CN109544195B - Information processing method and electronic equipment - Google Patents
Information processing method and electronic equipment Download PDFInfo
- Publication number
- CN109544195B CN109544195B CN201811581667.XA CN201811581667A CN109544195B CN 109544195 B CN109544195 B CN 109544195B CN 201811581667 A CN201811581667 A CN 201811581667A CN 109544195 B CN109544195 B CN 109544195B
- Authority
- CN
- China
- Prior art keywords
- session
- information
- determining
- current
- different
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 230000010365 information processing Effects 0.000 title claims abstract description 22
- 238000003672 processing method Methods 0.000 title claims abstract description 19
- 238000000034 method Methods 0.000 claims description 24
- 230000006399 behavior Effects 0.000 claims description 13
- 230000004044 response Effects 0.000 claims description 13
- 230000008451 emotion Effects 0.000 claims description 11
- 238000004364 calculation method Methods 0.000 description 17
- 230000000739 chaotic effect Effects 0.000 description 3
- 230000002159 abnormal effect Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/01—Customer relationship services
Landscapes
- Business, Economics & Management (AREA)
- Accounting & Taxation (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Finance (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Telephonic Communication Services (AREA)
Abstract
The application provides an information processing method and electronic equipment, which are applied to an intelligent customer service session system, wherein the intelligent customer service session system can respond to received input information to automatically provide feedback information, and the information processing method comprises the following steps: determining dialogue state rounds to the intelligent customer service session system end to acquire session information; determining a session reset degree based on the session information; the intelligent customer service session system can automatically respond to the reset operation of the current session based on the session information without manually resetting the user, thereby reducing the operation burden of the user and improving the user experience.
Description
Technical Field
The present invention relates to the field of information processing technologies, and in particular, to an information processing method and an electronic device.
Background
In an intelligent conversational system, the system is able to converse with the user and give relevant feedback.
In the process of the system and the user dialogue, the problem that the system cannot correctly understand the accurate intention of the user input information often occurs, and if the system continuously sends out inquiry information to the user, the session is terminated or the user is inspired. In this case, in the existing intelligent session system, a session reset operation may be performed by the user, for example, the user clicks a "reset" button or the user inputs a reset session statement, so that the session between the system and the user can be restarted.
However, the manner in which the user performs the session resetting operation increases the burden of the user operation and reduces the user experience.
Disclosure of Invention
In view of the above, the present invention provides an information processing method and an electronic device to solve the above technical problems.
In order to achieve the above purpose, the present invention provides the following technical solutions:
An information processing method applied to an intelligent customer service session system, the intelligent customer service session system being capable of responding to received input information to automatically provide feedback information, the information processing method comprising:
Determining dialogue state rounds to the intelligent customer service session system end to acquire session information;
determining a session reset degree based on the session information;
and determining that the session resetting degree meets a first preset condition, and responding to the resetting operation of the current session.
Preferably, the determining the session reset degree based on the session information includes:
determining current session attributes of different session categories based on the session information;
And determining a session reset degree based on the current session attribute.
Preferably, the determining the session resetting degree based on the current session attribute includes:
determining a target information value corresponding to the current session attribute based on the first corresponding relation; the first corresponding relation is used for representing the corresponding relation between different session attributes and different information values under the same session category;
determining weight values corresponding to different session categories based on the second correspondence; the second corresponding relation is used for representing the corresponding relation between different session categories and different weight values;
a session reset value is determined based on the weight values of the different session categories and the information values of the current session attributes.
Preferably, the determining that the session resetting degree meets the first preset condition includes:
And determining that the session reset value is greater than or equal to a first threshold.
Preferably, the method further comprises:
determining that the session reset value is smaller than a first threshold and larger than a second threshold, and prohibiting the reset operation of the current session; wherein the first threshold is greater than the second threshold.
Preferably, the first threshold is determined by:
Determining current session attributes of different session categories under a preset session threshold;
Determining a target information value corresponding to the current session attribute based on the third corresponding relation; the third corresponding relation is used for representing the corresponding relation between different session attributes and different information values under the same session category;
Determining weight values corresponding to different session categories based on the fourth correspondence; the fourth corresponding relation is used for representing the corresponding relation between different session categories and different weight values;
And determining the first threshold value based on the weight values of different session categories, the information values of the current session attribute and a preset first coefficient.
An electronic device for use in an intelligent customer service session system that is capable of automatically providing feedback information in response to received input information, the electronic device comprising:
A memory for storing a program;
The processor is used for executing the program, the program is used for determining dialogue state rounds to the intelligent customer service session system end, obtaining session information, determining the session resetting degree based on the session information, determining that the session resetting degree meets a first preset condition, and responding to the resetting operation of the current session.
Preferably, the memory stores a first corresponding relation and a second corresponding relation, wherein the first corresponding relation is used for representing the corresponding relation between different session attributes and different information values in the same session category, and the second corresponding relation is used for representing the corresponding relation between different session categories and different weight values;
The processor executes the program, and is specifically configured to determine current session attributes of different session categories based on the session information, determine a target information value corresponding to the current session attributes based on the first correspondence, determine weight values corresponding to the different session categories based on the second correspondence, and determine a session reset value based on the weight values of the different session categories and the information value of the current session attributes.
Preferably, the processor executes the program to determine that the session reset degree meets a first preset condition, including: determining that the session reset value is greater than or equal to a first threshold;
The memory is further used for storing a third corresponding relation and a fourth corresponding relation, the third corresponding relation is used for representing the corresponding relation between different session attributes and different information values in the same session category, and the fourth corresponding relation is used for representing the corresponding relation between different session categories and different weight values;
The processor executing the program determines the first threshold by:
Determining current session attributes of different session categories under a preset session threshold, determining a target information value corresponding to the current session attribute based on a third corresponding relation, determining a weight value corresponding to the different session categories based on a fourth corresponding relation, and determining the first threshold based on the weight values of the different session categories, the information value of the current session attribute and a preset first coefficient.
An electronic device for use in an intelligent customer service session system that is capable of automatically providing feedback information in response to received input information, the electronic device comprising:
the information acquisition unit is used for determining dialogue state rounds to the intelligent customer service session system end and acquiring session information;
a first determining unit configured to determine a session resetting degree based on the session information;
and the first response unit is used for determining that the session resetting degree meets a first preset condition and responding to the resetting operation of the current session.
Compared with the prior art, the embodiment of the application provides an information processing method, which is applied to an intelligent customer service session system, wherein the intelligent customer service session system can respond to received input information to automatically provide feedback information, and particularly, the session information is acquired, a dialogue state round is determined to an intelligent customer service session system end, the session reset degree is determined based on the session information, the session reset degree meets a first preset condition, and the current session reset operation is responded, so that the intelligent customer service session system can automatically respond to the current session reset operation based on the session information without manually resetting a user, thereby reducing the operation burden of the user and improving the user experience.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are required to be used in the embodiments or the description of the prior art will be briefly described below, and it is obvious that the drawings in the following description are only embodiments of the present invention, and that other drawings can be obtained according to the provided drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of an information processing method according to a first embodiment of the present application;
fig. 2 is a flow chart of an information processing method according to a second embodiment of the present application;
Fig. 3 is a schematic flow chart of an information processing method according to a third embodiment of the present application;
fig. 4 is a flow chart of a first threshold determining method according to a fourth embodiment of the present application;
Fig. 5 is a schematic structural diagram of an electronic device according to a first embodiment of the present application;
fig. 6 is a schematic structural diagram of an electronic device according to a fifth embodiment of the present application.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
The first embodiment of the application provides an information processing method which is applied to an intelligent customer service session system, wherein the intelligent customer service session system can respond to received input information to automatically provide feedback information. That is, when a user inputs information to the intelligent customer service session system, the intelligent customer service session system responds based on the input information of the user to give feedback information about the input information. It can be seen that the intelligent customer service session system can simulate a real person, so that a user can have a feeling of chatting with the real person in the intelligent customer service session system.
As shown in fig. 1, the method comprises the steps of:
Step 101: determining dialogue state rounds to an intelligent customer service session system end to acquire session information;
In the process of a conversation with the intelligent customer service conversation system, the conversation state can be rotated, for example, the user firstly presents a problem or describes information, then the intelligent customer service system can receive input information of the user, respond to the input information to output feedback information, and wait for the response of the user to the feedback information or input the feedback information again by the user. Then, determining the dialogue state round to the intelligent customer service session system end can be characterized as the state when the intelligent customer service session system end receives the input information, or the state before the intelligent customer service session system end outputs the feedback information.
The intelligent customer service session system acquires session information in the session process with the user, wherein the session information at least comprises input information received currently when the dialogue state turns to the intelligent customer service session system end, and the session information can also comprise all session information before the dialogue state turns to the intelligent customer service session system end.
Step 102: determining a session reset degree based on the session information;
the session reset degree is used for representing the degree of needing to reset the current session, specifically, the session reset degree is determined based on the session information, and the following processes can be included:
(1) Determining current session attributes of different session categories based on the session information;
The intelligent customer service session system is divided into a plurality of session categories aiming at the session resetting degree and a plurality of session attributes under each session category in advance, so that the session information is processed based on a natural language understanding algorithm to determine the current session attributes under different session categories.
As one implementation, the session category may include at least one of the following session categories:
A current session stage;
current user behavior;
Current user feedback;
current user emotion;
User history feedback;
Status confusion.
Among the session attributes that the current session stage has include, but are not limited to: a start session attribute, a locate problem attribute, a solve problem attribute, an end session attribute, a hello chat attribute.
Session attributes that current user behavior has include, but are not limited to: greeting attributes, chat attributes, offer information attributes, propose questions attributes, explain questions attributes.
Current user feedback has session attributes including, but not limited to: consent attribute, unknown attribute, objection attribute, strong objection attribute.
Session attributes that current user emotions have include, but are not limited to: happiness, normal, disappointment, anger.
The user history feedback has session attributes including, but not limited to: consent and objection.
The state confusion is used for representing the degree of session confusion between a user and the system, and specifically, the longer the path from the initial state to the completion state is, the more the repeated states are, so that the higher the state confusion is. Under a certain session scene, only d state jumps are needed from the problem solving to the problem solving, and the state jumps y times in the real dialogue process of the user and the intelligent customer service session system, if y < d, the session attribute of the state confusion is normal, and if y is more than or equal to d, the session attribute of the state confusion is abnormal.
Because the session information comprises the input information received currently when the dialogue state turns to the intelligent customer service session system end, the current session attribute of different session categories is determined based on the input information received currently. And the user history feedback and the state confusion degree need to use all session information before the dialogue state rotation reaches the intelligent customer service session system end, if the session information comprises all session information before the dialogue state rotation reaches the intelligent customer service session system end, the current session attribute of the user history feedback and the current session attribute of the state confusion degree can be determined.
For example, a user's dialogue with an intelligent customer service dialogue system is as follows:
user input 1-system feedback input 1-user input 2-system feedback input 2-user input 3-determine dialog state rounds to the intelligent customer service session system end- ….
The session information includes at least input3 entered by the current user, and based on input3, the current session attributes of the different session categories may be determined. Specifically, the current session attribute of the current session stage may be determined based on input3, and if input3 is the chat information of the intelligent customer service session system, the current session attribute of the current session stage is the hello chat attribute. The current session attribute of the current user behavior may also be determined based on input3, such as determining that the current session attribute of the current user behavior is a boring attribute. The current session attribute of the current user feedback can also be determined based on input3, and if input3 contains consent information for system feedback content 2, then the current session attribute of input3 fed back by the current user can be determined to be the consent attribute. The current session attribute of the current user emotion may also be determined based on input3, and if the emotion of the user when input3 is input is happy, the current session attribute of the current user emotion may be determined to be a happy attribute.
If the session information further includes all session information before the session state turns to the intelligent customer service session system end, the current session attribute of the user history feedback and the current session attribute of the state confusion can be determined based on the session information.
(2) And determining a session reset degree based on the current session attribute.
After determining the current session attribute of different session categories based on the session information, the session resetting degree can be determined based on the current session attribute, and the specific determination mode is not limited by the application, for example, different numerical values are given to different session attributes of each session category in advance, the total value of the current session attribute of different session categories is calculated, and the total value is used as the session resetting degree.
Step 103: and determining that the session resetting degree meets a first preset condition, and responding to the resetting operation of the current session.
The first preset condition may be set based on an actual situation, such as presetting a threshold, and if the calculated total value is greater than the preset threshold, responding to a reset operation of the current session.
Therefore, in the first embodiment of the method, the session information is acquired, the dialogue state turn is determined to the intelligent customer service session system end, the session resetting degree is determined based on the session information, the session resetting degree is determined to meet the first preset condition, and the resetting operation of the current session is responded, so that the resetting operation of the current session based on the automatic response of the session information is realized, the user does not need to manually reset, the operation burden of the user is reduced, and the user experience is improved.
A second embodiment of the present application provides an information processing method for describing in detail an implementation manner of determining a session reset degree based on a current session attribute, as shown in fig. 2, the information processing method includes the following steps:
Step 201: determining dialogue state rounds to the intelligent customer service session system end to acquire session information;
Step 202: determining current session attributes of different session categories based on the session information;
step 203: determining a target information value corresponding to the current session attribute based on the first corresponding relation;
Step 204: determining weight values corresponding to different session categories based on the second correspondence;
the intelligent customer service session system is provided with a first corresponding relation and a second corresponding relation in advance. The first correspondence is used for representing the correspondence between different session attributes and different information values in the same session category, that is, the first correspondence records information values corresponding to each session attribute in each session category.
The second correspondence is used for representing the correspondence between different session categories and different weight values, that is, the weight value corresponding to each session category is recorded in the second correspondence.
For example, the first correspondence may be as follows:
Current session stage: an initial session attribute-5, a locate problem attribute-4, a solve problem attribute-3, an end session attribute-2, and a hello chat attribute-1. Wherein, the larger the information value is, the more likely to cause confusion, and the more likely to need to be reset.
Current user behavior: greeting attribute-5, chat attribute-4, offer information attribute-3, ask question attribute-2, explain question attribute-1.
Current user feedback: consent attribute-0, unknown attribute-1, objection attribute-2, strong objection attribute-6.
Current user emotion: happiness-0, normal-1, disappointment-2, anger-6.
User history feedback: number of consent and number of objection. If the user continuously makes countermeasures for multiple times, the intention understanding before the system is described to be problematic or the dialogue state is disordered, and corresponding information values can be set according to the continuous countermeasures, wherein the larger the countermeasures are, the larger the information values are, the larger the agreeing times are, and the information values are smaller.
Status confusion degree: let the state jump from problem raising to normal need for problem solving be d, and the state jump in the real dialogue process of the user and the system be y, then the state confusion s is as follows:
the second correspondence may be as follows:
current session stage-1
Current user behavior-2
Current user feedback-2
Current user emotion-10
User history feedback-5
Status confusion degree-5
It should be noted that the first correspondence relationship and the second correspondence relationship are merely specific examples for easy understanding, and the present application is not limited thereto.
Step 205: determining a session reset value based on the weight values of the different session categories and the information values of the current session attribute;
Specifically, the session reset value may be calculated using a first calculation formula, where the first calculation formula is as follows:
Wherein R represents a session reset value, a i represents a weight value of the ith session class, p i represents an information value of the ith class, and n represents the number of session classes.
For ease of understanding, a specific example will be described, in which a target information value corresponding to a current session attribute is determined based on a first correspondence, and a weight value corresponding to a different session category is determined based on a second correspondence, as shown in table 1 below:
TABLE 1
Then, as can be derived from the first calculation formula described above,
The session reset value r=5×1+2×2+1×2+2×10+2×5+3×5=56
Step 206: and determining that the session reset value is greater than or equal to a first threshold, and responding to the reset operation of the current session.
In this embodiment, the system may preset a specific value of the first threshold. Other ways of determining the first threshold are also described in other embodiments of the present application, please refer to the fourth embodiment of the method of the present application in detail.
Therefore, in the second embodiment of the method, the session resetting degree can be represented by the session resetting value, and when the session resetting value is larger than the first threshold, the resetting operation of the current session is responded, so that the resetting operation of the current session is automatically responded, the user does not need to manually reset, the operation burden of the user is reduced, and the user experience is improved.
A third embodiment of the present application provides an information processing method, as shown in FIG. 3, including the following steps:
Step 301: determining dialogue state rounds to the intelligent customer service session system end to acquire session information;
Step 302: determining current session attributes of different session categories based on the session information;
Step 303: determining a target information value corresponding to the current session attribute based on the first corresponding relation;
Step 304: determining weight values corresponding to different session categories based on the second correspondence;
step 305: determining a session reset value based on the weight values of the different session categories and the information values of the current session attribute;
Step 306: determining that the session reset value is greater than or equal to a first threshold, and responding to a reset operation of a current session;
Step 307: determining that the session reset value is smaller than a first threshold and larger than a second threshold, and prohibiting the reset operation of the current session;
wherein the first threshold is greater than the second threshold.
In this embodiment, the system may preset specific values of the first threshold and the second threshold. Other ways of determining the first threshold and the second threshold are also described in other embodiments of the present application, please refer to the fourth embodiment of the method of the present application in detail.
Therefore, in the third embodiment of the method, the session resetting degree can be represented by the session resetting value, and when the session resetting value is larger than the first threshold, the resetting operation of the current session is responded, so that the resetting operation of the current session is automatically responded, the user does not need to manually reset, the operation burden of the user is reduced, and the user experience is improved; and when the session reset value is smaller than the first threshold and larger than the second threshold, the reset operation of the current session is forbidden, so that the error rate of the reset operation is reduced.
In a fourth embodiment of the method of the present application, as shown in fig. 4, the first threshold value is determined by:
step 401: determining current session attributes of different session categories under a preset session threshold;
the intelligent customer service session system is not only divided into a plurality of session categories aiming at the session resetting degree and a plurality of session attributes under each session category in advance, but also divided into a plurality of session categories aiming at the session threshold value and a plurality of session attributes under each session category in advance.
Note that the plurality of session categories for the session reset degree are not exactly the same as the plurality of session categories for the session threshold.
As one implementation, the session category for the meeting threshold may include at least one of the following session categories:
A current session stage;
User tolerance;
the user resets the preference.
Among the session attributes that the current session stage has include, but are not limited to: a start session attribute, a locate problem attribute, a solve problem attribute, an end session attribute, a hello chat attribute.
User tolerance has session attributes including, but not limited to: weak, general, strong.
Session attributes that user reset preferences have include, but are not limited to: dislike, like.
The current session attribute of the user tolerance and the current session attribute of the user reset preference can be obtained based on historical session data statistics of the user who performs a dialogue with the intelligent customer service session system. The current session attribute of the current session stage may be determined using the session information.
Step 402: determining a target information value corresponding to the current session attribute based on the third corresponding relation;
step 403: determining weight values corresponding to different session categories based on the fourth correspondence;
The intelligent customer service session system is provided with a third corresponding relation and a fourth corresponding relation in advance. The third correspondence is used for representing the correspondence between different session attributes and different information values in the same session category, that is, the third correspondence records information values corresponding to each session attribute in each session category.
The fourth correspondence is used for representing the correspondence between different session categories and different weight values, that is, the weight value corresponding to each session category is recorded in the fourth correspondence.
For example, the third correspondence may be as follows:
Current session stage: an initial session attribute-5, a locate problem attribute-4, a solve problem attribute-3, an end session attribute-2, and a hello chat attribute-1.
User tolerance: very weak-0, weak-1, general-2, strong-3, very strong-4.
User restart preferences: dislike-0, no so-called-1, like-2.
The fourth correspondence is as follows:
Current session stage-2
User tolerance-2
The user restarts preference-2.
It should be noted that the third correspondence relationship and the fourth correspondence relationship are only specific examples for easy understanding, and the present application is not limited thereto.
Step 404: and determining the first threshold value based on the weight values of different session categories, the information values of the current session attribute and a preset first coefficient.
The first coefficient is a preset coefficient used for calculating the first threshold value.
Specifically, the first threshold may be calculated using a second calculation formula, where the second calculation formula is as follows:
Wherein U represents a first threshold, f i represents a weight value of the ith session category, x i represents an information value of the ith category, m represents the number of session categories, and k is an upper limit coefficient. It should be noted that, E is a chaotic coefficient, and is an empirical value, and E may be added to the first threshold value as in the second calculation formula, or may be not added.
The manner of determining the second threshold may refer to the manner of determining the first threshold, and may be different from the manner of determining the first threshold only in that the value of k is different. K is a lower coefficient when calculating the second threshold. The upper limit coefficient and the lower limit coefficient are empirical values, and are usually set differently for different dialogue systems and different users. The specific intelligent session system can feed back the correction coefficient k according to the user behavior so as to meet the corresponding user.
For ease of understanding, a brief description is given with reference to table 2:
TABLE 2
Then, based on the second calculation formula, it can be derived,
First threshold u1=50+1.5 (5×1+2×2+1×2) =66.5
Second threshold u2=50+0.8 (5×1+2×2+1×2) =58.8
When the session reset value R is more than or equal to U1, immediately performing reset operation; when U1> R > U2, a reset operation for the current session is disabled.
Corresponding to the above-mentioned information processing method, the embodiment of the application also provides an electronic device, which is described below through several device embodiments.
The first embodiment of the application provides an electronic device which is applied to an intelligent customer service session system, wherein the intelligent customer service session system can respond to received input information to automatically provide feedback information. That is, when a user inputs information to the intelligent customer service session system, the intelligent customer service session system responds based on the input information of the user to give feedback information about the input information. It can be seen that the intelligent customer service session system can simulate a real person, so that a user can have a feeling of chatting with the real person in the intelligent customer service session system.
As shown in fig. 5, an electronic device includes a memory 100 and a processor 200;
wherein, the memory 100 is used for storing programs;
The processor 200 is configured to execute the program, where the program is configured to determine a session state round to the intelligent customer service session system side, obtain session information, determine a session reset degree based on the session information, determine that the session reset degree meets a first preset condition, and respond to a reset operation of a current session.
In the process of a conversation with the intelligent customer service conversation system, the conversation state can be rotated, for example, the user firstly presents a problem or describes information, then the intelligent customer service system can receive input information of the user, respond to the input information to output feedback information, and wait for the response of the user to the feedback information or input the feedback information again by the user. Then, determining the dialogue state round to the intelligent customer service session system end can be characterized as the state when the intelligent customer service session system end receives the input information, or the state before the intelligent customer service session system end outputs the feedback information.
The intelligent customer service session system acquires session information in the session process with the user, wherein the session information at least comprises input information received currently when the dialogue state turns to the intelligent customer service session system end, and the session information can also comprise all session information before the dialogue state turns to the intelligent customer service session system end.
The session reset level is used for representing the level of the current session to be reset, specifically, the processor executes the program to determine the session reset level based on the session information, which may include:
(1) Determining current session attributes of different session categories based on the session information;
the intelligent customer service session system is provided with a natural language understanding algorithm, and a plurality of session categories aiming at the session resetting degree and a plurality of session attributes under each session category are divided in advance, so that the session information is processed based on the natural language understanding algorithm to determine the current session attributes under different session categories.
As one implementation, the session category may include at least one of the following session categories:
A current session stage;
current user behavior;
Current user feedback;
current user emotion;
User history feedback;
Status confusion.
Among the session attributes that the current session stage has include, but are not limited to: a start session attribute, a locate problem attribute, a solve problem attribute, an end session attribute, a hello chat attribute.
Session attributes that current user behavior has include, but are not limited to: greeting attributes, chat attributes, offer information attributes, propose questions attributes, explain questions attributes.
Current user feedback has session attributes including, but not limited to: consent attribute, unknown attribute, objection attribute, strong objection attribute.
Session attributes that current user emotions have include, but are not limited to: happiness, normal, disappointment, anger.
The user history feedback has session attributes including, but not limited to: consent and objection.
The state confusion is used for representing the session confusion degree between the user and the system, and specifically, the longer the path from the initial state to the completion state is, the more the repeated states are, so that the higher the state wheel confusion degree is. Under a certain session scene, only d state jumps are needed from the problem solving to the problem solving, and the state jumps y times in the real dialogue process of the user and the intelligent customer service session system, if y < d, the session attribute of the state confusion is normal, and if y is more than or equal to d, the session attribute of the state confusion is abnormal.
Because the session information comprises the input information received currently when the dialogue state turns to the intelligent customer service session system end, the current session attribute of different session categories is determined based on the input information received currently. And the user history feedback and the state confusion degree need to use all session information before the dialogue state rotation reaches the intelligent customer service session system end, if the session information comprises all session information before the dialogue state rotation reaches the intelligent customer service session system end, the current session attribute of the user history feedback and the current session attribute of the state confusion degree can be determined.
(2) And determining a session reset degree based on the current session attribute.
After determining the current session attribute of different session categories based on the session information, the session resetting degree can be determined based on the current session attribute, and the specific determination mode is not limited by the application, for example, different numerical values are given to different session attributes of each session category in advance, the total value of the current session attribute of different session categories is calculated, and the total value is used as the session resetting degree.
The first preset condition may be set based on an actual situation, such as presetting a threshold, and if the calculated total value is greater than the preset threshold, responding to a reset operation of the current session.
Therefore, in the first embodiment of the device of the present application, session information is obtained, a session state round is determined to an intelligent customer service session system end, a session reset degree is determined based on the session information, the session reset degree is determined to satisfy a first preset condition, and a reset operation of a current session is responded, so that the reset operation of the current session based on automatic response of the session information is realized, the user does not need to manually reset, the operation burden of the user is reduced, and the user experience is improved.
In a second embodiment of the present application, the memory stores a first correspondence and a second correspondence.
The first correspondence is used for representing correspondence between different session attributes and different information values in the same session category, that is, the first correspondence records information values corresponding to each session attribute in each session category.
The second correspondence is used for representing the correspondence between different session categories and different weight values, that is, the weight value corresponding to each session category is recorded in the second correspondence.
The processor executes the program, and is specifically configured to determine current session attributes of different session categories based on the session information, determine a target information value corresponding to the current session attributes based on the first correspondence, determine weight values corresponding to the different session categories based on the second correspondence, and determine a session reset value based on the weight values of the different session categories and the information value of the current session attributes.
Specifically, the session reset value may be calculated using a first calculation formula, where the first calculation formula is as follows:
Wherein R represents a session reset value, a i represents a weight value of the ith session class, p i represents an information value of the ith class, and n represents the number of session classes.
Accordingly, the processor executes the program to determine that the session reset degree meets a first preset condition, including: and determining that the session reset value is greater than or equal to a first threshold.
In a third embodiment of the apparatus of the present application, the processor executing the program is further configured to determine that the session reset value is less than a first threshold and greater than a second threshold, and prohibit a reset operation on the current session.
Wherein the first threshold is greater than the second threshold.
In a fourth embodiment of the apparatus of the present application, the memory further stores a third correspondence and a fourth correspondence.
The third correspondence is used for representing the correspondence between different session attributes and different information values in the same session category, that is, the third correspondence records information values corresponding to each session attribute in each session category.
The fourth correspondence is used for representing the correspondence between different session categories and different weight values, that is, the weight value corresponding to each session category is recorded in the fourth correspondence.
The processor executing the program determines the first threshold by:
Determining current session attributes of different session categories under a preset session threshold, determining a target information value corresponding to the current session attribute based on a third corresponding relation, determining a weight value corresponding to the different session categories based on a fourth corresponding relation, and determining the first threshold based on the weight values of the different session categories, the information value of the current session attribute and a preset first coefficient.
The intelligent customer service session system is not only divided into a plurality of session categories aiming at the session resetting degree and a plurality of session attributes under each session category in advance, but also divided into a plurality of session categories aiming at the session threshold value and a plurality of session attributes under each session category in advance.
Note that the plurality of session categories for the session reset degree are not exactly the same as the plurality of session categories for the session threshold.
The first coefficient is a preset coefficient used for calculating the first threshold value.
Specifically, the first threshold may be calculated using a second calculation formula, where the second calculation formula is as follows:
Wherein U represents a first threshold, f i represents a weight value of the ith session category, x i represents an information value of the ith category, m represents the number of session categories, and k is an upper limit coefficient. It should be noted that, E is a chaotic coefficient, and is an empirical value, and E may be added to the first threshold value as in the second calculation formula, or may be not added.
The manner of determining the second threshold may refer to the manner of determining the first threshold, and may be different from the manner of determining the first threshold only in that the value of k is different. K is a lower coefficient when calculating the second threshold. The upper limit coefficient and the lower limit coefficient are empirical values, and are usually set differently for different dialogue systems and different users. The specific intelligent session system can feed back the correction coefficient k according to the user behavior so as to meet the corresponding user.
The fifth embodiment of the application also provides an electronic device which is applied to the intelligent customer service session system, and the intelligent customer service session system can respond to the received input information to automatically provide feedback information. That is, when a user inputs information to the intelligent customer service session system, the intelligent customer service session system responds based on the input information of the user to give feedback information about the input information. It can be seen that the intelligent customer service session system can simulate a real person, so that a user can have a feeling of chatting with the real person in the intelligent customer service session system.
As shown in fig. 6, an electronic device includes: an acquisition information unit 601, a first determination unit 602, a first response unit 603; wherein:
The information obtaining unit 601 is configured to determine a session state round to the intelligent customer service session system end, and obtain session information;
a first determining unit 602, configured to determine a session reset degree based on the session information;
Wherein the first determining unit includes: a first determination module and a second determination module.
A first determining module, configured to determine current session attributes of different session categories based on the session information;
And the second determining module is used for determining the session resetting degree based on the current session attribute.
A first response unit 603, configured to determine that the session resetting degree meets a first preset condition, and respond to a resetting operation of a current session.
In a sixth embodiment of the present application, the second determining module is specifically configured to determine, based on the first correspondence, a target information value corresponding to the current session attribute, determine, based on the second correspondence, a weight value corresponding to a different session category, and determine, based on the weight value of the different session category and the information value of the current session attribute, a session reset value.
The first correspondence is used for representing correspondence between different session attributes and different information values in the same session category.
The second correspondence is used for representing the correspondence between different session categories and different weight values.
Specifically, the second determining module may calculate the session reset value using a first calculation formula, where the first calculation formula is as follows:
Wherein R represents a session reset value, a i represents a weight value of the ith session class, p i represents an information value of the ith class, and n represents the number of session classes.
Correspondingly, the first response unit is specifically configured to determine that the session reset value is greater than or equal to a first threshold, and respond to a reset operation of a current session.
In a seventh embodiment of the present application, the electronic device further includes a first prohibiting unit, configured to determine that the session reset value is smaller than a first threshold and larger than a second threshold, and prohibit a reset operation on the current session; wherein the first threshold is greater than the second threshold.
In an eighth embodiment of the present application, the electronic device further includes a threshold determining unit, configured to determine a first threshold, and specifically configured to determine current session attributes of different session categories under a preset session threshold, determine a target information value corresponding to the current session attribute based on a third correspondence, and determine a weight value corresponding to the different session categories based on a fourth correspondence; the fourth correspondence is used for representing correspondence between different session categories and different weight values, and the first threshold is determined based on the weight values of the different session categories, the information values of the current session attribute and a preset first coefficient.
The third correspondence is used for representing the correspondence between different session attributes and different information values in the same session category. The fourth correspondence is used for representing the correspondence between different session categories and different weight values.
Specifically, the threshold determining unit may calculate the first threshold using a second calculation formula, where the second calculation formula is as follows:
Wherein U represents a first threshold, f i represents a weight value of the ith session category, x i represents an information value of the ith category, m represents the number of session categories, and k is an upper limit coefficient. It should be noted that, E is a chaotic coefficient, and is an empirical value, and E may be added to the first threshold value as in the second calculation formula, or may be not added.
It should be noted that the threshold determining unit may also determine the second threshold in the above manner, and the difference from the first threshold is only that the value of k is different. K is a lower coefficient when calculating the second threshold. The upper limit coefficient and the lower limit coefficient are empirical values, and are usually set differently for different dialogue systems and different users. The specific intelligent session system can feed back the correction coefficient k according to the user behavior so as to meet the corresponding user.
In the present specification, each embodiment is described in a progressive manner, and each embodiment is mainly described in a different point from other embodiments, and identical and similar parts between the embodiments are all enough to refer to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant points refer to the description of the method section.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (9)
1. An information processing method is applied to an intelligent customer service session system, and the intelligent customer service session system can respond to received input information to automatically provide feedback information, wherein the information processing method comprises the following steps:
Determining dialogue state rounds to the intelligent customer service session system end to acquire session information;
determining a session reset degree based on the session information;
Determining that the session resetting degree meets a first preset condition, and responding to resetting operation of a current session;
wherein the determining the session reset degree based on the session information includes:
Determining current session attributes of different session categories based on the session information; the intelligent customer service session system is divided into a plurality of session categories aiming at the session resetting degree and a plurality of session attributes under each session category in advance, and the current session attributes under different session categories are determined by processing the session information; the session category comprises at least one of a current session stage, a current user behavior, a current user feedback, a current user emotion and a state confusion;
And determining a session reset degree based on the current session attribute.
2. The method of claim 1, wherein the determining a degree of session reset based on the current session attribute comprises:
determining a target information value corresponding to the current session attribute based on the first corresponding relation; the first corresponding relation is used for representing the corresponding relation between different session attributes and different information values under the same session category;
determining weight values corresponding to different session categories based on the second correspondence; the second corresponding relation is used for representing the corresponding relation between different session categories and different weight values;
a session reset value is determined based on the weight values of the different session categories and the information values of the current session attributes.
3. The method of claim 2, wherein the determining that the session reset degree satisfies a first preset condition comprises:
And determining that the session reset value is greater than or equal to a first threshold.
4. The method of claim 2, further comprising:
determining that the session reset value is smaller than a first threshold and larger than a second threshold, and prohibiting the reset operation of the current session; wherein the first threshold is greater than the second threshold.
5. A method according to claim 3, wherein the first threshold is determined by:
Determining current session attributes of different session categories under a preset session threshold;
Determining a target information value corresponding to the current session attribute based on the third corresponding relation; the third corresponding relation is used for representing the corresponding relation between different session attributes and different information values under the same session category;
Determining weight values corresponding to different session categories based on the fourth correspondence; the fourth corresponding relation is used for representing the corresponding relation between different session categories and different weight values;
And determining the first threshold value based on the weight values of different session categories, the information values of the current session attribute and a preset first coefficient.
6. An electronic device for use in an intelligent customer service session system that is capable of automatically providing feedback information in response to received input information, wherein the electronic device comprises:
A memory for storing a program;
The processor is used for executing the program, the program is used for determining dialogue state rounds to the intelligent customer service session system end, acquiring session information, determining the session resetting degree based on the session information, determining that the session resetting degree meets a first preset condition, and responding to the resetting operation of the current session;
Wherein the processor determines a session reset degree based on the session information, comprising:
Determining current session attributes of different session categories based on the session information; the intelligent customer service session system is divided into a plurality of session categories aiming at the session resetting degree and a plurality of session attributes under each session category in advance, and the current session attributes under different session categories are determined by processing the session information; the session category comprises at least one of a current session stage, a current user behavior, a current user feedback, a current user emotion and a state confusion;
And determining a session reset degree based on the current session attribute.
7. The electronic device of claim 6, wherein the memory stores a first correspondence and a second correspondence, the first correspondence being used to characterize correspondence between different session attributes and different information values in a same session category, the second correspondence being used to characterize correspondence between different session categories and different weight values;
The processor executes the program, and is specifically configured to determine a target information value corresponding to a current session attribute based on the first correspondence, determine a weight value corresponding to a different session category based on the second correspondence, and determine a session reset value based on the weight value of the different session category and the information value of the current session attribute.
8. The electronic device of claim 7, wherein the processor executing the program for determining that the session reset degree satisfies a first preset condition comprises: determining that the session reset value is greater than or equal to a first threshold;
The memory is further used for storing a third corresponding relation and a fourth corresponding relation, the third corresponding relation is used for representing the corresponding relation between different session attributes and different information values in the same session category, and the fourth corresponding relation is used for representing the corresponding relation between different session categories and different weight values;
The processor executing the program determines the first threshold by:
Determining current session attributes of different session categories under a preset session threshold, determining a target information value corresponding to the current session attribute based on a third corresponding relation, determining a weight value corresponding to the different session categories based on a fourth corresponding relation, and determining the first threshold based on the weight values of the different session categories, the information value of the current session attribute and a preset first coefficient.
9. An information processing apparatus applied to an intelligent customer service session system capable of automatically providing feedback information in response to received input information, wherein the information processing apparatus comprises:
the information acquisition unit is used for determining dialogue state rounds to the intelligent customer service session system end and acquiring session information;
a first determining unit configured to determine a session resetting degree based on the session information;
A first response unit, configured to determine that the session resetting degree meets a first preset condition, and respond to a resetting operation of a current session;
Wherein the first determining unit includes: a first determination module and a second determination module;
The first determining module is used for determining current session attributes of different session categories based on the session information; the intelligent customer service session system is divided into a plurality of session categories aiming at the session resetting degree and a plurality of session attributes under each session category in advance, and the current session attributes under different session categories are determined by processing the session information; the session category comprises at least one of a current session stage, a current user behavior, a current user feedback, a current user emotion and a state confusion;
the second determining module is configured to determine a session resetting degree based on the current session attribute.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811581667.XA CN109544195B (en) | 2018-12-24 | 2018-12-24 | Information processing method and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811581667.XA CN109544195B (en) | 2018-12-24 | 2018-12-24 | Information processing method and electronic equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109544195A CN109544195A (en) | 2019-03-29 |
| CN109544195B true CN109544195B (en) | 2024-05-28 |
Family
ID=65856834
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811581667.XA Active CN109544195B (en) | 2018-12-24 | 2018-12-24 | Information processing method and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109544195B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101436961A (en) * | 2007-11-12 | 2009-05-20 | 国际商业机器公司 | Conversation management system and method |
| CN104580226A (en) * | 2015-01-15 | 2015-04-29 | 上海瀚之友信息技术服务有限公司 | Session data sharing system and method |
| CN106685752A (en) * | 2016-06-28 | 2017-05-17 | 腾讯科技（深圳）有限公司 | Information processing method and terminal |
| CN106790171A (en) * | 2016-12-29 | 2017-05-31 | 杭州朗和科技有限公司 | The method of session control, device and computer-readable recording medium |
| CN107633042A (en) * | 2012-07-20 | 2018-01-26 | 韦韦欧股份有限公司 | The method and system of user view in search input is inferred in talking with interactive system |
-
2018
- 2018-12-24 CN CN201811581667.XA patent/CN109544195B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101436961A (en) * | 2007-11-12 | 2009-05-20 | 国际商业机器公司 | Conversation management system and method |
| CN107633042A (en) * | 2012-07-20 | 2018-01-26 | 韦韦欧股份有限公司 | The method and system of user view in search input is inferred in talking with interactive system |
| CN104580226A (en) * | 2015-01-15 | 2015-04-29 | 上海瀚之友信息技术服务有限公司 | Session data sharing system and method |
| CN106685752A (en) * | 2016-06-28 | 2017-05-17 | 腾讯科技（深圳）有限公司 | Information processing method and terminal |
| CN106790171A (en) * | 2016-12-29 | 2017-05-31 | 杭州朗和科技有限公司 | The method of session control, device and computer-readable recording medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109544195A (en) | 2019-03-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112365894B (en) | AI-based composite voice interaction method and device and computer equipment | |
| CN109522564B (en) | Voice translation method and device | |
| CN112860756A (en) | Exercise test-based learning resource recommendation method and device and computer equipment | |
| CN113053388B (en) | Voice interaction method, device, equipment and storage medium | |
| CN110706696A (en) | Voice control method and device | |
| CN112331213A (en) | Intelligent household equipment control method and device, electronic equipment and storage medium | |
| CN108932947B (en) | Voice control method and household appliance | |
| DE112018007242T5 (en) | Data processing device, data processing method, program and data processing system | |
| CN110569344A (en) | Method and device for determining standard question sentence corresponding to dialog text | |
| CN109544195B (en) | Information processing method and electronic equipment | |
| CN112151034B (en) | Voice control method and device of equipment, electronic equipment and storage medium | |
| CN112182189B (en) | Dialogue processing method and device, electronic equipment and storage medium | |
| CN111046156B (en) | Method, device and server for determining rewarding data | |
| CN114898755B (en) | Voice processing method and related device, electronic equipment and storage medium | |
| CN109726002B (en) | Processing flow adjusting method and device | |
| WO2020168444A1 (en) | Sleep prediction method and apparatus, storage medium, and electronic device | |
| CN110400576A (en) | The processing method and processing device of voice request | |
| CN113539275B (en) | Method, device and storage medium for determining speech technology | |
| CN109670030A (en) | Question and answer exchange method and device | |
| CN114328821A (en) | Multi-round conversation control method and device based on control slot position and service data slot position | |
| CN109800122B (en) | Monitoring prompting method and device, computer equipment and storage medium | |
| CN111400475A (en) | Information processing method and device and electronic equipment | |
| CN113051386A (en) | Adjustment method and device for manual service conversion, electronic equipment and storage medium | |
| CN113380241A (en) | Semantic interaction adjusting method and device, voice equipment and storage medium | |
| CN112331185A (en) | Voice interaction method, system, storage medium and electronic equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |