CN115408511B

CN115408511B - Interactive node skipping method and component and conversation development system

Info

Publication number: CN115408511B
Application number: CN202211359057.1A
Authority: CN
Inventors: 李进峰; 高爱玲; 杨雯绚; 赖晓荣
Original assignee: Shenzhen Renma Interactive Technology Co Ltd
Current assignee: Shenzhen Renma Interactive Technology Co Ltd
Priority date: 2022-11-02
Filing date: 2022-11-02
Publication date: 2023-01-20
Anticipated expiration: 2042-11-02
Also published as: CN115408511A

Abstract

The application relates to an interactive node skipping method and component and a conversation development system. The method comprises the following steps: when jumping to a first plot interaction node based on first input information of a user, determining a target branch path matched with the first input information; judging whether the target branch path corresponds to an internal jump identifier or not; if the target branch path corresponds to the internal jump identifier, triggering internal jump to a corresponding second scenario interaction node based on the target branch path and hiding a node scenario of the second scenario interaction node; responding to the input information at the second scenario interaction node. According to the scheme, the branch path matched with the input information is determined, the scenario interaction node is jumped by combining the judgment result of the internal jump identification, the quick response to the input information and the quick jump of the scenario interaction node can be realized, the flexibility is good, and the intelligent interaction can be realized.

Description

Interactive node skipping method and component and conversation development system

Technical Field

The present application relates to the field of artificial intelligence technologies, and in particular, to an interactive node skipping method, an interactive node skipping apparatus, a computer device, a storage medium, and a session development system.

Background

With the development of artificial intelligence, the interactive robot can realize the jumping of nodes so as to gradually interact with users. In the node skipping process, the interactive robot can acquire input information of a user and perform node skipping based on the input information.

In the conventional technology, each node jump of the interactive robot needs to be triggered by user input information, namely, one node jump is performed only after one user input. This approach is often not flexible enough, resulting in less intelligent interaction.

It is to be noted that the information disclosed in the above background section is only for enhancement of understanding of the background of the present invention and therefore may include information that does not constitute prior art known to a person of ordinary skill in the art.

Disclosure of Invention

In view of the foregoing, it is necessary to provide an interactive node jump method, an apparatus, a computer device and a storage medium, and a dialog development system, which can implement intelligent interaction.

An interactive node jump method comprises the following steps:

when jumping to a first plot interaction node based on first input information of a user, determining a target branch path matched with the first input information;

judging whether the target branch path corresponds to an internal jump identifier or not;

if the target branch path corresponds to the internal jump identifier, triggering internal jump so as to jump to a corresponding second scenario interaction node based on the target branch path and hide the node scenario of the second scenario interaction node;

responding to the input information at the second plot interactive node.

In an optional embodiment, after the determining whether the target branch path corresponds to the internal jumping identifier, the method further includes:

if the target branch path does not have an internal jumping identifier correspondingly, jumping to a corresponding third scenario interaction node based on the target branch path;

acquiring second input information based on the third plot interaction node;

and if the second input information accords with the guiding condition of the third plot interaction node, continuing to carry out node skipping.

In an optional embodiment, said responding to the input information at the second scenario interaction node comprises:

determining a response scenario corresponding to the first input information;

and converting the response scenario into voice and outputting the voice to respond to the first input information.

determining a response answer corresponding to the first input information;

and converting the response answer into voice and outputting the voice to respond to the first input information.

In an optional embodiment, the determining a target branch path matching the first input information includes:

if the plot merging identifier is not set in the first plot interaction node, acquiring a plurality of branch paths preset by the first plot interaction node;

determining a branch path matching the first input information from the plurality of branch paths as the target branch path.

In an optional embodiment, the determining the target branch path matching the first input information includes:

if the first plot interaction node is provided with a plot merging identifier, eliminating a plurality of branch paths preset by the first plot interaction node;

determining a merged scenario corresponding to the node scenario of the first scenario interaction node, and taking the merged scenario as a unique outlet of the first scenario interaction node;

and directly taking the branch path corresponding to the combined scenario as the target branch path.

In an optional embodiment, before the responding the input information by the second scenario interaction node, the method further includes:

acquiring a historical variable label corresponding to historical input information of a user; when the first plot interactive node corresponds to input information, the historical input information comprises the input information corresponding to the first plot interactive node;

acquiring a predetermined variable label set;

comparing the historical variable labels with variable labels in the variable label set, and obtaining difference variable labels according to comparison results;

generating a node plot of a next plot interaction node of the second plot interaction node based on the difference variable label;

correspondingly, after the second scenario interaction node responds to the input information, the method further comprises the following steps:

and continuously jumping to the next scenario interaction node, and converting the node scenario of the next scenario interaction node into voice and outputting the voice.

An interactive node hopping apparatus, comprising:

the system comprises a branch path determining module, a target branch path determining module and a target branch path determining module, wherein the branch path determining module is used for determining a target branch path matched with first input information when jumping to a first plot interaction node based on the first input information of a user;

the identification judgment module is used for judging whether the target branch path corresponds to an internal jump identification or not;

the node skipping module is used for skipping to a corresponding second scenario interaction node based on the target branch path and hiding the node scenario of the second scenario interaction node if the target branch path corresponds to the internal skipping identification;

and the input response module is used for responding the input information at the second plot interactive node.

A dialog development system comprising: an internal jumping unit, a node jumping unit and a response unit;

when jumping to a first scenario interaction node based on first input information of a user, the node jumping unit determines a target branch path matched with the first input information;

the internal jumping unit judges whether the target branch path corresponds to an internal jumping identifier;

if the target branch path corresponds to the internal jumping identification, the internal jumping unit triggers internal jumping to jump to a corresponding second scenario interaction node based on the target branch path and hides the node scenario of the second scenario interaction node;

the response unit responds to the input information at the second scenario interactive node.

A computer device comprising a memory and a processor, the memory storing a computer program, the processor implementing the following steps when executing the computer program:

if the target branch path corresponds to the internal jumping identification, triggering internal jumping to jump to a corresponding second scenario interaction node based on the target branch path and hiding a node scenario of the second scenario interaction node;

responding to the input information at the second plot interactive node.

A computer-readable storage medium, on which a computer program is stored which, when executed by a processor, carries out the steps of:

responding to the input information at the second plot interactive node.

According to the interactive node skipping method, when skipping to a first plot interactive node based on first input information of a user, a target branch path matched with the first input information is determined; judging whether the target branch path corresponds to an internal jump identifier or not; if the target branch path corresponds to the internal jump identifier, triggering internal jump to a corresponding second scenario interaction node based on the target branch path and hiding a node scenario of the second scenario interaction node; responding to the input information at the second scenario interaction node. The method and the system have the advantages that the branch path matched with the input information is determined, the scenario interaction node is jumped according to the judgment result of the internal jump identification, the input information can be quickly responded, the scenario interaction node can be quickly jumped, the flexibility is good, and intelligent interaction can be realized. Correspondingly, the interactive node jumping device, the equipment and the storage medium provided by the application, and the conversation development system also have the technical effects.

Drawings

FIG. 1 is a diagram of an application environment for an interactive node hopping method in one embodiment;

FIG. 2 is a flowchart illustrating an interactive node hopping method according to an embodiment;

FIG. 3 is a block diagram of an exemplary embodiment of an interactive node jump apparatus;

FIG. 4 is a block diagram of the architecture of the dialog development system in one 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.

An automatic conversation development system for developing a chat robot capable of automatically conversing with a user, wherein the automatic conversation development system can be used for setting the structure of the chat robot and configuring various functions for the chat robot in the development process of the chat robot.

The chat robot consists of a plurality of plot interactive nodes, a plot interactive node frame forming the chat robot is built through an automatic conversation development system, and the connection relation of the plot interactive nodes is determined; and setting tasks to be executed by the chat robot at the plot interactive node so as to determine and realize the interaction between the chat robot and the user at the plot interactive node. The tasks to be executed by the chat robot at the scenario interaction node include setting a node scenario, guiding conditions, other response tasks aiming at the input information of the user in the scenario interaction node, and the like.

The node scenario is information sent by the chat robot to the user at a scenario interaction node, namely words output at the corresponding scenario interaction node. The node scenario may be used to guide the user in a conversation. The node scenario may be one or more utterances, and further, the node scenario may be a fiction scenario description, a dialogue text, an interactive guidance text, and the like in an application scenario such as an interactive fiction.

Specifically, if the first plot interactive node is connected with the second plot interactive node, the second plot interactive node is connected with the third plot interactive node. After entering a second plot interaction node from the first plot interaction node, the chat robot outputs a node plot arranged at the second plot interaction node; after entering the third plot interactive node from the second plot interactive node, the chat robot outputs the node plot arranged at the third plot interactive node.

The plot interaction node can set a plurality of guiding conditions, and the guiding conditions are used for executing the corresponding tasks associated with the guiding conditions when the guiding conditions are met.

The guide conditions include guide conditions that match user input information, functional conditions that do not need to match user input information, and other types of guide conditions. The guiding conditions matched with the user input information comprise guiding conditions matched with the user input information and having a specific matching range, and guiding conditions matched with the user input information and matching any range; the functional conditions which do not need to match the input information of the user comprise whether a plot merging identifier exists, whether an internal jumping identifier exists and other functional conditions. Wherein various types of guidance conditions may be applied in combination.

Optionally, a guiding condition matching the input information of the user and having a specific matching range may be set at the scenario interaction node, and when the guiding condition is satisfied, a task corresponding to the guiding condition is executed, including jumping from the current scenario interaction node to a next scenario interaction node connected with the current scenario interaction node and corresponding to the guiding condition according to the guidance of the guiding condition set at the current scenario interaction node.

Specifically, if the first scenario interaction node is connected with the second scenario interaction node, and the first scenario interaction node is further connected with the third scenario interaction node. The first plot interactive node is provided with a first guiding condition and a second guiding condition, the first guiding condition is set to guide jumping to the second plot interactive node, and the second guiding condition is set to guide jumping to the third plot interactive node. If the first user input information acquired at the first plot interactive node meets the first guide condition, the process of the chat robot jumps from the first plot interactive node to the second plot interactive node according to the guide of the first guide condition; and if the second user input information acquired at the first plot interactive node meets the second guiding condition, skipping the process of the chat robot from the first plot interactive node to a third plot interactive node according to the guidance of the second guiding condition.

Optionally, the scenario interaction node may set a guidance condition that matches the user input information and matches any range, and the setting of the guidance condition that matches the user input information and matches any range may be implemented by not setting a guidance condition that matches the user input information and has a specific matching range at the scenario interaction node. Therefore, the guide condition not specified thereafter defaults to a guide condition matching the user input information and having a specific matching range.

Specifically, the scenario interaction node is not provided with a guiding condition matching with the user input information and having a specific matching range, if the current scenario interaction node is not provided with a guiding condition matching with the user input information and having a specific matching range, and a unique next scenario interaction node connected with the current scenario interaction node exists, at this time, no matter the user inputs any information, the current scenario interaction node can enter the next scenario interaction node connected with the current scenario interaction node according to the guidance matching with the user input information and matching with the guiding condition of any range. That is, the user input information can be matched arbitrarily by matching the user input information and matching the guide condition of any range, and the satisfied condition is that the user input information exists.

Generally, a chat robot and a user are in a question-answer communication mode, the chat robot outputs a node plot of a current plot interaction node after entering the current plot interaction node, then is in a state of waiting for user input, after obtaining user input information, enters a next plot interaction node connected with the current plot interaction node according to guidance of a guidance condition which is arranged at the current plot interaction node and is matched with the input information, and outputs a node plot of the next plot interaction node as a response to the user input information.

However, in some cases, in order to make the chat robot more flexible to communicate with the user and increase the intelligence of the chat robot, a question-and-answer communication mode needs to be broken.

Optionally, a scenario merging identifier may be added to the current scenario interaction node. The plot merging identifier is used for enabling the node plot of the current plot interaction node and the node plot of the next plot interaction node to be continuously output.

Entering a current plot interactive node connected with a previous plot interactive node from the previous plot interactive node, and judging whether a plot merging identifier exists in the current plot interactive node or not after the chat robot outputs a node plot of the current plot interactive node; if the plot merging identifier exists, entering a next plot interaction node connected with the current plot interaction node from the current plot interaction node, outputting a node plot arranged at the next plot interaction node, and waiting for information input by a user at the next plot interaction node.

Specifically, if the scenario merging identifier exists, the scenario interaction node is skipped over to wait for the state of the user to input information, that is, the user is not required to input information at the current scenario interaction node at present, the next scenario interaction node connected with the current scenario interaction node can be directly entered through the current scenario interaction node, the node scenario set at the next scenario interaction node is continuously output, and the next scenario interaction node is in the state of waiting for the user to input information. Therefore, if the plot merging identifier exists, the chat robot can continuously output the node plot arranged at the current plot interaction node and the node plot arranged at the next plot interaction node connected with the current plot interaction node. Therefore, the plot merge sign may also be considered as one of the guiding conditions, which is a guiding condition that does not require the presence of user input information or does not require matching of user input information.

In addition, if a plurality of different guiding conditions are preset in the current plot interactive node, the current plot interactive node can jump to a plurality of different other next plot interactive nodes connected with the current plot interactive node according to the guiding of the plurality of different guiding conditions, then after the plot merging identifier is added to the current plot interactive node, all the guiding conditions preset by the current plot interactive node are removed, and the next plot interactive node connected with the current plot interactive node is uniquely specified, namely only the only plot interactive node is left to be connected with the current plot interactive node. It can be understood that adding the scenario merging identifier to the current scenario interaction node may cause the guidance condition that is preset by the current scenario interaction node and matches the user input information and has a specific matching range to be eliminated, so that the guidance condition that matches the user input information and matches any range still exists.

It can be understood that if the scenario merging identifier exists, the user does not need to input information at the current scenario interaction node, and the next scenario interaction node connected with the current scenario interaction node is directly accessed through the current scenario interaction node. In this embodiment, it may be considered that the scenario merging flag exists, the current scenario interaction node is skipped from the state of waiting for the user input information, but the guidance condition matching the user input information and matching any range is considered to be satisfied, so that the next scenario interaction node is entered according to the guidance matching the user input information and matching the guidance condition of any range. In other embodiments, the step of waiting for the user input information at the current scenario interaction node and the step of skipping the guidance condition matching input information matching the user input information may also be skipped according to the scenario merging identifier. In other embodiments, the next scenario interaction node linked with the current scenario interaction node may be entered according to the guidance of the scenario merging identifier, and there is no step of matching the guidance condition according to the user input information and then jumping to the next scenario interaction node linked with the current scenario interaction node according to the guidance of the guidance condition.

Optionally, an internal skip identifier may also be added to the current scenario interaction node. The internal jumping identification can be associated with other types of guiding conditions such as guiding conditions, plot merging identification and the like.

Preferably, the internal jump identification corresponds to the guiding condition association. If the internal skip identifier corresponds to the associated guiding condition, after the guiding condition is met, the chat robot judges whether the met guiding condition has the associated internal skip identifier. If the satisfied guide condition exists in the current plot interaction node, the corresponding internal jump identification is associated; entering a next scenario interaction node which is connected with the current scenario interaction node and corresponds to the guiding condition according to the guidance of the guiding condition; and executing a task corresponding to the internal jump identifier acting on the next scenario interaction node according to the internal jump identifier.

Specifically, if the current scenario interaction node has a first guide condition and a second guide condition. If the second directing condition is met, the chat robot judges whether the second directing condition has an internal skip mark corresponding to the association.

And if the second guidance condition does not have the internal jump identification corresponding to the association, entering a next scenario interaction node corresponding to the second guidance condition according to the guidance of the second guidance condition, outputting the node scenario of the next scenario interaction node and waiting for the input of the user.

And if the second guidance condition has the internal jump identification corresponding to the association, entering a next scenario interaction node corresponding to the second guidance condition according to the guidance of the second guidance condition, and executing a task corresponding to the internal jump identification acting on the next scenario interaction node according to the internal jump identification.

After entering the next scenario interaction node, hiding the node scenario of the next scenario interaction node, namely not outputting the node scenario of the next scenario interaction node to the outside (user), acquiring the latest user input information, continuously taking the latest user input information as the user input information of the next scenario interaction node, namely, at the next scenario interaction node, the state of waiting for the user input information does not exist, the user does not need to have new input information at the scenario interaction node, and the next scenario interaction node takes the latest user input information as the input information of the user at the scenario interaction node for processing, namely, the latest user input information is processed at different scenario interaction nodes once respectively and is processed twice in total. And if the next scenario interaction node has the third guiding condition which can be matched with the latest user input information, executing the task corresponding to the third guiding condition according to the guidance of the third guiding condition matched with the user input information by the next scenario interaction node.

Optionally, if the current scenario interaction node has user input information, the latest user input information is input information input by the user of the current scenario interaction node, and the user input information of the current scenario interaction node is continuously used as the user input information of the next scenario interaction node and is processed once again; if the current plot interaction node does not have the user input information, the latest user input information before the current plot interaction node is obtained, such as: and the previous plot interactive node connected with the current plot interactive node has user input information, the latest user input information is the user input information of the previous plot interactive node, the user input information of the previous plot interactive node is continuously used as the user input information of the next plot interactive node, and the processing is carried out once again. It can be understood that a first scenario interaction node processing user input information for the first time has a first guiding condition matching with the user input information, and a second scenario interaction node processing user input information for the second time has a second guiding condition matching with the user input information, where the first guiding condition and the second guiding condition may be the same or different, and when the first guiding condition and the second guiding condition are different, an inclusion relationship may be present or may not be present, such as: for the user input information of 'i want to eat watermelon', the first guiding condition can be that a cardinal relation exists, and the second guiding condition can be that a guest-moving relation exists. It can be understood that if the second scenario interaction node acted by the internal skip identifier cannot process the user input information input by the first scenario interaction node, that is, when the second scenario interaction node does not have the second indexing condition matching the user input information, a processing error will occur, and the second indexing condition in the second scenario interaction node needs to be modified to match the second indexing condition with the user input information. Therefore, it is preferable that the first guidance condition and the second guidance condition are the same, or that the second guidance condition range includes the first guidance condition range.

It is understood that the internal skip identifier can also be considered as a kind of guiding condition, which is a guiding condition without user input information acting on the next scenario interaction node.

In summary, the chat robot judges and identifies the sequence of the guide conditions of the types, namely, judges whether a plot merging identifier exists or not; and judging whether the user input can be matched with the guiding condition matched with the user input information, and judging whether an internal skip identifier corresponding to the guiding condition is existed after the guiding condition matched with the user input information is met.

Specifically, after entering a current plot interaction node, judging whether a plot combination identifier exists;

and if the plot merging identifier exists, judging whether an internal jumping identifier exists. In one embodiment, if a scenario merge flag exists, it is determined whether an internal condition flag exists that matches the user input information and matches the guidance condition association in any range.

If the plot merging identifier does not exist, judging whether the user input information meets the guiding condition matched with the user input information;

and if the user input information meets the guide condition matched with the user input information, judging whether an internal skip identifier corresponding to the met guide condition exists.

Alternatively, with respect to the combination of each type of guidance condition, there are cases where:

1. a plot merging identifier and an internal jumping identifier exist;

2. the scenario merging identification exists and the internal jumping identification does not exist;

3. the scenario merging identification does not exist, but the internal jumping identification exists;

4. there is no plot merging flag and no internal skip flag.

These situations are explained in detail below:

the first situation is as follows:

the chat robot acquires user input information at a previous plot interactive node, and if the user input information meets a first guide condition for entering the current plot interactive node from the previous plot interactive node, the chat robot enters the current plot interactive node from the previous plot interactive node;

the node plot of the current plot interactive node is output at the current plot interactive node, the chat robot judges whether the plot combination mark and the internal jumping mark exist in the current plot interactive node or not,

if the plot merging identifier and the internal jumping identifier exist in the current plot interaction node;

according to the plot combination identification, under the condition that the input information of the user is not obtained in the current plot interaction node, entering a next plot interaction node connected with the current plot interaction node;

and according to the internal jump identification, not outputting the node plot of the next plot interactive node at the next plot interactive node, acquiring the latest user input information, taking the latest user input information as the user input information of the next plot interactive node, and if the user input information meets a second guiding condition for entering a fourth plot interactive node connected with the next plot interactive node from the next plot interactive node, entering the fourth plot interactive node from the next plot interactive node.

In one embodiment, the scenario interaction nodes may correspond to scenario interaction node identifiers, each scenario interaction node has a unique scenario interaction node identifier, and the scenario interaction node identifiers may be numbers generated according to a certain rule, and the like. Such as scenario interaction nodes 1-X, scenario interaction nodes 1-X, and so on.

Assuming that the scenario interaction node 1-1 is an initial scenario, the scenario interaction node 1-2 is connected with the scenario interaction node 1-1, the scenario interaction node 1-3 is connected with the scenario interaction node 1-2, the scenario interaction node 1-3 is a transfer judgment scenario interaction node, a plurality of different guide conditions exist in the transfer judgment scenario interaction node 1-3, in the embodiment, 3 different guide conditions exist in the transfer judgment scenario interaction node 1-3, a first guide condition, a second guide condition and a third guide condition exist in the transfer judgment scenario interaction node 1-3, the scenario interaction node 1-4-X is connected with the scenario interaction node 1-3, the scenario interaction node 1-4-1, the scenario interaction node 1-4-2 and the scenario interaction node 1-4-3 respectively correspond to the 3 different guide conditions in the transfer judgment scenario interaction node 1-3 one by one, the scenario interaction node 1-4-1 is guided according to the first guide conditions, the scenario interaction node 1-4-2 is guided according to the second guide conditions, and the scenario interaction node 1-4-3 is guided according to the third guide conditions.

In a specific embodiment, the specific case of the indexing condition in the scenario interaction node 1-2 may be: the scenario merging identification exists, the guiding conditions which are matched with the user input information and are matched in any range exist, and the internal jumping identification which is associated and corresponding to the guiding conditions which are matched with the user input information and are matched in any range exists.

The chat robot outputs a node plot of a plot interactive node 1-1 at the plot interactive node 1-1 and is in a state of waiting for the input of a user, acquires user input information, if the user input information meets the guiding condition of guiding the plot interactive node 1-2 from the plot interactive node 1-1, the plot interactive node 1-2 is entered from the plot interactive node 1-1, the node plot of the plot interactive node 1-2 is output at the plot interactive node 1-2, the chat robot judges whether a plot merging identifier exists or not, and skips the step of waiting for the input of the user at the plot interactive node 1-2 according to the plot merging identifier, and directly matches the input information of the user and the guiding condition in any range through the existence of matching user input information without the input of the user, obtaining an internal jumping identifier corresponding to the guiding condition which matches with the user input information and matches with any range, entering a scenario interaction node 1-3 connected with the scenario interaction node 1-2, hiding (not outputting) the node scenario of the scenario interaction node 1-3 in the scenario interaction node 1-3 according to the internal jumping identifier, obtaining the latest user input information, using the latest user input information as the user input information of the scenario interaction node 1-3, and obtaining the user input information input by the user at the scenario interaction node 1-1 because the user input information does not exist in the scenario interaction node 1-2, and using the input information input by the scenario interaction node 1-1 as the input information input by the user at the scenario interaction node 1-3, the input information input by the user in the plot interactive node 1-1 is processed once again in the plot interactive node 1-3, the plot interactive node 1-3 is a transfer judgment plot interactive node, and the plot interactive node 1-3 enters the plot interactive node 1-4-X according to the guidance of the guidance condition matched with the input information of the user in the transfer judgment plot interactive node 1-3.

Therefore, after entering the plot interactive node 1-2 from the plot interactive node 1-1 and outputting the node plot of the plot interactive node 1-2, the system automatically jumps to the transfer judgment plot node 1-3 and hides the node plot of the transfer judgment plot node 1-3, and then can jump to the plot interactive node 1-4-X according to the matching result of the transfer judgment plot node 1-3 on the user input information input by the user at the plot interactive node 1-1 based on the user input information input by the user at the plot interactive node 1-1. Under the condition, a user inputs information only at the plot interactive node 1-1, and the user does not need to input the user input information at the plot interactive node 1-2 and the plot interactive node 1-3, so that the plot can jump from the plot interactive node 1-1 to the plot interactive node 1-4-X, and the chat robot continuously inputs the node plot of the plot interactive node 1-2 and the node plot of the plot interactive node 1-4-X and hides the node plot of the plot interactive node 1-3 which is not input. Therefore, the meaningless node scenarios of the relay judgment scenario nodes 1-3 can be hidden and not output.

Case two:

the plot interaction node 1-2 has plot merging identification and no internal jump identification.

The plot interactive node 1-1 is an initial plot, the plot interactive node 1-2 is connected with the plot interactive node 1-1, and the plot interactive node 1-3 is connected with the plot interactive node 1-2.

If the scenario interaction node 1-2 corresponds to the scenario merging identifier but has no internal jump identifier, the chat robot enters the scenario interaction node 1-2 from the scenario interaction node 1-1, then outputs the node scenario of the scenario interaction node 1-2, enters the scenario interaction node 1-3 according to the scenario merging identifier and outputs the node scenario of the scenario interaction node 1-3, so that the node scenario of the scenario interaction node 1-2 and the node scenario of the scenario interaction node 1-3 are continuously output and are in the scenario interaction node 1-3 to wait for the input of the user.

Case three:

the scenario interaction node 1-2 has no scenario merging identification, but has an internal jumping identification.

In one embodiment, the scenario interaction node 1-1 is an initial scenario, the scenario interaction node 1-2 is connected with the scenario interaction node 1-1, and the scenario interaction node 1-3 is connected with the scenario interaction node 1-2. The scenario interaction node 1-2 has a plurality of different guiding conditions, such as a first guiding condition and a second guiding condition, and the scenario interaction node 1-3-1 is guided by the first guiding condition and the scenario interaction node 1-3-2 is guided by the second guiding condition. The first guiding condition has an associated corresponding internal jump identification. The scenario interaction node 1-4-1 is connected with the scenario interaction node 1-3-1. The second indexing condition has no internal jump identification corresponding to the association. The plot interactive node 1-4-2 is connected with the plot interactive node 1-3-2.

If the scenario interaction node 1-2 has no scenario merging identifier and has an internal jumping identifier, the chat robot is in a state waiting for the user to input information after the scenario interaction node 1-2 outputs the corresponding node scenario, analyzes the user input information (i.e. the user replies to the node scenario of the scenario interaction node 1-2) to determine a guidance condition matched with the user input information, if the first guidance condition is successfully matched with the user input information, acquires the internal jumping identifier corresponding to the first guidance condition after the first guidance condition is satisfied, enters the scenario interaction node 1-3-1 according to the guidance of the first guidance condition, hides the node scenario of the scenario interaction node 1-3-1 according to the internal jumping identifier corresponding to the first guidance condition, and takes the user input information of the user at the scenario interaction node 1-2 as the user input information of the scenario interaction node 1-3-1, and if the third guidance condition exists in the scenario interaction node 1-3-1-4, the third guidance information can be matched with the user input information of the scenario interaction node 1-3-1 according to the guidance condition. If the third guide condition for guiding the scenario interaction node 1-4-1 to enter the scenario interaction node 1-3-1 cannot be matched with the information input by the user, an error occurs at the moment. If the second guiding condition is successfully matched with the user input information, after the second guiding condition is met, the internal jump identification corresponding to the second guiding condition is not acquired, the scenario enters the scenario interaction node 1-3-2 according to the guidance of the second guiding condition, the node scenario of the scenario interaction node 1-3-2 is output, and the scenario is in a state waiting for user input in the scenario interaction node 1-3-2.

Case four:

the plot interactive node 1-2 has no plot merging identification and no internal jumping identification.

If the scenario interaction node 1-2 does not have a scenario merging identifier or an internal jumping identifier correspondingly, the chat robot is in a state waiting for user input in the scenario interaction node 1-2 after the scenario interaction node 1-2 outputs the corresponding node scenario, acquires user input information, jumps to the scenario interaction node 1-3 and outputs the node scenario of the scenario interaction node 1-3 according to guidance of a guidance condition matched with the user input information in the scenario interaction node 1-2, and is in a state waiting for user input in the scenario interaction node 1-3. Under the situation, the user needs to input information at the scenario interaction node 1-2 and the scenario interaction node 1-3 to gradually realize the skipping of the scenario interaction node.

Therefore, by adding the internal jump identifier and/or the plot merging identifier, the input times of the user can be effectively reduced, the condition that the output is only performed when the input is performed is avoided, and the intelligence and the flexibility of interaction can be effectively improved. The internal jumping identification and/or the plot merging identification can be configured for the plot interaction node of the chat robot based on the process of developing or setting the chat robot by the automatic conversation development system.

Further, the first scenario interaction node (1-2) may be understood as a branch junction, which has many branch paths such as branch 1, branch 2, and branch 3 (different guiding conditions correspond to different branches), and these branches lead to different scenario interaction nodes, for example, the guiding condition of branch 1 is that "i want to eat" leads to 2-1, the guiding condition of branch 2 is that "i do not want to eat" leads to 2-2, and branch 3 "i do not know that i do not want to eat" leads to 2-3. Assume that there is an internal jump identification only at branch 1, and that branches 2 and 3 do not. Then, the internal jump function can be triggered only if the sentence of 'i want to eat', the matching is performed once in 1-2, the matching is performed once again in 2-1 according to the successfully matched branch 1, and meanwhile, the sentence of 'i want to eat' is performed once again in 2-1, and if the matching is successful, the linkage scenario of 2-1 (namely, the node scenario corresponding to the next scenario interaction node) is performed. If the branch 2 and the branch 3 are not provided with the internal jump function, the internal jump is not triggered if the two words of 'i do not want to eat' or 'i do not know' are input in 1-2, but if the 1-2 matches the guide conditions of the branch 2 and the branch 3, the matching guide conditions can be triggered to jump to the linkage scenario 2-2 or 2-3 if another word is input.

The response may include a node scenario of the scenario interaction node linked with the current scenario interaction node, reply information preset in the current scenario interaction node, and other response actions or response information in the scenario interaction node.

Optionally, the node scenario may be used as response information for the user to input information in the scenario interaction node of the previous round or the scenario interaction nodes of the previous rounds. Such as: the first plot interactive node is connected with the second plot interactive node, and the response aiming at the input information of the user in the first plot interactive node is the node plot of the second plot interactive node, namely, the user input information in the first plot interactive node is replied through the node plot of the second plot interactive node. At this time, the response task flow of the chat robot is as follows: the chat robot outputs a first node plot at a first plot interaction node, a user inputs certain input information at the first plot interaction node, the chat robot carries out semantic understanding on the input information, obtains a semantic understanding result, matches the semantic understanding result with a plurality of preset guiding conditions of the first plot interaction node, enters a second plot interaction node according to the guidance of the successfully matched guiding conditions, and the chat robot outputs a second node plot at the second plot interaction node to serve as reply information of the input information.

Optionally, the responding to the input information of the user in the scenario interaction node may further include presetting reply information in the scenario interaction node. Such as: and if the input information of the user in the first plot interactive node meets a certain guiding condition and the preset responding information have an incidence relation, outputting the preset responding information according to the guidance of the guiding condition, and then jumping to the next plot interactive node connected with the first plot interactive node according to the guidance of the guiding condition.

The interactive node skipping method provided by the application can be applied to the application environment shown in fig. 1. The application environment comprises a terminal 101 and a server, and a plurality of chat robots 102 developed based on an automatic conversation development system are deployed in the server. The terminal 101 communicates with the server through a network and performs information interaction with the chat robot 102. The terminal may be a terminal that the user is using. A user can input information through a terminal; the chat robot receives input information sent by a terminal, and judges whether an internal jumping identifier and a plot merging identifier exist in a first plot interaction node or not when the input information based on a user jumps to the first plot interaction node; and realizing the response to the input information and the interactive node jump according to the judgment result. Among them, the terminal 101 may be, but is not limited to, various personal computers, notebook computers, smart phones, tablet computers, and portable wearable devices. The server may be implemented as a stand-alone server or as a server cluster consisting of a plurality of servers. The chat robot can realize interaction with users in application scenes of interactive novels, ticket purchasing, sales promotion services and the like, and the scenarios are connected in series through one scenario interaction node to promote the scenario development.

In one embodiment, as shown in fig. 2, there is provided an interactive node hopping method, including the steps of:

s201, when jumping to a first scenario interaction node based on first input information of a user, determining a target branch path matched with the first input information.

The scenario interaction node is a node for interacting with a user, that is, a node capable of outputting a scenario, receiving a user input, and responding to the user input, and in some cases, the scenario interaction node may be simply referred to as an interaction node or a node, and may also be referred to as a scenario point. In the scenario interaction node, the chat robot may acquire input information of the user. And the first plot interactive node is the plot interactive node which enters currently. The input information may be input information obtained after a user performs interaction such as voice input, text input, interface operation and the like through the terminal device.

Optionally, after the chat robot acquires the first input information, the chat robot analyzes the first input information and judges the matching of the guide conditions, and jumps to the first scenario interaction node according to a judgment result. A target branching path corresponding to the first input information is determined in the first scenario interaction node, which may be understood as a response to the first input information.

S202, judging whether the target branch path corresponds to an internal jump identifier.

S203, if the target branch path corresponds to the internal jump identifier, triggering internal jump to a corresponding second scenario interaction node based on the target branch path and hiding the node scenario of the second scenario interaction node.

The chat robot judges whether the target branch path has an internal jump identifier, and if the internal jump identifier exists, the chat robot enters a related flow of internal jump (which can comprise an internal jump flow and an internal jump + scenario merging flow); if the internal jump identification does not exist, the related flow (which can comprise a plot merging flow and a conventional jump flow) of the non-internal jump is entered.

For the condition that the internal jumping identification exists, the chat robot enters the related flow of the internal jumping, at the moment, the voice output is carried out on the target node plot of the first plot interaction node, the second plot interaction node can be determined as a transfer judgment plot node, and if the plot merging identification exists in the first plot interaction node, the flow of the internal jumping and the plot merging is entered: skipping to a transfer judgment scenario node, hiding a node scenario corresponding to the transfer judgment scenario node, and responding to the first input information based on a judgment result of the transfer judgment scenario node; if the scenario merging identification does not exist in the first scenario interaction node, an internal skipping process is started: and receiving input information aiming at the first scenario interaction node, skipping to the transfer judgment scenario node, hiding the node scenario corresponding to the transfer judgment scenario node, and responding to the input information aiming at the first scenario interaction node based on a judgment result of the transfer judgment scenario node. Therefore, for the condition that the internal jumping identification exists, different responses are carried out based on the judgment result of the scenario merging identification, the required user input times are different, the multi-scene expansion can be realized, and meanwhile, the intelligent interaction of the chat robot is ensured.

And S204, responding to the input information at the second scenario interactive node.

And responding the input information when the first input information is matched with the response condition corresponding to the second plot interactive node.

Further, when the first plot interactive node corresponds to the plot merging flag, the second plot interactive node may respond to the first input information, and when the first plot interactive node corresponds to the plot merging flag, the third input information corresponding to the first plot interactive node may be acquired, and the second plot interactive node responds to the third input information. It should be noted that the implementation manners of responding to the first input information and the third input information may be referred to each other.

In the interactive node skipping method, when skipping to a first plot interactive node based on first input information of a user, a target branch path matched with the first input information is determined; judging whether the target branch path corresponds to an internal jump identifier or not; if the target branch path corresponds to the internal jump identifier, triggering internal jump to a corresponding second scenario interaction node based on the target branch path and hiding a node scenario of the second scenario interaction node; responding to the input information at the second scenario interaction node. The method and the system have the advantages that the branch path matched with the input information is determined, the plot interaction node is skipped by combining the judgment result of the internal skip identifier, the quick response to the input information and the quick skip of the plot interaction node can be realized, the flexibility is good, and the intelligent interaction can be realized.

In an optional embodiment, after the determining whether the target branch path corresponds to the internal jumping identifier, the method further includes: if the target branch path does not have an internal jumping identifier correspondingly, jumping to a corresponding third scenario interaction node based on the target branch path; acquiring second input information based on the third plot interaction node; and if the second input information accords with the guiding condition of the third plot interaction node, continuing to carry out node skipping.

When the target branch path has no internal jump mark correspondingly, the target branch path does not start the internal jump function, common node jump can be carried out at the moment, namely jump to a third scenario interaction node corresponding to the target branch path, second input information is obtained based on the third scenario interaction node, guiding conditions are judged for the second input information, and then node jump is realized.

In the general skip process of this embodiment, input information for a first scenario interaction node is received, a skip is made to a next scenario interaction node corresponding to the input information, and the input information for the next scenario interaction node is responded based on the next scenario interaction node.

In an optional embodiment, when it is determined that the target branch path does not have the internal jumping identifier, it may be determined whether the first scenario interaction node corresponds to the scenario combination identifier, and the jumping of the node is achieved in a targeted manner according to a determination result of the scenario combination identifier.

And when the plot merging identifier exists, entering a plot merging and skipping process:

in the scenario merging process, a merged scenario corresponding to the scenario of the target node is obtained, the second scenario interaction node is skipped to and the merged scenario is output, then a response answer aiming at the input information of the merged scenario is obtained, and the new scenario interaction node is skipped to and the response answer is output.

The implementation of the embodiments of the present application is further illustrated below:

1) The chat robot says 1-1.

2) And the user replies to obtain the input information.

3) And (4) judging the matching of the guide conditions on the input information, and if the matching is successful, carrying out 1-1 to (1-2 and 1-3).

4) The chat robots say words (1-2 and 1-3).

5) And the user replies to obtain the input information.

6) And judging the matching of the guiding conditions for the input information, and if the matching is successful, jumping from 1-3 to 1-4.

In the scenario merging process, merging output of the target node scenario and the merging scenario can be automatically carried out, two sections/sections of scenarios (such as 1-2 and 1-3) can be connected together, the chatting robot can speak at one time (the words of 1-2 and 1-3 are output to the user), and therefore the merging scenario can be flexibly set to continuously trigger the sections of scenarios, the flexibility of node skipping is effectively improved, and the intelligence of voice interaction is guaranteed.

And when the plot merging identifier does not exist, entering a conventional skipping process:

in a conventional jumping process, input information aiming at a target plot interactive node is received, jumping to a second plot interactive node corresponding to the input information is carried out, and responding to the input information aiming at the second plot interactive node based on the second plot interactive node.

1) The chat robot says 1-1.

2) And the user replies to obtain the input information.

3) And (4) judging the matching of the guiding conditions for the input information, and if the matching is successful, jumping from 1-1 to 1-2.

4) The chat robot says 1-2 words.

5) And the user replies to obtain the input information.

6) And judging the matching of the guiding conditions for the input information, and if the matching is successful, jumping from 1-2 to 1-3.

7) The chat robot speaks words 1-3.

8) And the user replies to obtain the input information.

9) And judging the matching of the guiding conditions for the input information, and if the matching is successful, jumping from 1-3 to 1-4.

In the conventional skipping process, skipping of interactive nodes and outputting of answers are performed step by step under the triggering of user input, and ordered voice interaction can be realized.

In an optional embodiment, said responding to the input information at the second scenario interaction node comprises: determining a response scenario corresponding to the first input information; and converting the response scenario into voice and outputting the voice to respond to the first input information.

Optionally, the response scenario corresponding to the first input information is a node scenario of the scenario interaction node linked with the first scenario interaction node, and the response scenario may guide the user to perform the next interaction.

In this embodiment, the response scenario corresponding to the first input information is determined, and the second scenario intersection node responds to the first input information, so that reliable answer to the input of the user can be realized, and the intelligence of voice interaction is ensured.

In an optional embodiment, said responding to the input information at the second scenario interaction node comprises: determining a response answer corresponding to the first input information; and converting the response answer into voice and outputting the voice to respond to the first input information.

The response answer is a reply to the first input message, and may be a reply message preset in the first scenario interaction node.

Optionally, after the response answer is obtained, the scenario interaction node is jumped to, and the response answer is output in the new scenario interaction node.

In this embodiment, a response answer corresponding to the first input information is determined, and the first input information is responded to the second scenario intersection, so that a reliable answer to the user input can be realized, and the intelligence of voice interaction is ensured.

In an optional embodiment, the determining a target branch path matching the first input information includes: if the plot merging identifier is not set in the first plot interaction node, acquiring a plurality of branch paths preset by the first plot interaction node; determining a branch path matching the first input information from the plurality of branch paths as the target branch path.

In this embodiment, under the condition that the scenario merging function is not started by the first scenario interaction node, a common branch path determination method is used to select from multiple branch paths to obtain a target branch path, and then the scenario node is skipped based on the target branch path, so that the voice interaction scenario can be effectively expanded.

Optionally, when the scenario merging function is not started by the first scenario interaction node and the internal skip identifier exists in the target branch path, the internal skip process is entered. In the internal skipping process, third input information aiming at the target node scenario is obtained, then a next scenario interaction node corresponding to the third input information is determined as a transfer judgment scenario node, skipping is conducted to the transfer judgment scenario node, the corresponding node scenario is hidden, a response answer is obtained, and then skipping is conducted to a new scenario interaction node, and the response answer is output.

1) The chat robot says 1-1.

2) And the user replies to obtain the input information.

3) And judging the matching of the guiding conditions for the input information, and if the matching is successful, jumping from 1-1 to 1-2.

4) The chat robot speaks 1-2 words at 1-2.

5) And judging the matching of the guiding conditions for the input information, and if the matching is successful, jumping from 1-2 to 1-3.

6) And (3) judging the label in 1-3, and if the chat robot does not say 1-3, walking the branch according to the judgment result, walking the branch 1-4 if the condition 1 is met, walking the branch 1-5 if the condition 2 is met, and walking the general branch if the condition is not met.

In the embodiment, the internal jumping function is realized and is used after the scenario 1-2; if the scenario reaches 1-3,1-3 from 1-2, the scenario will not be shown; in 1-3, no user input is required; if the 1-3 scenario does not stop, the 1-4 scenario is displayed after the 1-3 scenario enters the 1-4 scenario from the 1-3 scenario, and therefore after third input information aiming at the target node scenario is obtained, two node jumps can be automatically realized, the flexibility of the node jumps is effectively improved, and the intelligence of voice interaction is ensured.

In an optional embodiment, the determining a target branch path matching the first input information includes: if the first plot interaction node is provided with the plot merging identifier, eliminating a plurality of branch paths preset by the first plot interaction node; determining a merged scenario corresponding to the node scenario of the first scenario interaction node, and taking the merged scenario as a unique outlet of the first scenario interaction node; and directly taking the branch path corresponding to the combined scenario as the target branch path.

In this embodiment, when the scenario merging function is started, other branch paths except the branch path corresponding to the merged scenario are eliminated to obtain a unique target branch path, and then directional skip of the scenario interaction node is realized based on the target branch path, and no additional user input is required in the process of outputting the merged scenario, so that flexibility of voice interaction can be effectively improved.

Further, under the condition that the first scenario interaction node starts the scenario merging function and the target branch path corresponds to the internal skip identifier, the chat robot enters the internal skip + scenario merging flow, and the following example illustrates an implementation manner of the internal skip + scenario merging flow:

1) The chat robot says 1-1.

2) And the user replies to obtain the input information.

3) And judging the matching of the guide conditions of the input information, if the matching is successful, jumping from 1-1 to 1-2, when the chatting robot is in 1-2, saying 1-2, directly jumping from 1-2 to 1-3 without the reply of the user because of the effect of merging the scenarios and after saying 1-2, and the chatting robot does not say 1-3 because of the effect of internal jumping.

4) In 1-3, the label can be judged, branches are taken according to the judgment result, if the condition 1 is met, the branches are taken by 1-4, if the condition 2 is met, the branches are taken by 1-5, and if the condition is not met, the general branches are taken (namely, the general node scenario is output); wherein the source of the tags is the content extracted from the user responses of the previous rounds of conversations.

5) And (4) jumping to 1-4 connected with 1-3 by judging 1-3.

6) If the chat robot says 1-4, and 1-4 corresponds to the matching result in 1-3, the reply aiming at the input information of 1-1 by the user can be realized.

It can be seen that in the case of the combined scenario + the internal skip, after the user speaks to reply in 1-1, the chat robot speaks the words 1-2 and 1-4, that is, after the user speaks a sentence, the chat robot automatically realizes the skip of a plurality of nodes and outputs a plurality of voices.

Further, specific scenario examples are as follows:

in this example, the user says that he wants to eat apple, 1-3 transfers to the next scenario 1-4, replies "{ apple } with it, and eat fruit can beautify skin and nourish face! "; if the user says that the user wants to eat other foods, 1-3 can transfer different next-step scenarios, for example, the user wants to eat beef, the next step is 'beef', and resistance can be enhanced by eating more meat; when people want to eat the ice cream, people say that the ice cream is popular corresponding to the next step, people can enjoy the mood when eating more desserts, and the like.

The method simultaneously uses an 'internal jump' and a 'merge scenario', the former plays a role in judging the jump and hides 1-3 through the transfer scenario, the latter lets 1-2 and 1-4 appear simultaneously, otherwise, a user needs to say what after 1-2 and then 1-4 appears.

In the above embodiment, the first scenario interaction node is skipped based on the first input information, after the process of internal skipping and scenario merging is entered, the skipping of the scenario node can be automatically determined and the first input information is responded without new input by the user, and the two functions of internal skipping and scenario merging are combined for use, so that the linking effect can be achieved.

In an optional embodiment, before the responding the input information by the second scenario interaction node, the method further includes: acquiring a history variable label corresponding to history input information of a user; when the first plot interactive node corresponds to input information, the historical input information comprises input information corresponding to the first plot interactive node; acquiring a predetermined variable label set; comparing the historical variable labels with variable labels in the variable label set, and obtaining difference variable labels according to comparison results; generating a node plot of a next plot interaction node of the second plot interaction node based on the difference variable label; correspondingly, after the second scenario interaction node responds to the input information, the method further comprises the following steps: and continuing to jump to the next scenario interaction node, and converting the node scenario of the next scenario interaction node into voice and outputting the voice.

Optionally, in this embodiment of the application, each time the user speaks, the chat robot performs semantic understanding on the input information of the user, and extracts key information (which may be a keyword) in the input information according to a semantic understanding result to fill in a preset tag. And each label may be pre-set with a corresponding response answer or response scenario.

The historical input information of the user can include the first input information, and also can include a historical keyword corresponding to the first input information, wherein the historical keyword is a keyword which is determined based on the previous guide condition and is matched with the first input information (historical input), and the keyword can be used as a label.

Optionally, after entering the second scenario interaction node, the historical input information may be analyzed based on the second scenario interaction node to determine a node scenario of the next scenario interaction node.

The process of determining the scenario of the node is as follows:

different transfer branches are set under the second plot interaction node, different branches set different preconditions, and corresponding branches can be taken only if specific conditions are met, for example, a chat robot selling mobile phones needs a user to say a desired brand, color and budget to give corresponding recommendations. For example, for a keyword spoken by a user, the chat robot stores corresponding variables, judges which branch should be taken at the second scenario interaction node (judges which scenario should be taken next), asks for budget at the branch with 1 default, asks for color at the branch with 2 default, asks for brand at the branch with 3 default, and makes a recommendation at the branch with 4 default, and the chat robot judges at the second scenario interaction node by setting which variables are missing as preconditions for branching. In this example, assuming that the user has replied to the brand and the color, the chat robot will enter the next branch of inquiring the budget, and will not repeatedly inquire the existing variables such as the brand, the color, etc., so as to complete the whole process of selling the mobile phone.

According to the embodiment, the difference variable tags are determined through comparison of the history variable tags and the variable tag set, the node scenario of the next scenario interaction node is further determined based on the difference variable tags, the determined node scenario of the next scenario interaction node has strong pertinence, the true requirement of the user can be accurately positioned through the problem step-by-step selection process, the node scenario matched with the intention of the user is further output, and accurate commodity recommendation can be achieved in a commodity recommendation scene.

It should be understood that, although the steps in the above-described flowcharts 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 a part of the steps in the above flowcharts may include multiple steps or multiple stages, which are not necessarily performed at the same time, but may be performed at different times, and the order of performing the steps or stages is not necessarily performed in sequence, but may be performed alternately or alternately with other steps or at least a part of the steps or stages in other steps.

Based on the same idea as the interactive node skipping method in the above embodiment, the present invention further provides an interactive node skipping apparatus, which may be used to execute the interactive node skipping method. For convenience of illustration, in the structural schematic diagram of the embodiment of the interactive node skipping apparatus, only the part related to the embodiment of the present invention is shown, and those skilled in the art will understand that the illustrated structure does not constitute a limitation to the apparatus, and may include more or less components than those illustrated, or combine some components, or arrange different components.

In one embodiment, as shown in fig. 3, there is provided an interactive node jump apparatus 300, which may be a part of a computer device using a software module or a hardware module, or a combination of the two, and specifically includes:

a branch path determining module 301, configured to determine a target branch path matching first input information when jumping to a first scenario interaction node based on the first input information of a user;

an identifier determining module 302, configured to determine whether the target branch path corresponds to an internal skip identifier;

a node skipping module 303, configured to skip to a corresponding second scenario interaction node based on the target branching path and hide a node scenario of the second scenario interaction node if the target branching path corresponds to the internal skipping identifier;

an input response module 304, configured to respond to the input information at the second scenario interaction node.

In the interactive node skipping device, the branch path matched with the input information is determined, the skipping of the scenario interactive node is realized by combining the judgment result of the internal skipping identification, the quick response to the input information and the quick skipping of the scenario interactive node can be realized, the flexibility is good, and the intelligent interactive node skipping can be realized

In an optional embodiment, further comprising:

the interactive node skipping module is used for skipping to a corresponding third plot interactive node based on the target branch path if the target branch path does not have an internal skipping identifier correspondingly;

the input information acquisition module is used for acquiring second input information based on the third plot interaction node;

and the node continuous skipping module is used for continuing node skipping if the second input information accords with the guiding condition of the third plot interaction node.

In an alternative embodiment, the input response module includes:

a response scenario determination sub-module configured to determine a response scenario corresponding to the first input information;

and the first input response submodule is used for converting the response scenario into voice and outputting the voice so as to respond to the first input information.

In an alternative embodiment, the input response module includes:

a response answer determination submodule for determining a response answer corresponding to the first input information;

and the second input response submodule is used for converting the response answer into voice and outputting the voice so as to respond to the first input information.

In an alternative embodiment, the branch path determining module includes:

the branch path obtaining sub-module is used for obtaining a plurality of branch paths preset by the first plot interactive node if the plot merging identifier is not set by the first plot interactive node;

a branch path determination submodule configured to determine, as the target branch path, a branch path that matches the first input information from the plurality of branch paths.

In an alternative embodiment, the branch path determining module includes:

the branch path eliminating submodule is used for eliminating a plurality of branch paths preset by the first plot interactive node if the first plot interactive node is provided with the plot merging identifier;

the exit determining submodule is used for determining a combined scenario corresponding to the node scenario of the first scenario interaction node, and taking the combined scenario as the only exit of the first scenario interaction node;

and the target branch path determining submodule is used for directly taking the branch path corresponding to the combined scenario as the target branch path.

In an optional embodiment, further comprising:

the variable label acquisition module is used for acquiring a historical variable label corresponding to historical input information of a user; when the first plot interactive node corresponds to input information, the historical input information comprises input information corresponding to the first plot interactive node;

the tag set acquisition module is used for acquiring a predetermined variable tag set;

the tag comparison module is used for comparing the historical variable tags with the variable tags in the variable tag set and obtaining difference variable tags according to comparison results;

a node scenario generating module, configured to generate a node scenario of a next scenario interaction node of the second scenario interaction node based on the difference variable tag;

correspondingly, the method also comprises the following steps:

and the plot output module is used for continuously jumping to the next plot interaction node, converting the node plot of the next plot interaction node into voice and outputting the voice.

For the specific definition of the interactive node skipping apparatus, reference may be made to the above definition of the interactive node skipping method, which is not described herein again. All or part of the modules in the interactive node jumping device can be 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.

A dialog development system, as shown in fig. 4, comprising: an internal hop unit 401, a node hop unit 402, and a response unit 403. These functional units implement the following functions:

The dialogue development system comprises an internal jumping unit, a node jumping unit and a response unit, determines a branch path matched with input information by combining the unit modules, realizes jumping of scenario interaction nodes by combining a judgment result of an internal jumping identifier, can realize quick response to the input information and quick jumping of the scenario interaction nodes, has good flexibility, and can realize intelligent interaction.

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, and a network interface 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 operating system and the computer program to run on the non-volatile storage medium. The database of the computer device is used for storing data. 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 an interactive node hopping method.

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, a computer device is further provided, which includes a memory and a processor, the memory stores a computer program, and the processor implements the steps of the above method embodiments when executing the computer program.

In an embodiment, a computer-readable storage medium is provided, in which a computer program is stored which, when being executed by a processor, carries out the steps of the above-mentioned method embodiments.

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 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 can include at least one of non-volatile and volatile memory. Non-volatile Memory may include Read-Only Memory (ROM), magnetic tape, floppy disk, flash Memory, optical storage, or the like. Volatile Memory can include Random Access Memory (RAM) or external cache Memory. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM), among others.

It should be noted that "based on" in the embodiment of the present application is not meant in a closed manner, for example, performing B actions based on a means that not only B actions according to a may be performed, but also B actions according to other contents besides a, for example, performing B actions based on a may include performing B actions according to C except a and a.

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-mentioned embodiments 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, and these are all 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. An interactive node jump method is characterized by comprising the following steps:

responding to the input information at the second plot interactive node;

further comprising:

executing a task corresponding to the internal jumping identification acting on the second plot interaction node according to the internal jumping identification;

after entering a second plot interaction node, hiding the node plot of the second plot interaction node, acquiring latest user input information, and continuously taking the latest user input information as the user input information of the second plot interaction node;

if the first plot interaction node has the user input information, continuing to use the user input information of the first plot interaction node as the user input information of the second plot interaction node, and processing once again; and if the first plot interaction node does not have the user input information, acquiring the latest user input information before the first plot interaction node, and continuously using the latest user input information as the user input information of the second plot interaction node.

2. The method according to claim 1, wherein after determining whether the target branch path corresponds to the internal jump identifier, further comprising:

acquiring second input information based on the third plot interaction node;

3. The method of claim 1, wherein said responding to input information at said second storyline interactive node comprises:

determining a response scenario corresponding to the first input information;

4. The method of claim 1, wherein said responding to input information at said second storyline interactive node comprises:

determining a response answer corresponding to the first input information;

5. The method of claim 1, wherein determining the target branch path matching the first input information comprises:

6. The method of claim 1, wherein determining the target branch path matching the first input information comprises:

7. The method of any of claims 1 to 6, wherein prior to said responding to input information by said second scenario interaction node, further comprising:

acquiring a historical variable label corresponding to historical input information of a user; when the first plot interactive node corresponds to input information, the historical input information comprises input information corresponding to the first plot interactive node;

acquiring a predetermined variable label set;

and continuing to jump to the next scenario interaction node, and converting the node scenario of the next scenario interaction node into voice and outputting the voice.

8. An interactive node jump apparatus, comprising:

the node skipping module is used for skipping to a corresponding second scenario interaction node based on the target branch path and hiding the node scenario of the second scenario interaction node if the target branch path corresponds to the internal skipping identifier;

the input response module is used for responding the input information at the second plot interactive node;

further comprising:

if the first plot interaction node has the user input information, continuing to take the user input information of the first plot interaction node as the user input information of the second plot interaction node, and processing once again; and if the first plot interaction node does not have the user input information, acquiring the latest user input information before the first plot interaction node, and continuously using the latest user input information as the user input information of the second plot interaction node.

9. A dialog development system, comprising: an internal jumping unit, a node jumping unit and a response unit;

if the target branch path corresponds to the internal jump identifier, the internal jump unit triggers internal jump to a corresponding second scenario interaction node based on the target branch path and hides the node scenario of the second scenario interaction node;

the response unit responds to the input information at the second scenario interactive node;

further comprising:

10. A computer device comprising a memory and a processor, the memory storing a computer program, wherein the processor implements the steps of the method of any one of claims 1 to 7 when executing the computer program.

11. A computer-readable storage medium, storing a computer program, wherein the computer program, when executed by a processor, performs the steps of the method according to any one of claims 1 to 7.