dialog logs generated by a plurality of historical users on the corresponding user interface, and each historical dialog log records a real dialog process of one historical user and the intelligent/artificial customer service. The history user is a user who has performed a conversation on the user interface. A dialog log is a record of a user's dialog in text format. Alternatively, the server may directly read the dialog log in the text format from the database, or may record the dialog in the voice format, and obtain the dialog log in the text format through voice recognition.

Specifically, the dialog log corresponding to each user includes at least one user message and at least one system message. Optionally, an object identifier may be included in the conversation log, the object identifier identifying whether the message was entered by the user or by the system. As an embodiment, the system message may be input by the customer service robot, that is, the conversation log includes a conversation record of the user with the intelligent customer service. As another embodiment, the system message may be entered by a human customer service, i.e., the conversation log includes a record of the user's conversation with the human customer service.

It should be understood that the dialog log is not limited to the customer service log, and may be a dialog text log in other scenarios. For example, the dialog log may be a question set by one party and an answer provided by multiple parties under the question and answer platform; or a chat log of the non-question-and-answer type, etc.

S202: and clustering a plurality of user problems in the conversation log to obtain a plurality of problem clusters.

After obtaining the dialog log of the portal identification, a plurality of user questions in the dialog log may be extracted. The user question can reflect the user's dialog intention, i.e., the meaning that the user input question wants to express, i.e., the user inputs the question that the question wants to answer.

In particular, the server may extract user messages entered by the user from the dialog log based on the object identification. As an embodiment, each user message input by the user may be determined as each user question, thereby enabling more comprehensive analysis of the message input by the user. As another embodiment, question detection may be performed on the user message, for example, whether a question word or the like is included is detected, a question input by the user is obtained, and each question input by the user is determined as each user question, so that the intention of the user question can be analyzed more accurately.

In some embodiments, before clustering multiple user questions, user questions may also be filtered. Specifically, user questions may be filtered, with questions identified as chatting questions being filtered based on intent. By primarily screening the user problems, the operation cost can be reduced, and the memory consumption can be reduced.

The clustering process may be implemented by various algorithms. For example, a K-means algorithm, also known as a K-means or K-means clustering algorithm, a graph clustering algorithm, such as a label propagation algorithm, etc., a hierarchical clustering algorithm, density-based clustering, grid-based clustering, etc., may be used. It can be understood that the clustering process can adopt a clustering algorithm of unsupervised learning, the unsupervised learning does not need to manually mark a large number of samples in advance, and the limitation of the granularity of class division due to overhigh training cost can be avoided.

In some embodiments, the clustering process may be a text clustering process based on semantic similarity. The method can perform clustering processing on a plurality of user problems based on semantics to obtain a plurality of problem clusters. Wherein each question cluster comprises at least one user question with similar semantics. Specifically, please refer to the following embodiments.

In other embodiments, the clustering process may be a text clustering process that is performed according to the traffic type. For example, in a banking business scenario, the business type may be financial, bankcard, credit card, and so forth. The problem clusters corresponding to all service types can be obtained through clustering.

In still other embodiments, the server may classify a plurality of user questions according to preset labels to obtain a label of each user question, and perform clustering processing based on semantic matching on the user questions having the same label to obtain at least one question cluster corresponding to each label. The label may be a type of business associated with the user interface, or may be a representation of the user, such as the user's age, membership grade, gender, etc. For example, the preset tag may be an age group of the user, and after the user question corresponding to the user of each age group is obtained, the user questions of each age group are clustered respectively to obtain a plurality of question clusters corresponding to each age group.

S203: at least one push question is determined based on a plurality of question clusters.

Wherein the at least one push question is to be pushed and displayed on a user interface. The user interface is an interface corresponding to the portal identification that enables a dialog with the user. And an information display area is arranged on the user interface and is used for displaying the pushing problem. Optionally, when the push problem is a real user problem in the dialog log, after the push problem is determined, error correction processing may be performed on the push problem based on the dictionary and the rule, a wrongly written word or a wrongly grammatical word in the push problem is corrected, and the problem after error correction is pushed.

In some embodiments, the server may determine one or more target problem clusters from the plurality of problem clusters, and then determine a push problem based on the user problem of the target problem cluster.

As one approach, the target problem cluster may be determined according to the label of the current business scenario or the label of the current interactive user. For example, a user portrait of a current interactive user can be acquired, a problem cluster with the same user portrait label is determined as a target problem cluster, and attributes of the current interactive user are considered when a push problem is determined, so that the push problem is more targeted, and the push accuracy is higher.

As another way, the target problem cluster may be determined according to the number of the user problems in each problem cluster, and the problem cluster with a larger number is determined as the target problem cluster, so that the push problem is the user problem that the user frequently asks. By determining the target problem cluster first, power consumption required for processing all problem clusters can be reduced, and particularly under the condition that the problem clusters are more, the calculation amount can be greatly reduced, and the calculation efficiency is improved.

In some embodiments, the server may determine candidate questions corresponding to each question cluster, and determine at least one push question among the candidate questions corresponding to each question cluster. Specifically, please refer to the following embodiments, which are not described herein.

In some implementations, after determining the push question, an answer to the push question may also be determined and pushed.

As one way, a plurality of standard questions and an answer to each standard question may be set in advance, wherein a standard question is a question sentence having a standard format and content, which can be accurately recognized. A standard question with the highest semantic similarity to the push question may be determined, and the answer to the standard question may be determined as the answer to the push question. Alternatively, the standard questions may also be updated by the operator based on the push questions. For example, when there is no standard question similar to the push question, the standard question corresponding to the push question is configured and the answer is configured.

Alternatively, the answer to the push question may be determined in the conversation log based on the context of the user message and the system message in the conversation log. As an embodiment, a system message corresponding to the push question input by the user may be searched in the dialog log, and the system message may be determined as an answer to the push question. As another embodiment, a question cluster to which the pushed question belongs may be obtained, a system message corresponding to each user question in the question cluster is collected as an answer cluster, and a system message corresponding to a center point of the answer cluster is determined as an answer to the pushed question.

Referring to fig. 5, fig. 5 is a processing diagram illustrating an information pushing method according to an embodiment of the present application. As shown in FIG. 5, the credit card application interface is provided with a dialogue portal "customer service" button, which has a portal identification of 1. The server acquires a conversation log corresponding to the entrance identifier 1, clusters a plurality of user questions in the conversation log to obtain a plurality of question clusters, determines at least one pushing question 1 based on the plurality of question clusters, and pushes the pushing question 1 to a user interface which is corresponding to the entrance identifier 1 and used for conversation to display. And the terminal equipment responds to the triggering operation of the user on the conversation entrance of the entrance identifier 1 on the credit card application interface, and switches the current credit card application interface into a user interface corresponding to the entrance identifier 1, namely an intelligent customer service interface displaying the pushing question 1. As shown in fig. 5, the shopping mall interface and the personal information interface are also provided with session entries, and the session entries respectively correspond to entry identifiers, and similarly, the user interfaces corresponding to other entry identifiers can be pushed and display the push questions corresponding to the entry identifiers.

It will be appreciated that the push question on each user interface is derived from the user question corresponding to the portal identification of that user interface, with different portal identifications corresponding to different push questions. Since users who trigger a dialog through the same dialog entry usually have the same dialog intention, the user questions input by the users are often closer. Therefore, the pushing problem is determined based on the problem clusters obtained by clustering the problems of the users with the same entrance identifier, and the pushing problem is displayed on the user interface of the entrance identifier, so that the user intention of triggering the conversation entrance can be hit more accurately, and the accuracy of the pushing problem is improved.

In practical application, the process of determining the push problem can be automatically carried out by the server, a large number of conversation entries can be put in a business scene, so that the user problem in the conversation scene corresponding to each conversation entry can be divided into fine granularity, the push problem on the user interface corresponding to the entry identifier accords with the intention of the user triggering the conversation entry with larger probability, the user problem is hit, and the time for the user to input the problem and clarify the problem is greatly shortened.

The information pushing method provided by the embodiment of the application obtains a conversation log corresponding to an entrance identifier, wherein the entrance identifier is used for identifying a user interface for realizing conversation with a user; clustering a plurality of user problems in the conversation log to obtain a plurality of problem clusters; at least one push question is determined based on the plurality of question clusters, the at least one push question to be pushed and displayed on the user interface. The method and the device can determine the pushing problem according to the user problem input by the historical user based on the entrance identification, so that the obtained pushing problem is more in line with the question-asking intention of the user, and the accuracy of the pushing problem is improved.

Referring to fig. 6, fig. 6 is a schematic flow chart of an information pushing method according to an embodiment of the present application, and is applied to the server. The method includes S301 to S305.

S301: a conversation log corresponding to the portal identification is obtained.

S302: each user question entered by the user in the dialog log is extracted.

S303: and extracting semantic features of each user question to obtain a feature vector of each user question.

The feature vector of each user question can be used for representing the semantics of the user question, and the feature vector is a mathematical representation converted from a text form by performing semantic feature extraction on the user question. Specifically, the server may extract semantic features of the user question based on a preset semantic understanding model, input the user question into the semantic understanding model, and output a feature vector of the user question. As an embodiment, the semantic understanding model may be a model such as a Bidirectional encoding from transforms (BERT) network based on a transformer. The specific implementation of semantic feature extraction is not limited herein.

In some embodiments, the server may perform word segmentation processing on each user question, and divide the user question into independent words to obtain word segmentation phrases of each question; and semantic feature extraction is carried out on the word segmentation phrase of each user problem, the word vector of the word segmentation phrase of each user problem is calculated, and the feature vector of the sentence of the whole user problem is obtained according to the word vector of the word segmentation phrase of each user problem. Optionally, before word segmentation processing, error correction processing may be performed on the user problem to correct a wrongly written word or a wrongly grammatical word in the user problem, so as to obtain a more accurate feature vector.

In one embodiment, the feature vector may be generalized using the upper and lower features or the synonym relationship of the features. Generalization refers to the replacement of some similar words with the same expression, e.g. "how much money" a few money is generalized to. The semantics of the user problem can be accurately represented through generalization processing, and the accuracy of clustering processing in subsequent processing is improved.

S304: and clustering the characteristic vectors of the user problems to obtain a plurality of problem clusters.

The server can cluster the feature vectors according to the semantic similarity of the feature vectors of the user problems to obtain a plurality of problem clusters. The number of problem clusters can be preset or automatically determined by a clustering algorithm in the clustering process, and the central point of the clustering can be preset or randomly selected. Specifically, the distance between the feature vector of each user question and each clustering center is calculated, the clustering center to which the user question belongs is determined according to the distance, and the clustering center is updated. The distance between the feature vectors can represent the semantic similarity between the user questions corresponding to the feature vectors. The larger the distance, the smaller the semantic similarity. Alternatively, the distance may be a cosine distance, a euclidean distance, a manhattan distance, or the like.

And clustering the characteristic vectors of the user problems to obtain a plurality of problem clusters. The distance between the feature vectors of the user questions in the same question cluster is as small as possible, that is, the semantics of the user questions in the same question cluster are relatively close. The distance between the feature vectors of the user questions in different question clusters is as large as possible, that is, the semantic difference expressed by the user questions in different question clusters is large.

Due to the diversity of language expression, for the same problem, multiple user question sentences may exist, and multiple user problems containing the same semantics may be obtained by extracting the user log. By clustering, user problems with similar semantics can be classified into one category, which is beneficial to managing the user problems. For example, after the clustering process is performed, "how to open a credit card? "," how to handle a credit card? "how do I want to do with credit card? "these semantically identical or similar user questions are classified as a problem cluster.

S305: at least one push question is determined based on a plurality of question clusters.

It should be noted that, for portions of the foregoing method steps that are not described in detail, please refer to the foregoing embodiments, and further description is omitted here.

According to the information pushing method provided by the embodiment of the application, each user problem input by a user in a dialog log is extracted, semantic features of each user problem are extracted to obtain a feature vector of each user problem, and the feature vectors of the user problems are clustered to obtain a plurality of problem clusters. The semantic feature extraction is carried out to obtain the feature vector of each user question, then the feature vectors of the user questions are clustered to obtain a plurality of question clusters, the user questions in each question cluster can have similar semantics, and then the pushing question corresponding to the user intention is determined based on the obtained plurality of question clusters.

Referring to fig. 7, fig. 7 is a schematic flowchart of an information pushing method according to an embodiment of the present application, and the information pushing method is applied to the server. The method includes S401 to S406.

S401: a conversation log corresponding to the portal identification is obtained.

S402: each user question entered by the user in the dialog log is extracted.

S403: and extracting semantic features of each user question to obtain a feature vector of each user question.

S404: and clustering the characteristic vectors of the user problems to obtain a plurality of problem clusters.

S405: and determining the user question corresponding to the central point of each question cluster as a candidate question.

And the central point of the problem cluster is a clustering center obtained after clustering processing. Specifically, the central point is a feature vector of the user question located in the center of the cluster, and can represent the center of semantic distribution in each question cluster. That is, the user question corresponding to the center point of the question cluster may represent the semantics of the question cluster. By determining the user problem corresponding to the center point of each problem cluster as a candidate problem, a candidate problem corresponding to each problem cluster can be obtained, and the candidate problem can cover the semantics of more user problems in the problem cluster.

S406: at least one push question is determined among the candidate questions corresponding to each question cluster.

In some embodiments, the server may determine one or more target problem clusters from the plurality of problem clusters, and then determine the push problem from the candidate problems of the target problem clusters. Please refer to the foregoing embodiments for specific implementation of determining a target problem cluster, which is not described herein again.

In some embodiments, after the candidate questions corresponding to each question cluster are obtained, the candidate questions of the plurality of question clusters may be ranked, and the push question may be determined according to the ranking order. Specifically, the candidate questions can be scored through dimensions such as the length of the candidate questions, semantic consistency of the candidate questions, the number of preset keywords appearing in the candidate questions, the frequency of the candidate questions appearing in a dialog log and the like, the candidate questions are sorted in a descending order according to scores, and the questions with the designated number sorted in the front are determined as the pushing questions. The dimensionalities such as the length, semantic consistency and the like of the candidate problem can represent readability, and the higher the score is, the better the corresponding readability is; the keywords can be preset words related to the service scene, and the number of the preset keywords appearing in the candidate problems can represent the correlation degree of the user problems and the service scene; the frequency of occurrence of candidate questions may characterize the heat of the question. By determining the candidate question with a higher score as the push question, a more appropriate push question can be determined by combining multiple factors.

In some embodiments, as shown in fig. 8, S406 may include S4061 through S4063.

S4061: and acquiring the number of user questions in each question cluster.

And respectively counting the number of user problems in each problem cluster. Since the user questions are real questions input by the user in the historical dialog, each question cluster can represent a semantic meaning, namely a user intention, and the number of the user questions can represent the heat of the question cluster. It should be noted that recurring problems will be counted repeatedly. For example, if the question cluster includes 2 repeated user questions "how to apply for a credit card", the number of user questions in the question cluster is 2.

S4062: and sorting the candidate problems in a descending order according to the number.

And performing descending sorting on the problem clusters according to the number of the user problems in each problem cluster, and taking the sorting sequence of the problem clusters as the sorting sequence of the candidate problems of the problem clusters. For example, if the question cluster 1 includes 10 user questions, the candidate question is question 1, the question cluster 2 includes 5 user questions, and the candidate question is question 2, the sorted order is question 1 and question 2.

As an embodiment, the candidate questions in the question cluster may be ranked according to the distance of the candidate question from the center point, or the score value of the candidate questions. As an implementation mode, for the problem clusters with the same number, the sequence can be determined according to the average score values of the candidate problems in the problem clusters, and the problem cluster with higher score is arranged in the front.

S4063: determining a specified number of top ranked candidate questions as at least one push question.

Wherein the specified number is a preset number. As an embodiment, the specified number is a maximum number of questions that can be displayed on the user interface. After the candidate questions are sorted in a descending order, the candidate questions with the designated number sorted in the front can be determined as the push questions, so that the push questions are the questions with high popularity and frequently asked by the user, and the accuracy of the user intention hit by the push questions is improved. As another embodiment, the specified number may be operator configured.

In some embodiments, step S4073 may be performed by the server, and the server may push the push question to the terminal device to display the push question on a user interface of the terminal device after determining the at least one push question. In one mode, after pushing the push question to the terminal device, if the server acquires the request information sent by the terminal device, the server pushes other questions than the push question to the terminal device. The request information is used for representing a request for acquiring more push questions input by a user after the push questions are displayed on the user interface. For example, the request message may be sent in response to a "change question" operation triggered by the user on the interactive interface of the push question.

In other embodiments, step S4073 may also be performed by the terminal device, and the server may send the candidate questions and the ranking order to the terminal device, and the terminal device determines the top specified number of candidate questions as the at least one push question.

In some implementations, after determining the top-ranked specified number of candidate questions as the at least one push question, the at least one push question and an order of the at least one push question may be pushed to the user interface to display the at least one push question on the user interface in the order of the order. For example, the push questions can be displayed on the display interface from top to bottom or from left to right according to the arrangement sequence, so that the questions with higher popularity and frequently asked by the user are displayed at the position easy to see first according to the reading habit of the user, the display of the push questions is more in line with the reading habit of the user, and the user experience is improved.

According to the information pushing method provided by the embodiment of the application, the user problem corresponding to the central point of each problem cluster is determined as the candidate problem, at least one pushing problem is determined in the candidate problems corresponding to each problem cluster, the user problem in the semantic distribution center in each problem cluster can be determined as the candidate problem, the candidate problem can cover more semantics of the user problem in each problem cluster, and the pushing problem is enabled to better accord with the intention of a user.

Referring to fig. 7, fig. 7 is a schematic flowchart of an information pushing method according to an embodiment of the present application, and the information pushing method is applied to the server. The method includes S501 to S507.

S501: a conversation log corresponding to the portal identification is obtained.

S502: each user question entered by the user in the dialog log is extracted.

S503: and extracting semantic features of each user question to obtain a feature vector of each user question.

S504: and clustering the characteristic vectors of the user problems to obtain a plurality of problem clusters.

And S505, acquiring the distance between the feature vector of each user question and the center point of the question cluster.

For each problem cluster, the distance of the feature vector of each user problem in the problem cluster from the center point of the problem cluster can be calculated. The distance between the feature vectors can represent the semantic similarity between the user questions corresponding to the feature vectors. The larger the distance, the smaller the semantic similarity. Alternatively, the distance may be a cosine distance, a euclidean distance, a manhattan distance, or the like.

S506: and determining the user problem characterized by the feature vector less than the specified distance as a candidate problem.

The specified distance may be used to characterize a semantic similarity threshold. As an embodiment, the specified distance may be a preset distance. As another example, the specified distance may be an average distance between the center point of each problem cluster and other points of the problem cluster.

After the distance between the feature vector of each user question in the question cluster and the center point of the question cluster is obtained, the feature vector with the distance smaller than the specified distance is determined, and the user question represented by the feature vector is determined as a candidate question corresponding to the question cluster. That is, the candidate question is a question in each question cluster near the center of the semantic distribution, and may be used to represent the semantics of the question cluster.

It should be noted that the candidate questions include the user question corresponding to the center point of the question cluster. In some embodiments, if there is no feature vector smaller than the specified distance in the question cluster, the candidate question of the question cluster is the user question corresponding to the central point.

S507: at least one push question is determined among the candidate questions corresponding to each question cluster.

In some embodiments, the candidate questions corresponding to each question cluster may be sorted, the candidate question sorted at the top is determined as the target candidate question of the question cluster, and at least one push question is determined in the target candidate question corresponding to each question cluster. The target candidate questions can be determined by simultaneously considering semantics and various factors in each question cluster by sorting according to the scoring values of the candidate questions. For a specific implementation of the ranking according to the score value, reference may be made to the foregoing examples. Each problem cluster corresponds to one pushing problem, so that semantic difference among the pushing problems is large, and more user intentions are covered when the same number of pushing problems are displayed.

In some embodiments, S507 may include obtaining the number of user questions in each question cluster; sorting the candidate problems in a descending order according to the number; determining a specified number of top ranked candidate questions as at least one push question. Alternatively, the candidate question may be each candidate question in each question cluster, or may be a target candidate question.

According to the information pushing method provided by the embodiment of the application, the distance between the feature vector of each user question and the central point of the question cluster is obtained; determining the user problem represented by the feature vector smaller than the specified distance as a candidate problem; and determining at least one push question in the candidate questions corresponding to each question cluster. Because the center of each problem cluster can represent the semantic distribution center, candidate problems are determined according to the distance from the center point, more semantics of user problems can be covered in each problem cluster, and the problem pushing is more in line with the intention of a user.

Referring to fig. 10, fig. 10 is a schematic flowchart of an information pushing method according to an embodiment of the present application, and is applied to the server. The method includes S601 to S606.

S601: a conversation log corresponding to the portal identification is obtained.

S602: and clustering a plurality of user problems in the conversation log to obtain a plurality of problem clusters.

S603: at least one push question is determined based on a plurality of question clusters.

S604: and acquiring at least one preset specified problem and the priority of the specified problem.

The specified question is a preset question. For example, specifying at least one question that may be configured by an operator. As an embodiment, the assignment problem is the same for each portal identification. As another embodiment, a specific question may be set for each entry identification, respectively.

In particular, the priority is used to determine the ordering of the assignment problem and the push problem. For example, the priority may be specifying that the question is ranked before the push question, or specifying that the question is ranked after the push question. As another example, the priority is used to determine where to insert the specified question into at least one push question.

S605: and ranking and combining the at least one appointed question and the at least one pushing question according to the priority to obtain a question combination.

And ranking and combining the at least one appointed question and the at least one pushing question according to the priority to obtain a question combination comprising the at least one appointed question and the at least one pushing question. For example, when the priority is to specify a question to be arranged before the push question, the question is specified to be question 1, and the push question is question 2, the question combination is question 1, question 2.

As an embodiment, the specified question and the priority of the specified question are stored in a server, and S604 and S605 are performed by the server. As another embodiment, the specified question and the priority of the specified question are stored in the terminal device, and the server may push the push question to the terminal device, and the terminal device performs S604 and S605.

S606: the problem combinations are pushed to the user interface.

The problem combination is pushed to the user interface, the problem combination can be displayed according to the arrangement sequence of the problem combination, the pushed problem obtained by a dialog log of a user and a preset specified problem can be displayed on the user interface, and therefore the requirement for displaying specific content in some service scenes is met, and the application scenes of the information pushing method in the embodiment of the application are enriched.

According to the information push method provided by the embodiment of the application, after at least one push problem is determined based on a plurality of problem clusters, at least one preset appointed problem and the priority of the appointed problem are obtained, the at least one appointed problem and the at least one push problem are arranged and combined according to the priority to obtain a problem combination, and the problem combination is pushed to a user interface, so that the problem combination of the push problem and the appointed problem can be displayed on the user interface, the requirement for displaying specific content in some service scenes is met, and the application scenes of information push are enriched.

Referring to fig. 11, fig. 11 is a schematic flowchart of an information pushing method according to an embodiment of the present application, and is applied to the server. The method includes S701 to S706.

S701: a conversation log corresponding to the portal identification is obtained.

S702: and clustering a plurality of user problems in the conversation log to obtain a plurality of problem clusters.

S703: at least one push question is determined based on a plurality of question clusters.

S704: and acquiring at least one preset standard question.

The standard question is a preset question with a standard format and a text form of standard content, has definite semantics, and can be accurately identified by a server or a terminal device. Optionally, for each standard question, its corresponding standard answer may also be configured.

In some embodiments, the standard Questions may be Questions stored in a Frequent Ask Questions (FAQ) knowledge base after unified processing of user Questions with similar semantics. Since different users may have different expression modes for the same question, some common standard questions may be defined in the FAQ knowledge base for subsequent question determination operations. For example, "how to increase the credit card amount" is a standard problem, and similar user problems corresponding to the standard problem include "how the credit card amount is not enough," how to make the credit card amount a bit higher ", and so on.

S705: and acquiring semantic similarity between each standard question and each push question.

For each standard question, semantic similarity of the standard question to each push question may be obtained based on text matching.

As an implementation manner, semantic extraction can be performed on the standard problem and the push problem respectively based on a preset semantic understanding model to obtain vector representations of the standard problem and the user problem, and then semantic similarity is determined according to the distance between the vector representations, wherein the smaller the distance is, the larger the semantic similarity is. For example, the Semantic similarity of each standard question to each push question may be determined by a Deep Semantic model (DSSM).

As another embodiment, the cross features between the input standard problem and the push problem can be extracted through a preset depth model to obtain a matching signal tensor, and then the matching signal tensors are aggregated to obtain the semantic similarity. For example, the predetermined depth model may be ARC-II (volumetric New Network architecture for Matching Natural language), MatchPyramid (text Matching as Image recognition), or the like.

The embodiment of the present application does not limit the specific implementation manner of how to obtain the semantic similarity.

S706: and if the semantic similarity is greater than the specified threshold, updating the push problem corresponding to the semantic similarity into a standard problem.

Wherein, the designated threshold is a preset semantic similarity threshold. If the semantic similarity is larger than the specified threshold value, the semantic similarity of the push question corresponding to the semantic similarity and the standard question is judged, the push question can be updated to the standard question, and the updated push question is pushed to the user interface. As an embodiment, when the standard question is configured with a standard answer, the standard answer corresponding to the standard question may also be pushed to the user interface, and when the user touches or inputs the pushed question on the interactive interface, the standard answer is displayed on the terminal device.

In some embodiments, if the semantic similarity is less than or equal to the specified threshold, the pushing problem may be subjected to error correction based on a dictionary and rules, wrongly written words or wrongly grammars in the pushing problem may be corrected, and the pushing problem may be updated to an error-corrected problem, so as to improve readability of the pushing problem.

Because the standard problem is preset, the push problem before updating is the problem of real user input, and the standard problem is more smooth in semantics and easier to accurately identify compared with the problem input by a real user. By replacing the pushing problem with the standard problem with similar semantics, the readability of the pushing problem can be improved, the accuracy of problem identification can be improved, and the situation that the semantics are not smooth possibly existing in the pushing problem caused by real user problems can be avoided.

According to the information push method provided by the embodiment of the application, after at least one push problem is determined based on a plurality of problem clusters, the semantic similarity between each standard problem and each push problem is obtained by obtaining at least one preset standard problem, if the semantic similarity is larger than a specified threshold value, the push problem corresponding to the semantic similarity is updated to be the standard problem, the push problem can be replaced by the standard problem under the condition that the semantics are similar, and therefore the readability of the push problem is improved.

Referring to fig. 12, fig. 12 is a schematic flowchart of an information pushing method according to an embodiment of the present application, and is applied to the server. The method includes S801 to S803.

S801: and updating the dialog log corresponding to the entrance identifier into the dialog log of the specified time at every preset time.

The preset time is a preset time interval for updating the dialog log, and can represent the frequency of updating the dialog log. The designated time is the preset time length of the acquired dialog log. The preset time and the designated time may be set by the server by default or may be configured by the operator. Specifically, please refer to the foregoing embodiment, which is not described herein again.

As an embodiment, the specified time may be a length of time from the current time. For example, if the preset time is 30 minutes and the specified time is 12 hours from the current time, the dialog log corresponding to the entry identifier is updated to the dialog log of the entry identifier 12 hours in the past every 30 minutes. By updating the dialog logs at preset time intervals, the dialog logs with better timeliness can be obtained, and data support is provided for determining the user intention more accurately in the follow-up process. Alternatively, the server may save the dialog log corresponding to the specified time of the portal identification every preset time, and remove the previous dialog log, thereby saving the storage space.

In some embodiments, after the session log corresponding to the entry identifier at the specified time is obtained, it may be determined whether the user log at the specified time meets a preset updating condition, and if so, the session log is updated to the session log at the specified time. As one way, the preset update condition may be a preset number threshold for determining whether the user question in the dialog log at the specified time is greater than the number threshold. By presetting the updating condition, the subsequent processing can not be carried out under the condition that the conversation log does not meet the condition, thereby improving the data processing efficiency.

S802: and clustering a plurality of user problems in the dialog log at the specified time to obtain a plurality of updated problem clusters.

The server may perform clustering processing on a plurality of problems in the dialog log at the specified time, and reference may be made to the foregoing embodiment to obtain an updated implementation manner of a plurality of problem clusters, which is not described herein again.

S803: updating at least one push question based on the updated plurality of question clusters.

For an implementation of updating at least one push question based on a plurality of updated question clusters, reference may be made to the foregoing embodiments, which are not described herein again. The server can push the updated push problem to the terminal equipment, so that the timeliness change of the user intention can be captured more accurately under the condition of considering the timeliness information, and the information push accuracy is higher.

According to the information push method provided by the embodiment of the application, the conversation log corresponding to the entrance identifier is updated to the conversation log at the appointed time at intervals of the preset time, a plurality of user problems in the conversation log at the appointed time are clustered to obtain a plurality of updated problem clusters, at least one push problem is updated based on the plurality of updated problem clusters, the push problems can be updated at intervals of the preset time, so that the push problems are determined in real time, and the timeliness of the push problems is improved.

Referring to fig. 13, fig. 13 is a schematic flowchart of an information pushing method according to an embodiment of the present application, and is applied to the terminal device. The method includes S901 to S903.

S901: and if the conversation request of the user is detected, acquiring an entrance identifier of a user interface for realizing the conversation with the user.

The dialogue request is used for representing the operation of triggering the dialogue entrance by the user in the man-machine interaction process. Specifically, some interactive interfaces of the client application program are provided with a conversation inlet, and a conversation request of a user in a man-machine interaction process can be responded, and the current interactive interface can jump to a user interface for realizing conversation with the user. Each user interface for realizing the conversation corresponds to one conversation inlet, each conversation inlet corresponds to one inlet identification, and the inlet identification can be used for identifying the user interface. When a dialog request of a user is detected, a dialog entry triggered by the user can be determined, and an entry identification of the dialog entry is determined as an entry identification of a user interface for realizing a dialog with the user.

In some embodiments, after step S901, the information pushing method may further include obtaining a dialog log of the user corresponding to the portal identification, and sending the dialog log to the server.

Specifically, after the portal identifier is obtained, a dialog log of the user corresponding to the portal identifier on the current terminal device may be obtained, the terminal device establishes a connection with the server, and the dialog log is sent to the server. Wherein the dialog log is a historical dialog record of the user in text format, and the dialog log may include an entry identifier of a user interface generating the dialog, at least one user message, and at least one system message. Optionally, the dialog log may also include the time of each message or the user identification of the current terminal device, etc. As an embodiment, the terminal device may send a session log of the session to the server after each session is ended.

S902: at least one push question corresponding to the portal identification is obtained.

The at least one push question is a question determined based on the obtained plurality of question clusters after clustering a plurality of user questions, the plurality of user questions being questions input by users in a dialog log corresponding to the portal identification. Optionally, the terminal device may further obtain an answer corresponding to the push question.

In some embodiments, the terminal device may further obtain the ranking order of the push questions from the server.

Please refer to the foregoing embodiments for a specific implementation of the server sending the push problem, which is not described herein again.

S903: the push question is displayed on a user interface.

The user interface is used for realizing a dialog with a user. And an information display area is arranged on the user interface and is used for displaying the acquired push problem. As an embodiment, the push question may be arranged on a plurality of keys of the information display area, respectively, with a push question text being displayed on each key. If the click operation of the user on the push problem on the interactive interface is obtained, the push problem is quickly input on the interactive interface, namely the click operation on the push problem can replace the user typing or voice input push problem. Optionally, after the push question is input quickly, the terminal device may obtain and display an answer to the push question.

In some implementations, a specified number of push questions can be displayed at most on the user interface. Wherein the specified number may be used to limit the number of push questions per fetch. As an implementation manner, a push trigger for acquiring a push question may be set on the user interface, and the terminal device may send, in response to an operation of the user on the push trigger, request information for acquiring more push questions to the server, so as to acquire more push questions to update the currently displayed push questions. For example, the operation of the user on the push trigger may be clicking a "change question" button on the user interface.

In some embodiments, after the terminal device obtains the ranking order of the push questions, the terminal device may display the push questions in the ranking order on the user interface. For example, the push question may be displayed in a sorted order from top to bottom, or from left to right. Because the rank order can characterize the popularity of the push problem, according to the reading habit and the rank order of the user, the popularity can be higher, and the problem frequently asked by the user is displayed at the position easily seen first, so that the reading habit of the user is more met, and the user experience is improved.

According to the information pushing method provided by the embodiment of the application, if a user conversation request is detected, an entry identifier of a user interface for realizing conversation with a user is obtained, at least one pushing problem corresponding to the entry identifier is obtained, the at least one pushing problem is determined based on the obtained plurality of problem clusters after clustering processing is carried out on the plurality of user problems, the plurality of user problems are problems input by the user in a conversation log corresponding to the entry identifier, the pushing problem is displayed on the user interface, the pushing problem corresponding to the entry identifier can be displayed on the user interface, the pushing problem is determined according to historical user problems of the same entry identifier, the pushing problem is made to accord with the intention of the user, and user experience is improved.

It should be understood that the foregoing examples are merely illustrative of the application of the method provided in the embodiments of the present application in a specific scenario, and do not limit the embodiments of the present application. The method provided by the embodiment of the application can also be used for realizing more different applications.

Referring to fig. 14, fig. 14 is a block diagram illustrating a structure of an information pushing apparatus according to an embodiment of the present application. As will be explained below with respect to the structural block diagram shown in fig. 14, the information pushing apparatus 1400 includes: a log obtaining module 1410, a cluster processing module 1420, and a problem determining module 1430, wherein: a log obtaining module 1410, configured to obtain a dialog log corresponding to an entry identifier, where the entry identifier is used to identify a user interface for implementing a dialog with a user; a clustering module 1420, configured to perform clustering on multiple user questions in the call log to obtain multiple question clusters; the question determining module 1430 is configured to determine at least one push question based on the plurality of question clusters, the at least one push question being intended to be pushed and displayed on the user interface.

Further, the cluster processing module 1420 includes: the question extraction submodule is used for extracting each user question input by the user in the dialog log; the feature extraction submodule is used for extracting semantic features of each user question to obtain a feature vector of each user question; and the characteristic clustering submodule is used for clustering the characteristic vectors of the user problems to obtain the plurality of problem clusters.

Further, the problem determination module 1430 includes: a central problem determining submodule, configured to determine a user problem corresponding to the central point of each problem cluster as the candidate problem; and a push question determining sub-module for determining the at least one push question in the candidate questions corresponding to each question cluster.

Further, the problem determination module 1430 includes: the distance acquisition submodule is used for acquiring the distance between the feature vector of each user question and the central point of the question cluster; the candidate problem determining submodule is used for determining the user problem represented by the feature vector smaller than the specified distance as a candidate problem; and a push question determining sub-module for determining the at least one push question in the candidate questions corresponding to each question cluster.

Further, the information pushing apparatus 1400 further includes: the pushing module is used for pushing the at least one pushing question and the arrangement sequence of the at least one pushing question to the user interface so as to display the at least one pushing question on the user interface in the arrangement sequence.

Further, the information pushing apparatus 1400 further includes: the system comprises a specified problem acquisition module, a priority acquisition module and a priority management module, wherein the specified problem acquisition module is used for acquiring at least one preset specified problem and the priority of the specified problem; the problem combination obtaining module is used for carrying out permutation and combination on the at least one appointed problem and the at least one pushing problem according to the priority to obtain a problem combination; and the question combination pushing module is used for pushing the question combination to the user interface.

Further, the information pushing apparatus 1400 further includes: the standard problem acquisition module is used for acquiring at least one preset standard problem; the similarity obtaining module is used for obtaining the semantic similarity between each standard problem and each push problem; and the standard problem updating module is used for updating the push problem corresponding to the semantic similarity into the standard problem if the semantic similarity is greater than a specified threshold.

Further, the log obtaining module 1410 further includes: the log updating module is used for updating the conversation log corresponding to the entrance identifier into a conversation log at a specified time every preset time; the cluster processing module further comprises: the problem cluster updating module is used for clustering a plurality of user problems in the dialog logs of the specified time to obtain a plurality of updated problem clusters; a push question update module to update the at least one push question based on the updated plurality of question clusters.

Referring to fig. 15, fig. 15 is a block diagram illustrating an information pushing apparatus according to an embodiment of the present application. As will be explained below with respect to the block diagram of the structure shown in fig. 15, the information pushing apparatus 1500 includes: an entry identification acquisition module 1510, a question acquisition module 1520, and a question display module 1530, wherein: an entry identifier obtaining module 1510, configured to obtain an entry identifier of a user interface that implements a session with a user if a session request of the user is detected; a question obtaining module 1520, configured to obtain at least one push question corresponding to the entry identifier, where the at least one push question is a question determined based on the obtained plurality of question clusters after clustering a plurality of user questions, and the plurality of user questions are questions input by users in a dialog log corresponding to the entry identifier; and a question display module 1530 for displaying the push question on the user interface.

Further, the portal identification obtaining module 1510 further includes: and the log sending module is used for acquiring the conversation log of the user corresponding to the entrance identifier and sending the conversation log to a server.

It can be clearly understood by those skilled in the art that, for convenience and brevity of description, the specific working processes of the above-described apparatuses and modules may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.

In the several embodiments provided in the present application, the coupling between the modules may be electrical, mechanical or other type of coupling.

Referring to fig. 16, a block diagram of an electronic device according to an embodiment of the present disclosure is shown. The electronic device 1600 may be a smart phone, a tablet computer, an electronic book, or other electronic devices capable of running an application. The electronic device 1600 in the present application may include one or more of the following components: a processor 1610, a memory 1620, and one or more applications, wherein the one or more applications may be stored in the memory 1620 and configured to be executed by the one or more processors 1610, the one or more programs configured to perform a method as described in the aforementioned method embodiments.

Processor 1610 may include one or more processing cores. The processor 1610, using various interfaces and circuitry, connects various components throughout the electronic device 1600, performs various functions of the electronic device 1600 and processes data by executing or executing instructions, programs, code sets, or instruction sets stored in the memory 1620, and invoking data stored in the memory 1620. Alternatively, the processor 1610 may be implemented in hardware using at least one of Digital Signal Processing (DSP), Field-Programmable Gate Array (FPGA), and Programmable Logic Array (PLA). Processor 1610 may integrate one or a combination of a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), a modem, and the like. Wherein, the CPU mainly processes an operating system, a user interface, an application program and the like; the GPU is used for rendering and drawing display content; the modem is used to handle wireless communications. It is understood that the modem may not be integrated into the processor 1610, but may be implemented by a communication chip.

The Memory 1620 may include a Random Access Memory (RAM) and a Read-Only Memory (Read-Only Memory). The memory 1620 may be used to store instructions, programs, code sets, or instruction sets. The memory 1620 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for implementing at least one function (such as a touch function, a sound playing function, an image playing function, etc.), instructions for implementing various method embodiments described below, and the like. The data storage area may also store data created by the electronic device 1600 during use (e.g., phone books, audio-video data, chat log data), etc.

Referring to fig. 17, a block diagram of a computer-readable storage medium according to an embodiment of the present disclosure is shown. The computer-readable storage medium 1700 has stored therein program code that can be invoked by a processor to perform the methods described in the method embodiments above.

The computer-readable storage medium 1700 may be an electronic memory such as a flash memory, an electrically-erasable programmable read-only memory (EEPROM), an erasable programmable read-only memory (EPROM), a hard disk, or a ROM. Alternatively, the computer-readable storage medium 1000 includes a non-volatile computer-readable storage medium. The computer readable storage medium 1700 has storage space for program code 1710 for performing any of the method steps described above. The program code can be read from or written to one or more computer program products. The program code 1710 may be compressed, for example, in a suitable form.

Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit the same; although the present application has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; such modifications and substitutions do not necessarily depart from the spirit and scope of the corresponding technical solutions in the embodiments of the present application.

Claims

1. An information pushing method, comprising:

obtaining a conversation log corresponding to an entrance identifier, wherein the entrance identifier is used for identifying a user interface for realizing conversation with a user;

clustering a plurality of user problems in the dialog log to obtain a plurality of problem clusters; and

determining at least one push question based on the plurality of question clusters, the at least one push question to be pushed and displayed on the user interface.

2. The method of claim 1, wherein clustering the plurality of user questions in the dialog log to obtain a plurality of question clusters comprises:

extracting each of the user questions input by the user in the dialog log;

semantic feature extraction is carried out on each user question to obtain a feature vector of each user question; and

and clustering the characteristic vectors of the user problems to obtain the plurality of problem clusters.

3. The method of claim 2, wherein determining at least one push question based on the plurality of question clusters comprises:

determining the user question corresponding to the central point of each question cluster as the candidate question; and

determining the at least one push question among the candidate questions corresponding to each question cluster.

4. The method of claim 2, wherein determining at least one push question based on the plurality of question clusters comprises:

acquiring the distance between the feature vector of each user question and the central point of the question cluster;

determining the user problem represented by the feature vector smaller than the specified distance as a candidate problem; and

5. The method according to claim 3 or 4, wherein the determining the at least one push question among the candidate questions corresponding to each question cluster comprises:

acquiring the number of the user questions in each question cluster;

sorting the candidate problems in a descending order according to the number; and

determining a top-ranked specified number of the candidate questions as the at least one push question.

6. The method of claim 5, wherein after the determining the top-ranked specified number of the candidate questions as the at least one push question, the method further comprises:

pushing the at least one push question and the ranking order of the at least one push question to the user interface to display the at least one push question on the user interface in the ranking order.

7. The method of claim 1, wherein after the determining at least one push question based on the plurality of question clusters, the method further comprises:

acquiring at least one preset specified problem and the priority of the specified problem;

arranging and combining the at least one designated question and the at least one pushing question according to the priority to obtain a question combination; and

pushing the question combination to the user interface.

8. The method of claim 1, wherein after the determining at least one push question based on the plurality of question clusters, the method further comprises:

acquiring at least one preset standard problem;

obtaining semantic similarity between each standard question and each pushed question; and

and if the semantic similarity is greater than a specified threshold value, updating the push question corresponding to the semantic similarity to the standard question.

9. The method of claim 1, wherein obtaining the conversation log corresponding to the entry identifier comprises:

updating the conversation log corresponding to the entrance identifier into a conversation log at a specified time at intervals of preset time;

the clustering processing of the multiple user questions in the dialog log to obtain multiple question clusters includes:

clustering a plurality of user problems in the dialog logs of the specified time to obtain a plurality of updated problem clusters;

the determining at least one push question based on the plurality of question clusters comprises:

updating the at least one push question based on the updated plurality of question clusters.

10. An information pushing method, comprising:

if a conversation request of a user is detected, acquiring an entrance identifier of a user interface for realizing the conversation with the user;

obtaining at least one pushing question corresponding to the entrance identifier, wherein the at least one pushing question is a question determined based on the obtained plurality of question clusters after clustering a plurality of user questions, and the plurality of user questions are questions input by users in a dialog log corresponding to the entrance identifier; and

displaying the push question on the user interface.

11. The method of claim 10, wherein after the obtaining an entry identifier of a user interface for implementing a dialog with the user if the user's dialog request is detected, the method further comprises:

and acquiring a dialog log of the user corresponding to the entrance identifier, and sending the dialog log to a server.

12. An information pushing apparatus, comprising:

the log acquisition module is used for acquiring a conversation log corresponding to an entrance identifier, wherein the entrance identifier is used for identifying a user interface for realizing conversation with a user;

the cluster processing module is used for carrying out cluster processing on a plurality of user problems in the conversation log to obtain a plurality of problem clusters; and

a question determination module for determining at least one push question based on the plurality of question clusters, the at least one push question being for being pushed and displayed on the user interface.

13. An information pushing apparatus, comprising:

the entrance identification acquisition module is used for acquiring the entrance identification of a user interface realizing the conversation with the user if the conversation request of the user is detected;

a problem obtaining module, configured to obtain at least one push problem corresponding to the entry identifier, where the at least one push problem is a problem determined based on a plurality of obtained problem clusters after clustering a plurality of user problems, and the plurality of user problems are problems input by users in a dialog log corresponding to the entry identifier; and

a question display module for displaying the push question on the user interface.

14. An electronic device, comprising:

one or more processors;

a memory;

one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the one or more programs configured to perform the method of any of claims 1-11.

15. A computer-readable storage medium, characterized in that a program code is stored in the computer-readable storage medium, which program code can be called by a processor to perform the method according to any of claims 1-11.