CN112035615A

CN112035615A - Online inquiry data processing method and device and computer equipment

Info

Publication number: CN112035615A
Application number: CN202010898095.9A
Authority: CN
Inventors: 高丽
Original assignee: Kangjian Information Technology Shenzhen Co Ltd
Current assignee: Kangjian Information Technology Shenzhen Co Ltd
Priority date: 2020-08-31
Filing date: 2020-08-31
Publication date: 2020-12-04
Anticipated expiration: 2040-08-31
Also published as: CN112035615B

Abstract

The application relates to artificial intelligence and provides an on-line inquiry data processing method, an on-line inquiry data processing device and computer equipment. The method comprises the following steps: acquiring initial inquiry data corresponding to the user identification; querying a target interview tactical subset matched with the initial interview data; determining a target inquiry dialect according to the dialect path of the target inquiry dialect subset, and sending the target inquiry dialect to a terminal; receiving target inquiry data fed back by a terminal; when the target inquiry data is judged to be the professional term consultation data, feeding back spoken answering data matched with the corresponding professional terms to the terminal, and executing the step of receiving the target inquiry data fed back by the terminal; and when the target inquiry data is judged to be inquiry answer data and meets the corresponding answer conditions, executing the step of determining the target inquiry dialogues according to the dialogues paths of the target inquiry dialogues subset until each inquiry dialogues in the target inquiry dialogues subset is traversed. By adopting the method, the inquiry data processing efficiency can be improved.

Description

Online inquiry data processing method and device and computer equipment

Technical Field

The application relates to the technical field of artificial intelligence, in particular to an online inquiry data processing method and device and computer equipment.

Background

With the continuous development of artificial intelligence technology, the online intelligent inquiry mode based on artificial intelligence is gradually developed, and a great deal of convenience is brought to the life of people. In the current inquiry data processing mode in the intelligent inquiry mode, usually, an inquiry-answer pair is configured in advance, when inquiry data triggered by a user is acquired, target answer data matched with the inquiry data is inquired from the preconfigured inquiry-answer pair, the inquired target answer data is fed back to the user, and when the target answer data matched with the inquiry data is not inquired, an artificial answer flow is triggered aiming at the inquiry data so as to manually answer the inquiry data. However, this type of inquiry data processing method has a problem of low efficiency.

Disclosure of Invention

In view of the above, it is necessary to provide an online inquiry data processing method, an online inquiry data processing device and a computer device, which can improve the inquiry data processing efficiency.

A method of online interrogation data processing, the method comprising:

acquiring initial inquiry data corresponding to the user identification;

inquiring a matched target inquiry dialect subset from a pre-configured inquiry dialect set according to the initial inquiry data;

determining a target inquiry dialect according to the dialect path corresponding to the target inquiry dialect subset, and sending the target inquiry dialect to a corresponding terminal;

receiving target inquiry data fed back by the terminal aiming at the target inquiry dialect, and analyzing the target inquiry data to obtain a corresponding inquiry analysis result;

when the target inquiry data is judged to be professional term consultation data based on the inquiry analysis result, colloquial answer data matched with professional terms in the target inquiry data is inquired from a pre-configured professional term answer set, the colloquial answer data is fed back to the terminal, and the step of receiving the target inquiry data fed back by the terminal aiming at the target inquiry operation is continuously executed;

and when the target inquiry data is judged to be inquiry answer data based on the inquiry analysis result and the inquiry answer data meets the answer condition corresponding to the target inquiry dialect, returning to the step of determining the target inquiry dialect according to the dialect path corresponding to the target inquiry dialect subset and continuing to execute until each inquiry dialect in the target inquiry dialect subset is traversed.

In one embodiment, the querying a matching target questionnaire subset from a pre-configured questionnaire set according to the initial questionnaire data comprises:

inquiring a matched initial inquiry dialect subset from a pre-configured inquiry dialect set according to the initial inquiry data;

acquiring a user portrait corresponding to the user identifier;

when the user portrait meets the updating condition of the professional terms of the dialogies, acquiring spoken expressions matched with the professional terms in the initial inquiry and diagnosis tactics subset;

and updating the inquiry dialogs including the corresponding professional terms in the initial inquiry dialogs subset according to the spoken expressions to obtain a target inquiry dialogs subset.

In one embodiment, the method further comprises:

detecting a newly added high-frequency professional term in a preconfigured professional term spoken mapping set;

querying a spoken expression matched with each detected high-frequency professional term from the professional term spoken mapping set;

updating the interrogatories in the subset of interrogatories in the set of interrogatories comprising the corresponding high frequency professional terms according to each queried spoken expression.

In one embodiment, the method further comprises:

when the target inquiry data is judged to be the professional term inquiry data based on the inquiry analysis result, the target inquiry data is used as the inquiry data and is stored in a pre-configured professional term inquiry data set;

the detecting of the newly added high-frequency professional term in the preconfigured professional term spoken mapping set comprises:

carrying out statistical analysis on the consulting and inquiring data in the professional term consulting data set regularly according to a preset period to obtain consulting frequency corresponding to each professional term in the professional term consulting data set;

and determining the professional terms with the consultation frequency greater than or equal to the preset consultation frequency as the high-frequency professional terms detected from the pre-configured professional term spoken mapping set.

In one embodiment, after the feeding back the spoken answer data to the terminal, the method further includes:

when the preset waiting time is reached since the spoken answer data are fed back to the terminal, the target inquiry data fed back by the terminal aiming at the target inquiry dialogs are not received, the target inquiry dialogs are sent to the terminal again, and the step of returning to the step of receiving the target inquiry data fed back by the terminal aiming at the target inquiry dialogs is continuously executed.

In one embodiment, the method further comprises:

judging whether each inquiry tactical subset in the inquiry tactical set meets the updating condition of the tactical sentence pattern according to the tactical path;

updating the sentence pattern of the corresponding inquiry sentence in each inquiry sentence pattern subset according with the updating condition of the sentence pattern according to the updating mode of the pre-configured sentence pattern.

In one embodiment, the determining, according to the dialect path, whether each of the subset of interrogatories in the set of interrogatories complies with the dialect schema update condition includes:

sequentially calculating the text similarity between two adjacent interrogatories in each interrogatories talkaries subset in the interrogatories talkaries set according to the talkaries path to obtain a text similarity sequence corresponding to each interrogatories subset;

counting the number of continuous text similarities which are greater than or equal to a similarity threshold value in each text similarity sequence;

and when the number is greater than or equal to the number threshold value, judging that the corresponding inquiry dialect subset meets the updating condition of the dialect sentence pattern.

An online interrogation data processing apparatus, the apparatus comprising:

the inquiry data acquisition module is used for acquiring initial inquiry data corresponding to the user identification;

a query module for a subset of interrogatories, configured to query a matching subset of target interrogatories from a pre-configured set of interrogatories according to the initial interrogation data;

the inquiry-call-operation determining module is used for determining a target inquiry-call operation according to the operation path corresponding to the target inquiry-call-operation subset and sending the target inquiry-call operation to a corresponding terminal;

the inquiry data processing module is used for receiving target inquiry data fed back by the terminal aiming at the target inquiry dialect and analyzing the target inquiry data to obtain a corresponding inquiry analysis result;

the inquiry data processing module is further configured to, when it is determined that the target inquiry data is professional term inquiry data based on the inquiry analysis result, query spoken answer data matched with professional terms in the target inquiry data from a preconfigured professional term inquiry and answer set, feed back the spoken answer data to the terminal, and receive the target inquiry data fed back by the terminal for the target inquiry and diagnosis;

the inquiry data processing module is further configured to, when it is determined based on the inquiry analysis result that the target inquiry data is inquiry reply data and the inquiry reply data meets a reply condition corresponding to the target inquiry dialect, enable the inquiry dialect determining module to determine the target inquiry dialect according to the dialect path corresponding to the target inquiry dialect subset until each inquiry dialect in the target inquiry dialect subset is traversed.

A computer device comprising a memory storing a computer program and a processor implementing the steps of the above method embodiments when executing the computer program.

A computer storage medium having stored thereon a computer program which, when executed by a processor, carries out the steps in the above-described method embodiments.

After the initial inquiry data corresponding to the user identifier is obtained, dynamically determining a target inquiry word subset based on the initial inquiry data, sequentially sending a target inquiry word to the terminal according to the dynamically determined target inquiry word subset, and in the process of sequentially sending the target inquiry word, determining further operations to be executed based on an inquiry analysis result corresponding to the target inquiry data fed back by the terminal aiming at the current target inquiry word, when the target inquiry data is judged to be professional term inquiry data triggered by professional terms in the target inquiry word, obtaining spoken answer data matched with the professional terms specified by the professional term inquiry data based on a pre-configured professional term inquiry set, and feeding the spoken answer data back to the user through the terminal, and when the target inquiry data is judged to be inquiry answer data meeting the answer condition corresponding to the target inquiry call, determining the next target inquiry call according to the call route corresponding to the target inquiry call subset, and executing the processing flow aiming at the next target inquiry call until each inquiry call in the target inquiry call subset is traversed. Therefore, the target inquiry dialect subset is dynamically determined based on the initial inquiry data, namely the target inquiry dialect is dynamically determined, and the subsequent processing flow is dynamically determined based on the target inquiry data correspondingly fed back by the terminal, so that the quality of the inquiry dialect can be improved, and the processing efficiency of the inquiry data can be improved.

Drawings

FIG. 1 is a diagram illustrating an exemplary scenario for processing on-line interrogation data;

FIG. 2 is a schematic flow chart illustrating a method for processing on-line interrogation data in one embodiment;

FIG. 3 is a schematic flow chart illustrating a method for processing on-line interrogation data in another embodiment;

FIG. 4 is a block diagram of an on-line interrogation data processing apparatus according to an embodiment;

FIG. 5 is a diagram illustrating an internal structure of a computer device according to an embodiment.

Detailed Description

In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.

The online inquiry data processing method provided by the application can be applied to the application environment shown in fig. 1. Wherein the terminal 102 and the server 104 communicate via a network. The server 104 obtains initial inquiry data corresponding to the user identifier, queries a matched target inquiry dialect subset from a pre-configured inquiry dialect set according to the initial inquiry data, determines a target inquiry dialect according to a dialect path corresponding to the target inquiry dialect subset, sends the target inquiry dialect to the terminal 102, receives target inquiry data fed back by the terminal 102 aiming at the target inquiry dialect, analyzes the target inquiry data to obtain a corresponding inquiry analysis result, queries spoken answer data matched with professional terms in the target inquiry data from the pre-configured term professional inquiry set when the target inquiry data is judged to be professional term inquiry data based on the inquiry analysis result, sends the spoken answer data to the terminal 102, and returns to the receiving terminal 102 to continue execution of the target inquiry data fed back by the terminal 102 aiming at the target inquiry, and when the target inquiry data is judged to be inquiry answer data based on the inquiry analysis result and the inquiry answer data meets the answer condition corresponding to the target inquiry dialogs, returning to the step of determining the target inquiry dialogs according to the dialogs path corresponding to the target inquiry dialogs subset and continuing to execute until each inquiry dialogs in the target inquiry dialogs subset is traversed. The terminal 102 may be, but not limited to, various personal computers, notebook computers, smart phones, tablet computers, and portable wearable devices, and the server 104 may be implemented by an independent server or a server cluster formed by a plurality of servers.

In one embodiment, an intelligent inquiry client operates in the terminal 102, and the intelligent inquiry client is used as an AI (Artificial Intelligence) inquiry doctor to interact with the user so as to realize online intelligent inquiry. The relevant steps performed by the terminal 102 in the present application may be specifically performed by an intelligent inquiry client running on the terminal 102. In the intelligent inquiry process realized based on the intelligent inquiry client, the intelligent inquiry client processes corresponding inquiry data through the server.

In one embodiment, as shown in fig. 2, an online inquiry data processing method is provided, which is described by taking the method as an example applied to the server in fig. 1, and includes the following steps:

step 202, obtaining initial inquiry data corresponding to the user identifier.

The initial inquiry data refers to the inquiry data or inquiry data initially provided by the user during inquiry, that is, the inquiry data which is firstly acquired by the terminal and is sent by the user for a single user.

Specifically, the terminal dynamically detects initial inquiry data triggered by a user, the detected initial inquiry data are used as initial inquiry data corresponding to a user identifier of the user, and the initial inquiry data are sent to the server, so that the server can determine the current inquiry intention of the user based on the initial inquiry data, and further query a matched target inquiry dialect subset based on the inquiry intention.

In one embodiment, the terminal can detect initial inquiry data input by a user on the user operation interface, can also detect initial inquiry voice triggered by the user, and performs voice recognition on the detected initial inquiry voice to obtain corresponding initial inquiry data.

For example, the initial inquiry data may be, for example, "do i want to consult with i am i need to be supplemented with folic acid? ". The terminal may transmit the detected initial inquiry data to the server together with the corresponding user identification.

Step 204, the matched target interrogatories subset is queried from the pre-configured set of interrogatories according to the initial interrogation data.

The inquiry-call set comprises a plurality of inquiry-call subsets, each inquiry-call subset corresponds to one or more inquiry intents, so that the matched inquiry-call subset can be quickly positioned from the inquiry-call set based on the inquiry intents to serve as the target inquiry-call subset. Each inquiry talk operation subset comprises a plurality of inquiry talk operations, the plurality of inquiry talk operations have an arrangement sequence, and the talk operation path corresponding to the inquiry talk operation subset is used for appointing the arrangement sequence of the plurality of inquiry talk operations in the inquiry talk operation subset. For example, the interrogatories in the subset of interrogatories ordered according to the tactical path include in sequence: "does ask for a question during pregnancy? "," ask for questions during lactation? "," ask for question during the future pregnancy ".

Specifically, the server processes the acquired initial inquiry data to obtain a corresponding inquiry intention, inquires an inquiry talk subset matched with the inquiry intention from a pre-configured inquiry talk set according to the inquiry intention, and uses the inquiry talk subset as a target inquiry talk subset matched with the initial inquiry data.

In one embodiment, the server processes the initial inquiry data by using the existing natural language processing technology to obtain the corresponding inquiry intention, which is not described herein again. For example, the server performs semantic parsing on the initial inquiry data to obtain a corresponding inquiry intention.

In one embodiment, a plurality of inquiry talk subsets in the inquiry talk set are stored in a partition mode according to the disease types, and one or more inquiry talk subsets matched with the disease types are stored in the partition corresponding to each disease type. The server determines the disease type corresponding to the initial inquiry data, screens inquiry talk sub-sets matched with the disease type from the inquiry talk sub-sets according to the disease type, and queries matched target inquiry talk sub-sets from the screened inquiry talk sub-sets according to the inquiry intention corresponding to the initial inquiry data. In this way, the inquiry dialogue subsets in the inquiry dialogue set are stored in a partitioned mode according to the disease types, so that when the target inquiry dialogue subsets matched with the inquiry intentions are inquired according to the disease types corresponding to the initial inquiry data, the inquiry efficiency of the target inquiry dialogue subsets can be improved. The classification of the disease types can be customized according to actual conditions, for example, the disease types include traditional Chinese medicine and western medicine, and each department can be used as a disease type.

And step 206, determining the target inquiry dialogs according to the dialogs paths corresponding to the target inquiry dialogs subsets, and sending the target inquiry dialogs to corresponding terminals.

Wherein the phone path is used to specify a ranking order for each of the interrogatories in the corresponding subset of interrogatories. The term "phrase path" corresponding to the subset of interrogatories is understood to mean an sequence of interrogatories in the order in which the interrogatories in the subset of interrogatories are arranged.

Specifically, the server determines the not-traversed and most-ranked one of the target inquiry dialogs in the target inquiry dialogs subset as the current target inquiry dialogs according to the dialogs path corresponding to the target inquiry dialogs subset, and sends the currently determined target inquiry dialogs to the terminal. Therefore, after the target inquiry operation subset matched with the initial inquiry data is obtained in the above manner, if the target inquiry operation is determined from the target inquiry operation subset for the first time, the most forward-ranked inquiry operation in the corresponding inquiry path is determined as the target inquiry operation, that is, the first inquiry operation in the inquiry path is determined as the target inquiry operation, and if the target inquiry operation is not determined from the target inquiry operation subset for the first time, the most forward-ranked inquiry operation not traversed in the corresponding inquiry path is determined as the target inquiry operation, that is, the first inquiry operation after the previously-determined target inquiry operation in the inquiry path is determined as the current target inquiry operation.

For example, it is assumed that the dialogical paths corresponding to the target inquiry dialogical subset sequentially include: "does ask for a question during pregnancy? "," ask for questions during lactation? "," ask for question during the future pregnancy ". The server first asks the interviewing phrase "ask for a question during pregnancy? "determine as the target inquiry dialect to send to the terminal, and in the subsequent inquiry dialect feedback process, sequentially ask for the inquiry dialect" do you ask for questions during lactation? And the question is fed back to the terminal as a target inquiry dialogue.

In one embodiment, the terminal presents the received target inquiry dialogs to the corresponding user according to a preset presentation form. The predetermined presentation forms include, but are not limited to, text, animation, and voice. For example, the terminal converts the target inquiry dialogs into animations and displays the animations to the user through the intelligent inquiry interface. For example, the terminal converts the target inquiry dialogue into voice through voice synthesis, and broadcasts the voice to the user.

And step 208, receiving the target inquiry data fed back by the terminal aiming at the target inquiry dialect, and analyzing the target inquiry data to obtain a corresponding inquiry analysis result.

The target inquiry data is the inquiry data fed back by the user corresponding to the target inquiry dialogues displayed by the terminal, and specifically may be professional term consultation data or inquiry reply data triggered by the target inquiry dialogues. For example, when a term that cannot be understood by the user is included in the target interview, the user typically triggers the interview data of the term, and when a term that cannot be understood by the user is not included in the target interview, the user typically triggers the corresponding interview response data of the target interview.

Specifically, the terminal detects target inquiry data triggered by the user aiming at the currently displayed target inquiry dialogs, and sends the detected target inquiry data to the server. And after receiving the target inquiry data fed back by the terminal aiming at the target inquiry dialect, the server analyzes the received target inquiry data to obtain a corresponding inquiry analysis result.

In one embodiment, when the server parses the currently received target inquiry data, the server further obtains a corresponding inquiry parsing result in combination with the context understanding, so as to improve the accuracy of the inquiry parsing result.

In one embodiment, the server analyzes the received target inquiry data through the existing natural language processing method to obtain a corresponding inquiry analysis result, which is not described herein again.

Step 210, when the target inquiry data is judged to be the professional term inquiry data based on the inquiry analysis result, the spoken answer data matched with the professional terms in the target inquiry data is inquired from the pre-configured professional term inquiry and answer set, the spoken answer data is fed back to the terminal, and the step 208 is returned to continue to be executed.

Wherein, the professional term consulting data refers to consulting data triggered by the professional term for consulting the meaning or colloquial expression of the professional term. The term-specific question-answer set is a question-answer set for specifying the corresponding relationship between a specific term and a spoken expression, and specifically may include a spoken expression corresponding to each specific term by a plurality of specific terms, or include candidate question-answer data pre-configured for each specific term in a plurality of specific terms, and spoken answer data pre-configured based on the spoken expression corresponding to each specific term, where the candidate question-answer data corresponds to the spoken answer data, that is, the candidate question-answer data and the spoken answer data constitute a question-answer pair. The spoken answer data interprets the corresponding professional term by a spoken expression.

Specifically, when the target inquiry data is judged to be the professional term consultation data triggered by the user aiming at the target inquiry dialogues based on the inquiry analysis result, the fact that the user does not understand at least one professional term in the target inquiry dialogues is indicated, the server extracts the professional terms which cannot be understood by the user from the target inquiry data, queries spoken answer data matched with the extracted professional terms from a pre-configured professional term inquiry set, and feeds the spoken answer data back to the terminal for displaying so as to explain the unappreciated professional terms to the user. And the terminal displays the spoken answer data to the user, detects target inquiry data further triggered by the user aiming at the corresponding target inquiry dialogue on the basis of the spoken answer data, and sends the detected target inquiry data to the server. And the server analyzes the received target inquiry data to obtain a corresponding inquiry analysis result, and when the corresponding target inquiry data is judged to be professional term inquiry data based on the inquiry analysis result, the server continues to execute the related steps of inquiring and feeding back spoken answer data.

It can be understood that, after the server displays the spoken answer data to the user through the terminal, the user may trigger corresponding inquiry answer data for the corresponding target inquiry dialogues on the basis of understanding the professional terms, and may further trigger corresponding professional term consultation data for the corresponding target inquiry dialogues or the received spoken answer data. When detecting inquiry answer data or professional term consultation data triggered by a user, the terminal sends the detected inquiry answer data or professional term consultation data as target inquiry data to the server so that the server can execute corresponding operation based on the received target inquiry data.

For example, suppose the target inquiry words sent by the server to the terminal are "ask for a question during ovulation? "what is the meaning of the ovulation period requested? "indicating that the user does not understand the term" ovulation period "in the target inquiry dialogue, the server can determine that the target inquiry data is the term inquiry data based on the inquiry analysis result of the target inquiry data, and query spoken answer data matched with the term from the term inquiry answer set, where the queried spoken answer data is, for example," normal menstruation ", the female ovulation period is counted from the first day of the next menstruation, 14 days before last is the ovulation day, and 5 days before and 4 days after the ovulation day are collectively referred to as the ovulation period", the server feeds back the spoken answer data to the user through the terminal, and on the basis of understanding the term "ovulation period", the user does not understand the term "ovulation period" for the target inquiry "at the ovulation period? Further trigger the corresponding target interrogation data.

In one embodiment, the server may match the professional terms in the target inquiry data with the professional terms in the professional term question-answer set, obtain the matched spoken expressions from the professional term question-answer set according to the matching result, and feed the spoken expressions back to the terminal as spoken answer data. The server can also calculate the text similarity between the target inquiry data and each candidate inquiry data in the professional term inquiry and answer set, determine the spoken answer data corresponding to the candidate inquiry data with the highest text similarity as the final spoken answer data, and feed back the final spoken answer data to the terminal.

In one embodiment, the terms included in the term of expertise set are low frequency terms. The server dynamically detects new low-frequency professional terms and high-frequency professional terms in the preconfigured professional term spoken mapping set, updates the preconfigured professional term question-answer set according to the new low-frequency professional terms, and updates the preconfigured inquiry call-art set according to the new high-frequency professional terms. The term of the low frequency term refers to a term of the low frequency of consulting, that is, a term of the low frequency triggering the term of consulting data, and correspondingly, the term of the high frequency term refers to a term of the high frequency of consulting, that is, a term of the high frequency triggering the term of consulting data. For example, a term of art having a consultation frequency greater than or equal to a preset consultation frequency may be determined as a high-frequency term of art, and a term of art having a consultation frequency less than the preset consultation frequency may be determined as a low-frequency term of art.

Step 212, when the target inquiry data is determined to be inquiry answer data based on the inquiry analysis result and the inquiry answer data meets the answer condition corresponding to the target inquiry dialogs, returning to step 206 to continue executing until each inquiry dialogs in the target inquiry dialogs subset is traversed.

The inquiry response data refers to data triggered by the target inquiry dialogs and representing responses to the target inquiry dialogs. The answer condition refers to a basis or condition, which is pre-configured for the target inquiry dialogs, for determining whether the target inquiry data is answer data triggered by the user for the target inquiry dialogs. For example, assume that the target interrogation is "ask for a question during ovulation? "accordingly, the answer condition is" yes "or" no ", or answer data associated with the targeted interview. If the target inquiry data corresponding to the target inquiry dialect is yes, no or the target inquiry data is the first day of ovulation, the target inquiry data can be judged to be the inquiry answer data meeting the answer condition corresponding to the target inquiry dialect.

Specifically, when the target inquiry data is determined to be inquiry answer data triggered by the target inquiry dialect based on the inquiry analysis result of the target inquiry data and the inquiry answer data meets the answer condition corresponding to the target inquiry dialect, it indicates that the user has understood the meaning of the target inquiry dialect, and the server triggers the inquiry answer data meeting the answer condition of the target inquiry dialect according to the target inquiry dialect, determines the next target inquiry dialect from the target inquiry dialect subset according to the dialect path corresponding to the target inquiry dialect subset, and sends the determined target inquiry dialect to the terminal, and then when target inquiry data fed back by the terminal aiming at the target inquiry dialogs are received, executing the related processes based on the target inquiry data until each inquiry dialogs in the target inquiry dialogs subset are traversed.

In one embodiment, when the target inquiry data is judged to be inquiry answer data based on the inquiry analysis result, but the inquiry answer data does not meet the answer condition corresponding to the target inquiry dialect, the answer of the user is indicated, the server triggers corresponding prompt information, and the prompt information is fed back to the user through the terminal.

After the initial inquiry data corresponding to the user identification is obtained, dynamically determining a target inquiry dialect subset based on the initial inquiry data, sequentially sending the target inquiry dialect to the terminal according to the dynamically determined target inquiry dialect subset, and in the process of sequentially sending the target inquiry dialect, determining further required operation based on an inquiry analysis result corresponding to the target inquiry data fed back by the terminal aiming at the current target inquiry dialect, and when the target inquiry data is judged to be the professional term inquiry data triggered by the professional terms in the target inquiry, obtaining spoken answer data matched with the professional terms specified by the professional term inquiry data based on a pre-configured professional term inquiry set, and feeding the spoken answer data back to the user through the terminal so that the user can further answer the current target inquiry dialect based on the spoken answer data, therefore, the target inquiry data fed back by the terminal is correspondingly processed according to the process, when the target inquiry data is judged to be inquiry answer data meeting the answer condition corresponding to the target inquiry dialogs, the next target inquiry dialogs are determined according to the dialogs path corresponding to the target inquiry dialogs subset, and the processing process is executed according to the next target inquiry dialogs until each inquiry dialogs in the target inquiry dialogs subset is traversed. Therefore, the target inquiry dialect subset is dynamically determined based on the initial inquiry data, namely the target inquiry dialect is dynamically determined, and the subsequent processing flow is dynamically determined based on the target inquiry data correspondingly fed back by the terminal, so that the quality of the inquiry dialect can be improved, and the processing efficiency of the inquiry data can be improved.

In one embodiment, step 204 comprises: querying a matched initial inquiry dialect subset from a pre-configured inquiry dialect set according to the initial inquiry data; acquiring a user portrait corresponding to a user identifier; when the user portrait meets the updating condition of the professional terms of the dialogies, acquiring spoken expressions matched with the professional terms in the initial inquiry and diagnosis dialogies subset; and updating the inquiry dialogs including the corresponding professional terms in the initial inquiry dialogs subset according to the spoken expressions to obtain a target inquiry dialogs subset.

The term updating condition for the dialogistic specialty term is a basis or condition for updating the specialty term in the dialogistic art, such as the age of the user is greater than a first preset age, or the age of the user is less than a second preset age, or the user is a remote mountain area, which is not listed here. A first predetermined age, such as 70 years, and a second predetermined age, such as 10 years. The user representation is used to describe characteristics of the user, and may specifically include, but is not limited to, the age of the user, the gender of the user, the occupation of the user, the place where the user lives, and the like.

Specifically, the server queries an initial inquiry dialect subset matched with the initial inquiry data from a pre-configured inquiry dialect set according to an inquiry intention corresponding to the initial inquiry data. Correspondingly, after the server acquires the initial inquiry data corresponding to the user identification, the server acquires the user portrait corresponding to the user identification and compares the user portrait with the pre-configured updating conditions of the jargon technical terms. When the user image is judged to meet the language professional term updating condition, the server extracts professional terms from the initial inquiry and diagnosis language subset and inquires a spoken expression matched with each professional term from a pre-configured professional term spoken mapping set. And the server updates the inquiry dialogs including the corresponding professional terms in the initial inquiry dialogs subset according to each inquired spoken expression to obtain a target inquiry dialogs subset.

In one embodiment, the step of updating, by the server, the corresponding interrogatories in the initial subset of interrogatories according to the spoken utterance comprises: and replacing the professional terms in the inquiry dialogs with corresponding spoken expressions, or interpreting the spoken expressions as remarks of the corresponding professional terms and adding the remarks into the corresponding inquiry dialogs. For example, the initial inquiry phrase is "ask for a question during ovulation period", and the updated inquiry phrase is "ask for a question during ovulation period? Under normal menstruation, the ovulation period of a female is counted from the first day of the next menstruation, 14 days after the last menstruation is the ovulation day, and the first 5 days and the last 4 days of the ovulation day are collectively called the ovulation period ".

In the above embodiment, the inquiry dialogs in the initial inquiry dialogs subset matched with the initial inquiry data are dynamically updated based on the user images, so that different dialogs professional term updating processes are triggered for users with different user images, that is, the inquiry dialogs in the initial inquiry dialogs subset are dynamically adjusted for users with different user images, and a target inquiry dialogs subset suitable for the user images is obtained, so that when a subsequent inquiry data processing process is executed based on the target inquiry dialogs subset, the inquiry data processing efficiency and the inquiry efficiency can be improved while the inquiry dialogs quality is ensured.

In one embodiment, the above online inquiry data processing method further includes: detecting a newly added high-frequency professional term in a preconfigured professional term spoken mapping set; inquiring the spoken expressions matched with each detected high-frequency professional term from the professional term spoken mapping set; updating the interrogatories in the subset of interrogatories in the set of interrogatories that include the corresponding high frequency professional term according to each spoken utterance queried.

The high-frequency professional term refers to a professional term with higher frequency for triggering the professional term to consult data, namely, a professional term in cognitive blind areas of most users. The professional term spoken mapping set comprises a plurality of professional terms and spoken expressions corresponding to the professional terms, and mapping relations are established between the professional terms and the corresponding spoken expressions, so that the corresponding spoken expressions can be quickly determined based on the professional terms.

Specifically, the server dynamically detects a high-frequency professional term newly added in the professional term spoken mapping set, and when the high-frequency professional term newly added is detected, the server inquires spoken expressions matched with each high-frequency professional term detected by the server from the professional term spoken mapping set. The server queries an interview subset including at least one high-frequency term from a preconfigured interview corpus and updates a corresponding interview in the interview subset including the high-frequency term in the interview corpus according to the spoken utterance queried for each high-frequency term.

In one embodiment, since there is a case that medical professional terms of doctors are not adapted to the cognitive level of users in the intelligent inquiry, by establishing the professional term spoken mapping set, the professional terms can be converted into spoken expressions which are easy to understand by the users, so that the users can understand the professional terms conveniently. It can be understood that the high-frequency professional terms have a higher priority in the professional term spoken language mapping set, and the high-frequency professional terms in the interviewing terminology set can be remarked and explained or directly replaced in advance based on the spoken language expression corresponding to the high-frequency professional terms, while the low-frequency professional terms have a lower priority in the professional term spoken language mapping set, and the timing for updating or pushing the spoken language expression corresponding to the low-frequency professional terms can be dynamically determined based on the user image or the target interviewing data fed back by the user for the target interviewing terminology in the process of interviewing data processing.

In the above embodiment, the high-frequency professional terms in the inquiry dialogs set are updated to the corresponding spoken expressions in advance, so that the inquiry dialogs quality can be improved, and meanwhile, the time for a user to think and ask can be reduced, thereby reducing the time consumed by unit inquiry, improving the effective inquiry amount of the unit, and further improving the inquiry data processing efficiency and the inquiry efficiency.

In one embodiment, the above online inquiry data processing method further includes: when the target inquiry data is judged to be the professional term inquiry data based on the inquiry analysis result, the target inquiry data is used as the inquiry data and is stored in a pre-configured professional term inquiry data set; detecting newly added high-frequency professional terms in a pre-configured professional term spoken mapping set, wherein the high-frequency professional terms comprise: carrying out statistical analysis on the consultation and inquiry data in the professional term consultation data set regularly according to a preset period to obtain consultation frequency corresponding to each professional term in the professional term consultation data set; and determining the professional terms with the consultation frequency greater than or equal to the preset consultation frequency as the high-frequency professional terms detected from the pre-configured professional term spoken mapping set.

Wherein, the professional term consulting data set is a set composed of a plurality of pieces of consulting and consulting data, and one piece of consulting and consulting data is a piece of target consulting and consulting data with the type of the professional term consulting and consulting data. The consulting frequency refers to the frequency of occurrence of consulting inquiry data corresponding to the professional term, that is, the frequency of triggering the professional term consulting data for the professional term. The preset period refers to a time interval between two adjacent high-frequency professional term detection operations, such as 1 month or 1 year, and is particularly customizable. The preset consultation frequency may be specifically customized according to a statistical dimension of the consultation frequency, for example, the consultation frequency is 50 times of consultation in a day, or the consultation frequency is 1000 times of consultation in a preset period, which is not specifically limited herein.

Specifically, in the process of processing the inquiry data based on the target inquiry dialect subset, when the target inquiry data is judged to be the professional term inquiry data based on the inquiry analysis result corresponding to the target inquiry data, the server stores the target inquiry data as the inquiry data into the pre-configured professional term inquiry data set. Further, the server periodically performs statistical analysis on the consultation inquiry data in the preconfigured term consultation data set according to a preset period to obtain a consultation frequency corresponding to each term included in the term consultation data set, and compares the consultation frequency corresponding to each term with the preconfigured preset consultation frequency. And when the consultation frequency corresponding to the professional term is judged to be greater than or equal to the preset consultation frequency, the server determines the professional term as a high-frequency professional term newly added in the professional term spoken mapping set.

In one embodiment, when the server periodically performs statistical analysis on the consultation and inquiry data in the professional term consultation data set according to a preset period, the server performs statistical analysis on newly-added consultation and inquiry data in the current preset period, and determines the obtained professional terms with consultation frequency greater than or equal to the preset consultation frequency as the newly-added high-frequency professional terms in the professional term spoken mapping set.

In one embodiment, in the process of processing the inquiry data based on the target inquiry dialect subset, when the target inquiry data is determined to be the professional term inquiry data based on the inquiry analysis result corresponding to the target inquiry data, the server extracts the professional terms from the target inquiry data and dynamically updates the inquiry frequency corresponding to the professional terms. In this way, the server periodically detects the consultation frequency corresponding to each term in the term spoken mapping set according to a preset period, and determines a high-frequency term based on the detected consultation frequency.

In the above embodiment, the consultation frequency of the professional terms is periodically detected according to the preset period, and the high-frequency professional terms are dynamically determined based on the consultation frequency, so that the inquiry dialogs including the high-frequency professional terms in the inquiry dialogs set are periodically updated, and the inquiry data processing efficiency can be improved.

In one embodiment, after the spoken answer data is fed back to the terminal, the online inquiry data processing method further includes: when the preset waiting time is reached since the spoken answer data is fed back to the terminal, the target inquiry data fed back by the terminal for the target inquiry dialect is not received, the target inquiry dialect is sent to the terminal again, and the step 208 is returned to continue to execute.

The preset waiting time is a preset time length for waiting the terminal to feed back the target inquiry data, and can be specifically defined by user, for example, 1 minute.

Specifically, after the server feeds back spoken answer data queried for the professional terms in the target inquiry data to the terminal, the real-time detection terminal further feeds back the target inquiry data for the target inquiry dialogue based on the spoken answer data, counts the waiting time for the waiting terminal to feed back the target inquiry data, and compares the counted waiting time with the preset waiting time. And when the statistical waiting time is judged to be greater than or equal to the preset waiting time and the server does not receive the target inquiry data fed back by the terminal aiming at the target inquiry dialogues, the target inquiry dialogues are sent to the terminal again, the target inquiry data fed back by the terminal aiming at the target inquiry dialogues which are sent again are received, and then the relevant inquiry data processing flow is executed based on the received target inquiry data.

In the above embodiment, when the preset waiting time is reached after the spoken data is fed back to the terminal, the target inquiry data fed back by the terminal for the target inquiry dialect is not received yet, which indicates that the user does not respond to the target inquiry dialect for a long time, and then the target inquiry data is sent to the user again, so that unnecessary waiting time is avoided from being wasted, and the inquiry efficiency can be improved.

In one embodiment, the above online inquiry data processing method further includes: judging whether each inquiry tactical subset in the inquiry tactical set meets the updating condition of the tactical sentence pattern according to the tactical path; updating the sentence pattern of the corresponding inquiry sentence in each inquiry sentence pattern subset according with the updating condition of the sentence pattern according to the updating mode of the pre-configured sentence pattern.

The query dialect schema updating condition is a condition or basis for determining whether to update the schema of the query dialect in the query dialect subset, and may specifically refer to that a preset number of continuous query dialects with similar schemas exist in the query dialect subset, for example, that the schemas with 3 or more continuous query dialects are similar. The phrase updating method refers to a method of updating the inquiry phrase in the subset of inquiry phrases, such as updating through a trained phrase updating model, and updating based on a pre-configured phrase updating template.

Specifically, for each of the pre-configured inquiry tactical subsets, whether the inquiry tactical subset meets the updating condition of the tactical sentence pattern is determined according to the tactical path corresponding to the inquiry tactical subset. And when the inquiry dialect subset is judged to meet the updating condition of the dialect sentence pattern, updating the sentence pattern of the corresponding inquiry dialect in the inquiry dialect subset according to the preset dialect sentence pattern updating mode.

In one embodiment, the server may update the sentence patterns of the corresponding interrogatories in the interrogatories ' sentence pattern subset through the trained sentence pattern update model, and may also query a pre-configured sentence pattern update template based on the interrogatories ' sentence pattern subset to be updated, and update the sentence patterns of the corresponding interrogatories in the interrogatories ' sentence pattern subset based on the queried sentence pattern update template, such as querying a consistent number of sentence pattern update templates based on the number of consecutive interrogatories similar to the sentence pattern.

In the above embodiment, the sentence patterns of the corresponding inquiry dialects are updated in advance for the inquiry dialects subset satisfying the update condition of the sentence pattern of the dialects, so that the situation that the inquiry dialects in the inquiry dialects subset are too hard, mechanical and monotonous and the quality of the dialects is reduced is avoided, and the processing efficiency of the inquiry data can be improved.

In one embodiment, determining whether each of the subset of interrogatories in the set of interrogatories according to the phonetics route meets the phonetics sentence pattern updating condition includes: sequentially calculating the text similarity between two adjacent interrogatories in each interrogatories talkaries subset in the interrogatories talkaries set according to the talkaries path to obtain a text similarity sequence corresponding to each interrogatories subset; counting the number of continuous text similarities which are greater than or equal to a similarity threshold value in each text similarity sequence; and when the quantity is greater than or equal to the quantity threshold value, judging that the corresponding inquiry dialect subset meets the updating condition of the dialect sentence pattern.

The text similarity is used for representing the similarity degree between two inquiry dialects, and particularly can be used for representing the similarity degree of sentence patterns of the two inquiry dialects. The text similarity sequence is a sequence formed by a plurality of text similarity arrangements, each inquiry talk operation subset corresponds to one text similarity sequence, the arrangement sequence of each text similarity in the text similarity sequence is determined by the arrangement sequence of the corresponding inquiry talk operation in the inquiry talk operation subset, for example, the text similarity calculated by the first inquiry talk operation and the second inquiry talk operation in the inquiry talk operation subset is at the head of the text similarity sequence corresponding to the inquiry talk operation subset. The similarity threshold may be customized, such as 80%. The quantity threshold may be customized, such as 3.

Specifically, for each inquiry-dialect subset in the inquiry-dialect subset, the server sequentially calculates the text similarity between two adjacent inquiry dialects in the inquiry-dialect subset according to the dialect path corresponding to the inquiry-dialect subset, obtains the text similarity between any two adjacent inquiry dialects in the inquiry-dialect subset, and obtains the text similarity sequence corresponding to the inquiry-dialect subset according to the obtained text similarity and the arrangement sequence of the corresponding inquiry dialects in the inquiry-dialect subset. Further, the server counts a number of consecutive text similarities greater than or equal to a similarity threshold in each text similarity sequence, and compares the counted number of consecutive text similarities with a preconfigured number threshold. And when the continuous text similarity number obtained by counting aiming at the text similarity sequence is greater than the number threshold value, the server judges that the inquiry dialect subset corresponding to the text similarity sequence conforms to the updating condition of the dialect sentence pattern.

For example, assume that the initial interview phonetics subset corresponds to a phonetics path of: "does ask for a question during pregnancy? "," ask for questions during lactation? "," ask for questions during the future pregnancy? If the number threshold is 3, the number of similarity degrees of the continuous text corresponding to the subset of the interrogatories is determined to be 3 based on the text similarity degrees, and since the number of similarity degrees of the continuous text is consistent with the number threshold, the interrogatories subset needs to be updated with the dialect sentence pattern, for example, the updated interrogatories subset is: "does ask for a question during pregnancy? "," is that during lactation? "," preparatory pregnancy worshipping? ". It can be appreciated that if the initial subset of interrogatories is assumed to correspond to a dialogistic path: "does ask for a question during pregnancy? "," ask for questions during lactation? "," is a preparatory pregnancy? ", since the number of similarity of the continuous text corresponding to the sub-set of the inquiry operation is 2, which is smaller than the number threshold value 3, it is determined that the sub-set of the inquiry operation does not conform to the updating condition of the dialect sentence pattern.

In the above embodiment, the text similarity sequence is obtained according to the dialect path corresponding to the dialect subset, and whether the dialect subset meets the dialect sentence pattern updating condition is determined based on the text similarity sequence, so as to update the sentence pattern of the corresponding dialect for the dialect subset meeting the dialect sentence pattern updating condition. Thus, when the sentence patterns of a plurality of continuous inquiry dialects in the dialect path are similar or consistent, the sentence patterns of the corresponding inquiry dialects in the corresponding inquiry dialect subset are updated, so as to improve the quality of the inquiry dialects.

As shown in fig. 3, in an embodiment, an online inquiry data processing method is provided, which specifically includes the following steps:

step 302, obtaining initial inquiry data corresponding to the user identifier.

In step 304, a matching initial interrogatories subset is queried from the pre-configured set of interrogatories based on the initial interrogation data.

Step 306, obtaining the user portrait corresponding to the user identification.

Step 308, when the user image meets the language professional term updating condition, obtaining the spoken language expression matched with the professional term in the initial inquiry language skill subset.

Step 310, updating the inquiry dialogs including the corresponding professional terms in the initial inquiry dialogs subset according to the spoken expressions to obtain the target inquiry dialogs subset.

Step 312, determining the target inquiry dialogs according to the dialogs path corresponding to the target inquiry dialogs subset, and sending the target inquiry dialogs to the corresponding terminals.

And step 314, receiving the target inquiry data fed back by the terminal aiming at the target inquiry dialect, and analyzing the target inquiry data to obtain a corresponding inquiry analysis result.

And step 316, when the target inquiry data is judged to be the professional term inquiry data based on the inquiry analysis result, inquiring spoken answer data matched with the professional terms in the target inquiry data from a pre-configured professional term inquiry and answer set, feeding the spoken answer data back to the terminal, and when the preset waiting time is reached from the time of feeding the spoken answer data back to the terminal, not receiving the target inquiry data fed back by the terminal aiming at the target inquiry dialogue, re-sending the target inquiry dialogue to the terminal, and returning to the step 314 to continue execution.

Step 318, when the target inquiry data is judged to be inquiry answer data based on the inquiry analysis result and the inquiry answer data meets the answer condition corresponding to the target inquiry dialogs, returning to step 312 to continue executing until each inquiry dialogs in the target inquiry dialogs subset is traversed.

And 320, detecting the newly added high-frequency professional term in the pre-configured professional term spoken mapping set.

Step 322, from the spoken term mapping set, queries the spoken expressions matching each detected high frequency professional term.

Step 324, updating the interrogatories in the subset of interrogatories in the set of interrogatories that include the corresponding high frequency professional term according to each spoken utterance queried.

In the above embodiment, for a high-frequency professional term in the professional term spoken mapping set, according to a spoken expression corresponding to the high-frequency professional term, a corresponding inquiry operation in an inquiry operation subset including the high-frequency professional term in the inquiry operation set is updated in advance. Further, in the inquiry data processing process based on the target inquiry word operation subset, if the user image meets the word operation professional term updating condition, the corresponding inquiry word operation in the target inquiry word operation subset is dynamically updated according to the spoken expression corresponding to the professional terms in the target inquiry word operation subset, if the user image does not meet the word operation professional term updating condition, the target inquiry word operation is sequentially determined according to the word operation path and is sent to the terminal, and the corresponding inquiry data processing flow is executed according to the target inquiry data corresponding to the target inquiry word operation based on the terminal, so that the corresponding inquiry data processing flow is provided for different conditions, and the inquiry data processing efficiency can be improved.

It should be understood that although the various steps in the flow charts of fig. 2-3 are shown in order as indicated by the arrows, the steps are not necessarily performed in order as indicated by the arrows. The steps are not performed in the exact order shown and described, and may be performed in other orders, unless explicitly stated otherwise. Moreover, at least some of the steps in fig. 2-3 may include multiple sub-steps or multiple stages that are not necessarily performed at the same time, but may be performed at different times, and the order of performance of the sub-steps or stages is not necessarily sequential, but may be performed in turn or alternating with other steps or at least some of the sub-steps or stages of other steps.

In one embodiment, as shown in fig. 4, there is provided an online interrogation data processing apparatus 400 comprising: an inquiry data acquisition module 401, a dialect subset query module 402, an inquiry dialect determination module 403, and an inquiry data processing module 404, wherein:

an inquiry data obtaining module 401, configured to obtain initial inquiry data corresponding to the user identifier;

a subset of interrogatories query module 402 for querying a matching target subset of interrogatories from a pre-configured set of interrogatories according to the initial interrogation data;

an inquiry-dialect determining module 403, configured to determine a target inquiry-dialect according to the dialect path corresponding to the target inquiry-dialect subset, and send the target inquiry-dialect to the corresponding terminal;

the inquiry data processing module 404 is configured to receive target inquiry data fed back by the terminal for the target inquiry dialect, and analyze the target inquiry data to obtain a corresponding inquiry analysis result;

the inquiry data processing module 404 is further configured to, when it is determined that the target inquiry data is the professional term inquiry data based on the inquiry analysis result, query spoken answer data matched with the professional terms in the target inquiry data from a preconfigured professional term inquiry-answer set, feed back the spoken answer data to the terminal, and receive the target inquiry data fed back by the terminal for the target inquiry;

the inquiry data processing module 404 is further configured to, when it is determined that the target inquiry data is inquiry reply data based on the inquiry analysis result and the inquiry reply data meets a reply condition corresponding to the target inquiry dialect, enable the inquiry dialect determining module 403 to determine the target inquiry dialect according to the dialect path corresponding to the target inquiry dialect subset until each inquiry dialect in the target inquiry dialect subset is traversed.

In one embodiment, the dialect subset query module 402 is further configured to query a matching initial interrogation dialect subset from a pre-configured set of interrogatories based on the initial interrogation data; acquiring a user portrait corresponding to a user identifier; when the user portrait meets the updating condition of the professional terms of the dialogies, acquiring spoken expressions matched with the professional terms in the initial inquiry and diagnosis dialogies subset; and updating the inquiry dialogs including the corresponding professional terms in the initial inquiry dialogs subset according to the spoken expressions to obtain a target inquiry dialogs subset.

In one embodiment, the above-mentioned online inquiry data processing apparatus 400 further includes: a dialect subset update module;

the dialect subset updating module is used for detecting a newly added high-frequency professional term in the preconfigured professional term spoken mapping set; inquiring the spoken expressions matched with each detected high-frequency professional term from the professional term spoken mapping set; updating the interrogatories in the subset of interrogatories in the set of interrogatories that include the corresponding high frequency professional term according to each spoken utterance queried.

In one embodiment, the inquiry data processing module 404 is further configured to, when it is determined that the target inquiry data is the term inquiry data based on the inquiry analysis result, use the target inquiry data as the inquiry data and store the inquiry data in a preconfigured term inquiry data set; the language and technology subset updating module is also used for periodically carrying out statistical analysis on the consultation and inquiry data in the professional term consultation data set according to a preset period to obtain the consultation frequency corresponding to each professional term in the professional term consultation data set; and determining the professional terms with the consultation frequency greater than or equal to the preset consultation frequency as the high-frequency professional terms detected from the pre-configured professional term spoken mapping set.

In one embodiment, the inquiry data processing module 404 is further configured to, when a preset waiting time is reached since the spoken answer data is fed back to the terminal, not receive the target inquiry data fed back by the terminal for the target inquiry dialogue, resend the target inquiry dialogue to the terminal, and receive the target inquiry data fed back by the terminal for the target inquiry dialogue.

In one embodiment, the dialect subset updating module is further configured to determine, according to the dialect path, whether each of the interrogatories in the interrogatories dialect set conforms to the dialect schema updating condition; updating the sentence pattern of the corresponding inquiry sentence in each inquiry sentence pattern subset according with the updating condition of the sentence pattern according to the updating mode of the pre-configured sentence pattern.

In one embodiment, the dialect subset updating module is further configured to sequentially calculate a text similarity between two adjacent dialects in each of the inquiry dialect subsets in the inquiry dialect set according to the dialect path, and obtain a text similarity sequence corresponding to each inquiry dialect subset; counting the number of continuous text similarities which are greater than or equal to a similarity threshold value in each text similarity sequence; and when the quantity is greater than or equal to the quantity threshold value, judging that the corresponding inquiry dialect subset meets the updating condition of the dialect sentence pattern.

For the specific limitations of the on-line inquiry data processing device, reference may be made to the above limitations of the on-line inquiry data processing method, which are not described herein again. The modules in the above-mentioned on-line inquiry data processing device can be wholly or partially realized by software, hardware and a combination thereof. The modules can be embedded in a hardware form or independent from a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.

In one embodiment, a computer device is provided, which may be a server, the internal structure of which may be as shown in fig. 5. The computer device includes a processor, a memory, a network interface, and a database connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The database of the computer device is used to store a pre-configured set of interrogatories and a set of terms of expertise. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement a method of on-line interrogation data processing.

Those skilled in the art will appreciate that the architecture shown in fig. 5 is merely a block diagram of some of the structures associated with the disclosed aspects and is not intended to limit the computing devices to which the disclosed aspects apply, as particular computing devices may include more or less components than those shown, or may combine certain components, or have a different arrangement of components.

In one embodiment, there is provided a computer device comprising a memory storing a computer program and a processor implementing the following steps when the processor executes the computer program: acquiring initial inquiry data corresponding to the user identification; querying a matched target interrogatory dialect subset from a pre-configured interrogatory dialect set according to the initial interrogation data; determining a target inquiry dialect according to the dialect path corresponding to the target inquiry dialect subset, and sending the target inquiry dialect to a corresponding terminal; receiving target inquiry data fed back by the terminal aiming at the target inquiry dialogs, and analyzing the target inquiry data to obtain a corresponding inquiry analysis result; when the target inquiry data is judged to be the professional term inquiry data based on the inquiry analysis result, the spoken answer data matched with the professional terms in the target inquiry data is inquired from a pre-configured professional term inquiry set, the spoken answer data is fed back to the terminal, and the step of returning to the target inquiry data fed back by the receiving terminal aiming at the target inquiry tactics is continuously executed; and when the target inquiry data is judged to be inquiry answer data based on the inquiry analysis result and the inquiry answer data meets the answer condition corresponding to the target inquiry dialogs, returning to the step of determining the target inquiry dialogs according to the dialogs path corresponding to the target inquiry dialogs subset and continuing to execute until each inquiry dialogs in the target inquiry dialogs subset is traversed.

In one embodiment, the processor, when executing the computer program, further performs the steps of: querying a matched initial inquiry dialect subset from a pre-configured inquiry dialect set according to the initial inquiry data; acquiring a user portrait corresponding to a user identifier; when the user portrait meets the updating condition of the professional terms of the dialogies, acquiring spoken expressions matched with the professional terms in the initial inquiry and diagnosis dialogies subset; and updating the inquiry dialogs including the corresponding professional terms in the initial inquiry dialogs subset according to the spoken expressions to obtain a target inquiry dialogs subset.

In one embodiment, the processor, when executing the computer program, further performs the steps of: detecting a newly added high-frequency professional term in a preconfigured professional term spoken mapping set; inquiring the spoken expressions matched with each detected high-frequency professional term from the professional term spoken mapping set; updating the interrogatories in the subset of interrogatories in the set of interrogatories that include the corresponding high frequency professional term according to each spoken utterance queried.

In one embodiment, the processor, when executing the computer program, further performs the steps of: when the target inquiry data is judged to be the professional term inquiry data based on the inquiry analysis result, the target inquiry data is used as the inquiry data and is stored in a pre-configured professional term inquiry data set; carrying out statistical analysis on the consultation and inquiry data in the professional term consultation data set regularly according to a preset period to obtain consultation frequency corresponding to each professional term in the professional term consultation data set; and determining the professional terms with the consultation frequency greater than or equal to the preset consultation frequency as the high-frequency professional terms detected from the pre-configured professional term spoken mapping set.

In one embodiment, the processor, when executing the computer program, further performs the steps of: and when the preset waiting time is reached since the spoken answer data are fed back to the terminal, the target inquiry data fed back by the terminal aiming at the target inquiry dialogs are not received, the target inquiry dialogs are sent to the terminal again, and the step of returning to the target inquiry data fed back by the receiving terminal aiming at the target inquiry dialogs is continuously executed.

In one embodiment, the processor, when executing the computer program, further performs the steps of: judging whether each inquiry tactical subset in the inquiry tactical set meets the updating condition of the tactical sentence pattern according to the tactical path; updating the sentence pattern of the corresponding inquiry sentence in each inquiry sentence pattern subset according with the updating condition of the sentence pattern according to the updating mode of the pre-configured sentence pattern.

In one embodiment, the processor, when executing the computer program, further performs the steps of: sequentially calculating the text similarity between two adjacent interrogatories in each interrogatories talkaries subset in the interrogatories talkaries set according to the talkaries path to obtain a text similarity sequence corresponding to each interrogatories subset; counting the number of continuous text similarities which are greater than or equal to a similarity threshold value in each text similarity sequence; and when the quantity is greater than or equal to the quantity threshold value, judging that the corresponding inquiry dialect subset meets the updating condition of the dialect sentence pattern.

In one embodiment, a computer storage medium is provided, having a computer program stored thereon, the computer program, when executed by a processor, implementing the steps of: acquiring initial inquiry data corresponding to the user identification; querying a matched target interrogatory dialect subset from a pre-configured interrogatory dialect set according to the initial interrogation data; determining a target inquiry dialect according to the dialect path corresponding to the target inquiry dialect subset, and sending the target inquiry dialect to a corresponding terminal; receiving target inquiry data fed back by the terminal aiming at the target inquiry dialogs, and analyzing the target inquiry data to obtain a corresponding inquiry analysis result; when the target inquiry data is judged to be the professional term inquiry data based on the inquiry analysis result, the spoken answer data matched with the professional terms in the target inquiry data is inquired from a pre-configured professional term inquiry set, the spoken answer data is fed back to the terminal, and the step of returning to the target inquiry data fed back by the receiving terminal aiming at the target inquiry tactics is continuously executed; and when the target inquiry data is judged to be inquiry answer data based on the inquiry analysis result and the inquiry answer data meets the answer condition corresponding to the target inquiry dialogs, returning to the step of determining the target inquiry dialogs according to the dialogs path corresponding to the target inquiry dialogs subset and continuing to execute until each inquiry dialogs in the target inquiry dialogs subset is traversed.

In one embodiment, the computer program when executed by the processor further performs the steps of: querying a matched initial inquiry dialect subset from a pre-configured inquiry dialect set according to the initial inquiry data; acquiring a user portrait corresponding to a user identifier; when the user portrait meets the updating condition of the professional terms of the dialogies, acquiring spoken expressions matched with the professional terms in the initial inquiry and diagnosis dialogies subset; and updating the inquiry dialogs including the corresponding professional terms in the initial inquiry dialogs subset according to the spoken expressions to obtain a target inquiry dialogs subset.

In one embodiment, the computer program when executed by the processor further performs the steps of: detecting a newly added high-frequency professional term in a preconfigured professional term spoken mapping set; inquiring the spoken expressions matched with each detected high-frequency professional term from the professional term spoken mapping set; updating the interrogatories in the subset of interrogatories in the set of interrogatories that include the corresponding high frequency professional term according to each spoken utterance queried.

In one embodiment, the computer program when executed by the processor further performs the steps of: when the target inquiry data is judged to be the professional term inquiry data based on the inquiry analysis result, the target inquiry data is used as the inquiry data and is stored in a pre-configured professional term inquiry data set; carrying out statistical analysis on the consultation and inquiry data in the professional term consultation data set regularly according to a preset period to obtain consultation frequency corresponding to each professional term in the professional term consultation data set; and determining the professional terms with the consultation frequency greater than or equal to the preset consultation frequency as the high-frequency professional terms detected from the pre-configured professional term spoken mapping set.

In one embodiment, the computer program when executed by the processor further performs the steps of: and when the preset waiting time is reached since the spoken answer data are fed back to the terminal, the target inquiry data fed back by the terminal aiming at the target inquiry dialogs are not received, the target inquiry dialogs are sent to the terminal again, and the step of returning to the target inquiry data fed back by the receiving terminal aiming at the target inquiry dialogs is continuously executed.

In one embodiment, the computer program when executed by the processor further performs the steps of: judging whether each inquiry tactical subset in the inquiry tactical set meets the updating condition of the tactical sentence pattern according to the tactical path; updating the sentence pattern of the corresponding inquiry sentence in each inquiry sentence pattern subset according with the updating condition of the sentence pattern according to the updating mode of the pre-configured sentence pattern.

In one embodiment, the computer program when executed by the processor further performs the steps of: sequentially calculating the text similarity between two adjacent interrogatories in each interrogatories talkaries subset in the interrogatories talkaries set according to the talkaries path to obtain a text similarity sequence corresponding to each interrogatories subset; counting the number of continuous text similarities which are greater than or equal to a similarity threshold value in each text similarity sequence; and when the quantity is greater than or equal to the quantity threshold value, judging that the corresponding inquiry dialect subset meets the updating condition of the dialect sentence pattern.

It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware related to instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in the embodiments provided herein may include non-volatile and/or volatile memory, among others. Non-volatile memory can include read-only memory (ROM), Programmable ROM (PROM), Electrically Programmable ROM (EPROM), Electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDRSDRAM), Enhanced SDRAM (ESDRAM), Synchronous Link DRAM (SLDRAM), Rambus Direct RAM (RDRAM), direct bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

The technical features of the above embodiments can be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the above embodiments are not described, but should be considered as the scope of the present specification as long as there is no contradiction between the combinations of the technical features.

The above examples only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present patent shall be subject to the appended claims.

Claims

1. A method of online interrogation data processing, the method comprising:

acquiring initial inquiry data corresponding to the user identification;

2. The method of claim 1, wherein querying a matching target questionnaire subset from a pre-configured questionnaire set according to the initial questionnaire data comprises:

acquiring a user portrait corresponding to the user identifier;

3. The method of claim 1, further comprising:

4. The method of claim 3, further comprising:

5. The method of claim 1, wherein after feeding back the spoken answer data to the terminal, the method further comprises:

6. The method according to any one of claims 1 to 5, further comprising:

7. The method of claim 6, wherein said determining whether each of the subset of interrogatories in the set of interrogatories according to the phony path meets a phony sentence pattern update condition comprises:

8. An online interrogation data processing apparatus, the apparatus comprising:

9. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor, when executing the computer program, implements the steps of the method of any of claims 1 to 7.

10. A computer storage medium on which a computer program is stored, characterized in that the computer program, when being executed by a processor, carries out the steps of the method of any one of claims 1 to 7.