CN115766947A - Intelligent management and control method and system for power grid customer service center - Google Patents
Intelligent management and control method and system for power grid customer service center Download PDFInfo
- Publication number
- CN115766947A CN115766947A CN202310029432.4A CN202310029432A CN115766947A CN 115766947 A CN115766947 A CN 115766947A CN 202310029432 A CN202310029432 A CN 202310029432A CN 115766947 A CN115766947 A CN 115766947A
- Authority
- CN
- China
- Prior art keywords
- user
- users
- customer service
- incoming call
- intention
- 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.)
- Granted
Links
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Landscapes
- Telephonic Communication Services (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides an intelligent management and control method and an intelligent management and control system for a power grid customer service center, which relate to the technical field of intelligent management and control of power grids and comprise the following steps: receiving a manual service instruction sent by a user; carrying out regional classification and priority classification on the users based on the incoming call numbers, and matching and mastering the manual customer service of the corresponding language for the users; judging the incoming call intention of the user and the age bracket of the user according to the tone of the user, the speed of the user and the incoming call content of the user; classifying the intentions of the users to obtain users with problems to be solved, users with emotions to be pacified and other users; providing service suggestions for manual customer service and predicting the incoming call ending time; and adjusting a telephone traffic control strategy of the power grid customer service center based on the incoming call end time. The invention combines man-machine as user service, greatly increases the working efficiency and quality of the power grid customer service center, improves the user experience and reduces the complained rate.
Description
Technical Field
The invention relates to the technical field of intelligent management and control of a power grid, in particular to an intelligent management and control method and system for a customer service center of the power grid.
Background
The national power grid is mainly responsible for power transaction and scheduling, power supply, power grid operation and maintenance, repair and management and the like in all jurisdictions, plays an important pivotal role in a modern energy supply system and is related to the electricity utilization life of residents of thousands of households. The national power grid is particularly provided with a power grid customer service center, so that users can inquire about and solve the problem of power utilization at any time, but when the users face a series of conditions such as large-area power failure caused by rapid change of weather or power failure caused by peak power utilization load, the number of the users who inquire about the power grid customer service center is increased suddenly, and an intelligent management and control method for the power grid customer service center is urgently needed to improve the processing efficiency and the user experience.
Disclosure of Invention
The invention provides an intelligent management and control method and system for a power grid customer service center, which are used for improving the working efficiency of the power grid customer service center.
The invention provides an intelligent management and control method for a power grid customer service center, which comprises the following steps:
receiving a manual service instruction sent by a user;
based on the incoming call number, regionally classifying the users, associating the power condition for the users based on the regionally classified result, and carrying out priority classification on the users according to the power condition;
based on regional classification results and priority classification results, matching and mastering the manual customer service of the corresponding language for the user;
after the artificial customer service connects the incoming call, judging the intention of the incoming call of the user and the age bracket of the user according to the tone of the user, the speed of the user and the content of the incoming call of the user;
classifying intentions of the users based on the incoming call intentions of the users to obtain users with problems to be solved, users with emotions to be pacified and other users;
providing service suggestions for the artificial customer service and predicting the incoming call ending time based on the intention classification result and the age group of the user;
and adjusting a telephone traffic control strategy of the power grid customer service center based on the incoming call end time.
According to the intelligent management and control method for the power grid customer service center provided by the invention,
the classifying the users regionally based on the incoming call numbers, associating the power conditions for the users based on the regional classification results, and classifying the users according to the priority according to the power conditions comprises the following steps:
based on the attribution information of the incoming call number, regionally classifying the users;
acquiring the power condition of the corresponding area based on the regional classification result, and associating the power condition label to the user;
clustering the users according to the power condition labels, and dividing the users into primary users, secondary users or tertiary users, wherein the processing priority of the primary users is higher than that of the secondary users, and the processing priority of the secondary users is higher than that of the tertiary users;
and the manual customer service for matching and mastering the corresponding language for the user based on the regional classification result and the priority classification result comprises the following steps:
calibrating the processing serial number of the user according to the priority classification result;
providing Chinese artificial customer service options and the artificial customer service options for mastering the language of the attribution to the user according to the processing sequence number of the user for the user to select;
and matching and mastering the manual customer service of the corresponding language for the user based on a user selection result.
According to the intelligent management and control method for the power grid customer service center, the age bracket to which the user belongs is obtained through an age identification model according to the tone features and the speech speed features of the user, and the method comprises the following steps:
inputting the user tone features and the user speech rate features into an input layer of the age identification model;
according to the user tone features and the user speech speed features, obtaining the user age through a calculation layer of the age identification model by using a first calculation formula;
according to the age of the user, obtaining the age group of the user, and outputting through an output layer of the age identification model;
wherein the first calculation formula is:
，
the A (user) represents the age of the user, the alpha represents a first weight, the T represents a tone feature vector of the user, the beta represents a second weight, the second weight is smaller than the first weight, and the S represents a speech rate feature vector of the user.
According to the intelligent management and control method for the power grid customer service center, provided by the invention, the user incoming call intention is obtained through an intention analysis model according to the age group of the user, the content text and the user tone characteristics, and the method comprises the following steps:
performing shortest path word segmentation on the content text;
selecting a calling language network graph according to the age group to which the user belongs, and extracting keywords from the segmented content text based on the calling language network graph to obtain keyword characteristics;
inputting the keyword features and the user mood features into an input layer of the intent analysis model;
analyzing by a first analysis layer of the intention analysis model by using a multi-head attention mechanism based on the keyword characteristics to obtain a text intention;
based on the user tone characteristics, three-dimensional emotion value calculation is carried out through a second analysis layer of the intention analysis model, and a user tone type is obtained;
and according to the text intention and the user tone type, carrying out comprehensive analysis through a third analysis layer of the intention analysis model to obtain the user incoming call intention, and outputting the user incoming call intention through an output layer of the intention analysis model.
According to the intelligent management and control method for the power grid customer service center, the users are subjected to intention classification based on the incoming call intention of the users, and users with problems to be solved, users with emotions to be pacified and other users are obtained, and the method comprises the following steps:
performing relevance analysis on the text intention and a pre-stored text intention to obtain an intention relevance value;
when the intention related value is within an interval (0, 1) and the three-dimensional emotion value is within an interval (4, 9), the user is a problem-resolved user;
when the intention related value is located in an interval (0, 1) and the three-dimensional emotion value is located in an interval [1,4], the user is an emotion to be placated user;
when the intention related value is in the interval [ -1,0], the user is the other user.
According to the intelligent management and control method for the power grid customer service center, provided by the invention, service suggestions are provided for the artificial customer service based on the intention classification result and the age bracket to which the user belongs, and the incoming call ending time is predicted, and the method comprises the following steps:
when the user is a user with a problem to be solved, sending a problem solution to the artificial customer service to assist the artificial customer service user;
when the user is an emotion-to-be-pacified user, sending a jargon suggestion for soothing emotion to the manual customer service;
when the user is other users, sending a service suggestion for recording newly added questions to the manual customer service;
and sending a query of the call duration to the manual customer service at regular time, and predicting the call ending time based on the query result.
The invention also provides an intelligent management and control system for the power grid customer service center, which is characterized by comprising the following steps:
an instruction receiving module to: receiving a manual service instruction sent by a user;
the artificial customer service matching module is used for: based on the incoming call number, carrying out regional classification on the user, and matching and mastering the manual customer service of the corresponding language for the user based on the regional classification result;
an analysis module to: after the artificial customer service connects the incoming call, judging the intention of the incoming call of the user and the age bracket of the user according to the tone of the user, the speed of the user and the content of the incoming call of the user;
a user classification module to: classifying the intentions of the users based on the incoming call intentions of the users to obtain users with problems to be solved, users with emotions to be pacified and other users;
a service suggestion module to: providing service suggestions for the artificial customer service and predicting the incoming call ending time based on the intention classification result and the age group of the user;
a policy adjustment module to: and adjusting a telephone traffic control strategy of the power grid customer service center based on the incoming call end time.
The invention also provides an electronic device, which comprises a memory, a processor and a computer program which is stored on the memory and can run on the processor, wherein when the processor executes the program, the intelligent management and control method of the power grid customer service center is realized.
The present invention further provides a non-transitory computer readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the computer program implements the method for intelligently managing and controlling a power grid customer service center as described in any one of the above.
The invention also provides a computer program product, which comprises a computer program, and when the computer program is executed by a processor, the intelligent management and control method for the power grid customer service center is realized.
According to the intelligent management and control method and system for the power grid customer service center, after receiving the manual service instruction sent by the user, the users are classified regionally, then the users are classified according to the priority of the power conditions corresponding to the regions, the users needing priority processing (such as the users in the regions with unsatisfactory power conditions) can be screened out firstly, the processing efficiency is improved, then the manual customer service of the corresponding language is matched and mastered for the users based on the regional classification result and the priority classification result, the smooth communication between the users and the manual customer service is ensured, the user experience is improved, and then the user incoming call intention and the age group of the users are judged based on the user tone, the user speed and the user incoming call content, the method comprises the steps of determining the intention category of a user according to the incoming call intention of the user, obtaining a user with problems to be solved and needing priority processing, a user with emotions to be soothed and needing priority smothering, or other users with other problems, quickly providing accurate service suggestions for manual customer service according to the intention category and the belonging age bracket of the user, combining human and machine into user service, greatly increasing the effectiveness and efficiency of incoming call service, improving user experience, reducing complaint rate and avoiding sharp increase of telephone traffic.
Drawings
In order to more clearly illustrate the technical solutions of the present invention or the prior art, the drawings needed for the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and those skilled in the art can also obtain other drawings according to the drawings without creative efforts.
Fig. 1 is a schematic flow diagram of an intelligent management and control method for a power grid customer service center according to the present invention;
FIG. 2 is a schematic structural diagram of an intelligent management and control system of a power grid customer service center provided by the invention;
fig. 3 is a schematic structural diagram of an electronic device provided in the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings, and it is obvious that the described embodiments are some, but not all embodiments of the present invention. All other embodiments, which can be obtained by a person skilled in the art without inventive step based on the embodiments of the present invention, are within the scope of protection of the present invention.
The following describes an intelligent management and control method and system for a power grid customer service center according to the present invention with reference to fig. 1 to 3.
Fig. 1 is a schematic flow chart of an intelligent management and control method for a power grid customer service center. The embodiment of the application provides an intelligent management and control method for a power grid customer service center, which can comprise the following steps:
s110, receiving a manual service instruction sent by a user;
s120, performing regional classification on users based on incoming call numbers, associating power conditions for the users based on regional classification results, and performing priority classification on the users according to the power conditions;
s130, based on the regional classification result and the priority classification result, matching and mastering the manual customer service of the corresponding language for the user;
s140, after the manual customer service connects the incoming call, judging the incoming call intention of the user and the age bracket of the user according to the tone of the user, the speed of the user and the content of the user;
s150, classifying intentions of the users based on the incoming call intentions of the users to obtain users with problems to be solved, users with emotions to be pacified and other users;
s160, providing service suggestions for the artificial customer service based on the intention classification result and the age group of the user, and predicting the incoming call ending time;
and S170, adjusting a traffic control strategy of the power grid customer service center based on the incoming call end time.
It should be noted that an execution main body of the intelligent management and control method for the power grid customer service center provided in the embodiment of the present application may be any network side device, for example, the power grid customer service center, the incoming call service system, and the like.
In step S110, the network side device receives a manual service instruction from a user.
It should be noted that, when a user calls the power grid customer service center, the network side device may first ask the user through an intelligent customer service (robot) whether to need manual customer service, and when the user needs manual customer service, the network side device may send a manual service instruction to the network side device, and the manual service instruction is received and processed by the network side device. For example, the intelligent customer service will confirm to the user whether the manual customer service is needed, and if so, press the number "1", and then the user can issue the manual service instruction by pressing the number "1".
Therefore, the intention of the user can be confirmed very conveniently, some users have social terrorism and do not like to communicate with real people, and the users can solve the problem through the intelligent customer service, but some users tend to directly communicate with the artificial customer service, so that the problem can be solved more quickly, and the difficulty in understanding the semantics by the intelligent customer service and the mechanical reply of the semantics by the intelligent customer service are avoided.
Further, the network side device may preset a blacklist number, that is, record all the historical crank call numbers in the blacklist, otherwise, the incoming call number that is not in the blacklist is a white list number, and if it is found that a certain incoming call number is also a crank call subsequently, the incoming call number may be updated to the blacklist.
In the invention, the network side equipment judges whether the incoming call number of the user is the number in the white list or not, so that the incoming call disturbance of the blacklist user can be avoided, the working effectiveness of the power grid customer service center is ensured, more time is saved for serving the user really needing the service, and the service efficiency is improved.
In step 120, the network side device may classify the user regionally based on the incoming call number, associate the power condition for the user based on the regional classification result, and classify the user according to the priority of the power condition.
It should be noted that the network side device may perform approximate regional classification on the user according to the incoming call number of the user before the user call is connected, and then obtain the power condition of the corresponding region according to the regional classification result, where the power condition may be, for example, normal power supply, peak power supply, abnormal power supply due to device failure, abnormal power supply due to weather, and the like, and then associate the power condition with the user, which is beneficial for the manual customer service to know the approximate condition of the user in advance, so as to better serve the user, and the network side device may perform priority classification on the user according to the quality of the power condition, so that the user call can be handled orderly and efficiently when a large number of user calls are encountered, thereby improving the service efficiency and quality.
In one embodiment, step S120 may include:
s1201, based on the attribution information of the incoming call number, regionally classifying the users;
s1202, acquiring the power condition of the corresponding region based on the regional classification result, and associating the power condition label to the user;
s1203, clustering the users according to the power condition labels, and dividing the users into first-level users, second-level users or third-level users, wherein the processing priority of the first-level users is prior to the second-level users, and the processing priority of the second-level users is prior to the third-level users.
It should be noted that the incoming call number generally has information such as a home location, a country code, a type of a mobile phone card, and a telephone area code, and the network side device may directly classify the user regionally through the home location information of the incoming call number, or may classify the user regionally through the telephone area code, for example, "020" is an area code of guangzhou city, and then the language corresponding to guangzhou city is cantonese.
If the regional classification of the user is Guangzhou, the network side device may search for the electricity utilization situation corresponding to Guangzhou, and also search for more detailed areas where abnormal electricity utilization situations exist in the Guangzhou area, such as a seapearl area and a litchi bay area, and the network side device may associate the power situation labels of "Guangzhou-seapearl area-abnormal power supply due to weather" and "Guangzhou-litchi bay area-abnormal power supply due to weather" with the user.
The network side equipment can cluster the users according to the labels of the power conditions through a K-Means (K mean value) clustering algorithm, a mean shift clustering algorithm or a hierarchical clustering algorithm, cluster the users adjacent to the region and similar to the power conditions to form a population, and process priority marks on the population according to the severity of the power conditions, for example, the population with abnormal power utilization conditions and the number of users exceeding a first preset threshold value L is marked as a first-level user, the population with abnormal power utilization conditions and the number of users not reaching L is marked as a second-level user, and the population with normal power utilization conditions is marked as a third-level user. Thus, the user can be efficiently and orderly served, and the user who really needs the service can be quickly responded.
In step S130, the network side device matches and grasps the manual customer service of the corresponding language for the user based on the regional classification result and the priority classification result.
It should be noted that, at present, most users purchase telephone numbers in a real-name system, and generally purchase the users at their own places, the network side device classifies the users regionally based on the incoming call numbers, and matches and masters the manual customer service in the corresponding language for the users orderly based on the regional classification result and the priority classification result, so that the problems that the users cannot communicate with the manual customer service due to language barriers and need to be redistributed can be avoided, and the language adopted by the manual customer service and the language adopted by the users can be unified, so that the users can feel intimate, the comfort level of the users is greatly improved, the understanding of the meaning of the users by the manual customer service is facilitated, and the working efficiency is improved.
In one embodiment, step S130 may include:
s1301, calibrating the processing sequence number of the user according to the priority classification result;
s1302, providing Chinese artificial customer service options and the artificial customer service options for mastering the language of the attribution to the user according to the processing sequence number of the user for the user to select;
and S1303, matching and mastering the manual customer service of the corresponding language for the user based on the selection result of the user.
It should be noted that, the network side device performs user clustering in step S120 to implement classification of processing priorities, and further, in each population, ranks the users in the population again to make clear the order of manual customer service for processing the users, where the ranking condition may be the incoming call duration of the user, and the longer the incoming call duration is, the user with the incoming call duration may be ranked in the front position, so as to avoid the risk of complaints caused by the user with the incoming call duration.
It should be noted that the incoming call number generally has information such as a home location, a country code, a mobile phone card type, a telephone area code, etc., the network side device can directly classify the user regionally through the home location information of the incoming call number, or can classify the user regionally through the telephone area code, for example, "020" is an area code of guangzhou city in Guangdong province, the language corresponding to the guangzhou city is cantonese, the network side device can provide the user with a Chinese artificial customer service option and an artificial customer service option for mastering cantonese, the user can select the artificial customer service, the Chinese artificial customer service is provided because Chinese is an unfamiliar language, the situation that some users know the local language is avoided, smooth communication between the user and the artificial customer service can be further guaranteed, a freely selected right is provided for the user, after the user selects, the network side device can rapidly match the artificial customer service of the corresponding language for the user, if the user does not select in one minute, the network side device can select the default artificial customer service option for Chinese, the manual service option for matching for the user, and the number of the subsequent incoming calls is avoided, which is too many.
In step S140, after the manual customer service connects the incoming call, the network side device may determine the user ' S incoming call intention and the age of the user according to the user ' S tone, the user ' S voice tone, the user ' S speed, and the user ' S incoming call content.
It should be noted that the network-side device may obtain the user tone (which may be obtained by detecting the tone word and volume intensity of the user speaking), the user tone (which may be obtained by detecting the sound wave level of the user speaking), the user speech rate (which may be obtained by calculating the time interval between words), and the user incoming call content (which may be obtained by converting speech into text) according to the speech of the user speaking, and may also determine the user tone (for example, a statement tone, a pragmatic tone, an exclamatory tone, a query tone), the user tone (for example, high, medium, low), and the user speech rate (fast, medium, slow) through artificial customer service.
The network side equipment can know the emotion, attitude and problems to be solved of the user through the tone of the user, the speed of the user and the incoming call content of the user so as to accurately judge the incoming call intention of the user and the age group of the user. The user belonging age group can comprise a teenage age group (13-17 years), a young age group (18-45 years), a middle age group (46-68 years) and an old age group (69 years and above), the specific ages can be adjusted according to actual conditions, the network side equipment obtains the user belonging age group, and the network side equipment can be helpful for manual customer service users, for example, the user in the old age group needs to be patience more, speak slowly and express clearly, so that the user can be really helped.
In one embodiment, step S140 may include:
s1401, after the manual customer service connects the incoming call, obtaining user tone characteristics, user speed characteristics and user incoming call contents in a preset time period (for example, 30 seconds, 45 seconds, 60 seconds or 120 seconds after the incoming call is connected);
s1402, obtaining the age bracket of the user through an age identification model according to the tone features of the user and the speech speed features of the user, wherein the age identification model is obtained through training according to voice sample data and voice sample label data;
s1403, converting the content of the user call to a content text (a voice for the content of the call can be converted to the content text by using a suitable existing technology);
and S1404, obtaining the incoming call intention of the user through an intention analysis model according to the age group of the user, the content text and the tone characteristics of the user, wherein the intention analysis model is obtained through training according to user intention sample data and user intention sample label data.
Specifically, the network side device may implement step S1402 by the following steps: inputting the tone features and the speech speed features of the user into an input layer of the age identification model; obtaining the age of the user by utilizing a first calculation formula through a calculation layer of an age identification model according to the tone features of the user and the speech speed features of the user; and obtaining the age bracket to which the user belongs according to the age of the user, and outputting through an output layer of the age identification model.
Wherein the first calculation formula is:
，
said a (user) represents the user ' S age, said α represents a first weight, preferably 0.75, said T represents a user ' S intonation feature vector, said β represents a second weight, preferably 0.25, said S represents a user ' S speech rate feature vector, the values of the first weight and the second weight can be set according to actual conditions, but the setting rule is that the second weight is smaller than the first weight because the user ' S intonation has a greater influence than the user ' S speech rate for age determination.
It should be noted that a (user) obtained by the network side device through the first calculation formula is a numerical value converted by a result vector obtained by adding the user intonation feature vector and the user speech rate feature vector (for example, obtained by calculating a value of a result vector coordinate, or assuming that the result vector is (a, B, \8230; Z), a calculation formula of the value is
The obtained F value is a decimal with one decimal, that is, a (user)), which is not a true value representing the age, so after a (user) is obtained, it is also necessary to compare a (user)' of the age group to which the user belongs, so that the age group to which the user belongs can be obtained. For example, the A (user)' of the adolescent age group is [0.5,0.7]The A (user)' of young age group is (0.7, 1.2)]The a (user) ' of the middle age group is greater than 1.2, the a (user) ' of the old age group is less than 0.5, the a (user) ' of the age group to which the user belongs may also be specifically set according to actual situations, and if the a (user) obtained by the network side device is 0.8, it indicates that the age group to which the user belongs is the young age group.
More specifically, the network side device may implement step S1404 by: performing shortest path word segmentation on the content text through a shortest path word segmentation device based on a HanLP (natural language processing toolkit formed by a series of models and algorithms) technology; according to the age group to which the user belongs, selecting a calling language network graph (the network side equipment can set the calling language network graph corresponding to calling language data through cytoscape software (software focusing on open source network visualization and analysis) for different age groups in advance, because the same sentence possibly has different meanings for users in different age groups to ensure the accuracy of subsequent intention analysis), and extracting accurate keywords by using a word frequency-inverse document frequency technology based on the content text after word segmentation of the calling language network graph to obtain keyword characteristics; inputting the keyword characteristics and the user tone characteristics into an input layer of the intention analysis model; analyzing by using a multi-head attention mechanism through a first analysis layer of an intention analysis model based on the characteristics of the keywords to obtain text intentions; based on the tone characteristics of the user, performing three-dimensional emotion value calculation by using a VAP (value-area-Power, arousal-joy one control degree) three-dimensional emotion theory, a PAD (plus-display joy, area-not activation, dominance domino-submissiveness) emotion three-dimensional theory or a PAD emotion scale through a second analysis layer of the intention analysis model to obtain a tone type of the user, wherein the tone type of the user comprises happiness, anger, surprise, sadness, fear, disgust, neutrality and the like; and according to the text intention and the tone type of the user, carrying out comprehensive analysis through a third analysis layer of the intention analysis model to obtain the user incoming call intention, and outputting through an output layer of the intention analysis model.
It should be noted that, the network side device may obtain the user mood type through the VAP three-dimensional emotion model. The VAP three-dimensional emotion model analyzes the emotion of a user through three dimensions of arousal degree, pleasure degree and control degree, wherein the arousal degree represents the height of the arousal degree of the emotion, the pleasure degree represents the height of the positive emotion, the control degree represents the control state of an individual on a scene and other people, and the three dimensions can represent the height of the individual through numerical values. For example, the value interval [1,9], 1 represents very low addiction/negative/low control desire, 9 represents very high agonism/positive/high control desire, three-dimensional emotion values can be obtained by adding the values of the three dimensions, and then the corresponding user tone type is output, for example, the value interval [4,6] represents a neutral user tone type. Alternatively, the network-side device may obtain the user mood type by externally connecting a second analysis layer of the intention analysis model to an appropriate auditory model (e.g., a ZCPA model) in the prior art.
When the network side equipment obtains a text intention and a user tone type, the text intention is associated with a three-dimensional emotion value, if the user tone type is anger, sadness and disgust, a label of a first-level priority service is marked on the text intention of the user, if the user tone type is surprise and fear, a label of a second-level priority service is marked on the text intention of the user, if the user tone type is happy and neutral, a label of a third-level priority service is marked on the text intention of the user to obtain a user call intention, and the sequence of serving the user can be arranged according to the user call intention subsequently.
In step S150, the network side device may classify the user intentions based on the user incoming call intentions to obtain a user with a problem to be solved, a user with an emotion to be soothed, and other users.
Specifically, the pre-stored text intention may be preset by the network side device based on historical data, the network side device may perform correlation analysis on the text intention and the pre-stored text intention to obtain an intention correlation value, for example, calculate a covariance or a covariance matrix of the text intention and the pre-stored text intention, and if the variation trends of the two variables are consistent, the covariance is a positive value, which indicates that the two variables are positively correlated; if the variation trends of the two variables are opposite, the covariance is a negative value, and the two variables are in negative correlation; if the two variables are independent of each other, then the covariance is 0, indicating that the two variables are uncorrelated. When the intention-related value is within the interval (0, 1) and the three-dimensional emotion value is within the interval (4, 9), the user is a user with a problem to be solved, when the intention-related value is within the interval (0, 1) and the three-dimensional emotion value is within the interval [1,4], the user is a user with an emotion to be soothed, and when the intention-related value is within the interval [ -1,0], the user is another user.
The method and the system have the advantages that the intention classification is carried out on the users based on the incoming call intention of the users, the users with problems to be solved and needing to be processed preferentially, the users with emotions needing to be soothed preferentially or other users with other problems are obtained, the method and the system are beneficial to providing accurate service suggestions for manual customer service according to the intention categories and the age groups of the users, and the working efficiency, quality and user experience of the manual customer service are improved.
In step S160, the network side device provides a service suggestion for the manual customer service based on the intention classification result and the age group to which the user belongs, and predicts an incoming call ending time.
Specifically, when the user is a user with a problem to be solved, the network side device may send a problem solution to the manual customer service to assist the manual customer service user, where the specific problem solution may be a solution obtained by matching the network side device with the knowledge base according to the user's intention of incoming call; when the user is a mood-to-be-pacified user, the network side device can send a jargon suggestion for mood pacifying to the manual customer service, for example, when the mood type of the user is sad, the network side device can send some jokes or small stories for exciting people to the manual customer service; when the user is other user, the network side device can send the service suggestion for recording the new question to the manual customer service, so as to update the knowledge base subsequently.
Furthermore, the network side equipment can also send out a query of the time length of the call to be called to the manual customer service at regular time, and the call ending time can be predicted based on the query result.
It should be noted that when the busy call exceeds three minutes, the network side device may send an inquiry about the time length of the call required to be made to the manual customer service every two minutes, for example, send "a" call required to be made for one minute, b "call required to be made for two minutes, c" call required to be made for more than two minutes, d "call is about to be ended", and after the manual customer service replies, the network side device may calculate the incoming call end time according to the option of the manual customer service, which is helpful for the network side device to adjust the traffic control policy, and avoids the traffic management imbalance and resource waste.
In step S170, the network side device may adjust a traffic volume management and control policy of the power grid customer service center based on the incoming call ending time.
It should be noted that the traffic control policy of the power grid customer service center includes an extensibility control and a protective control, the extensibility control is to use a resource with a relatively light load in the network for traffic suffering from congestion, and the protective control is to cancel the traffic with a relatively low call completing rate when the network is congested so as to prevent congestion from spreading and avoid the influence of part of the traffic on the whole network. The network side equipment can adjust and regulate the traffic control strategy of the power grid customer service center based on the incoming call end time of all the incoming calls of the users in the network, such as adjusting and shunting, increasing an emergency echelon and the like, so that the network side equipment is favorable for fully utilizing network resources, avoiding the blockage of the communication network and ensuring the long-term and stable operation of the communication network.
According to the intelligent management and control method and system for the power grid customer service center, after receiving the manual service instruction sent by the user, the users are classified regionally, then the users are classified according to the priority of the power conditions corresponding to the regions, the users needing priority processing (such as the users in the regions with unsatisfactory power conditions) can be screened out firstly, the processing efficiency is improved, then the manual customer service of the corresponding language is matched and mastered for the users based on the regional classification result and the priority classification result, the smooth communication between the users and the manual customer service is ensured, the user experience is improved, and then the user incoming call intention and the age group of the users are judged based on the user tone, the user speed and the user incoming call content, the method comprises the steps of determining the intention category of a user according to the incoming call intention of the user, obtaining a user with problems to be solved and needing to be treated preferentially, a user with emotion to be appealed and needing to appeals preferentially, or other users with other problems, and rapidly providing accurate service suggestions for artificial customer service according to the intention category and the belonging age bracket of the user, namely combining a man machine with the user service, so that the service effectiveness and efficiency of a power grid customer service center can be greatly increased, the user experience is improved, the complaint rate is reduced, and the telephone traffic is prevented from being increased sharply.
The following describes the intelligent management and control system for the power grid customer service center provided by the present invention, and the intelligent management and control system for the power grid customer service center described below and the intelligent management and control method for the power grid customer service center described above can be referred to in a corresponding manner.
Referring to fig. 2, the intelligent management and control system for a power grid customer service center provided by the invention comprises:
an instruction receiving module 310, configured to: receiving a manual service instruction sent by a user;
a first user classification module 320 to: based on the incoming call number, regionally classifying the users, associating the power condition for the users based on the regionally classified result, and carrying out priority classification on the users according to the power condition;
an artificial customer service matching module 330, configured to: matching and mastering manual customer service of corresponding languages for the user based on the regional classification result and the priority classification result;
an analysis module 340 for: after the manual customer service connects the incoming call, judging the incoming call intention of the user and the age bracket of the user according to the tone of the user, the speed of the user and the content of the user;
a second user classification module 350 for: classifying the intentions of the users based on the incoming call intentions of the users to obtain users with problems to be solved, users with emotions to be pacified and other users;
a service suggestion module 360 to: providing service suggestions for the artificial customer service and predicting the incoming call ending time based on the intention classification result and the age group of the user;
a policy adjustment module 370 for: and adjusting a telephone traffic control strategy of the power grid customer service center based on the incoming call end time.
Specifically, the first user classification module 320 includes:
a regional classification submodule to: based on the attribution information of the incoming call number, regionally classifying the users;
a power condition acquisition sub-module to: acquiring the power condition of the corresponding area based on the regional classification result, and associating the power condition label to the user;
a clustering submodule for: and clustering the users according to the power condition labels, and dividing the users into primary users, secondary users or tertiary users, wherein the processing priority of the primary users is higher than that of the secondary users, and the processing priority of the secondary users is higher than that of the tertiary users.
Specifically, the artificial customer service matching module 330 includes:
a sorting submodule for: calibrating the processing serial number of the user according to the priority classification result;
a manual customer service selection submodule for: according to the processing sequence number of the user, providing the Chinese artificial customer service option and the artificial customer service option for mastering the language of the attribution place to the user for the user to select;
the artificial customer service matching submodule is used for: and matching and mastering the manual customer service of the corresponding language for the user based on a user selection result.
Specifically, the analysis module 340 includes:
an information acquisition submodule for: after the manual customer service connects the incoming call, obtaining user tone characteristics, user speech speed characteristics and user incoming call content in a preset time period;
an age identification submodule to: obtaining the age group to which the user belongs through an age identification model according to the tone features of the user and the speech speed features of the user, wherein the age group to which the user belongs comprises a teenager age group, a youth age group, a middle age group and an old age group, and the age identification model is obtained by training according to voice sample data and voice sample label data;
a text conversion sub-module to: converting the content of the incoming call of the user into a content text;
an intent analysis submodule to: and obtaining the user incoming call intention through an intention analysis model according to the age group of the user, the content text and the user tone characteristics, wherein the intention analysis model is obtained through training according to user intention sample data and user intention sample label data.
Specifically, the age identifier module comprises:
a first input submodule for: inputting the user tone features and the user speech rate features into an input layer of the age identification model;
a computation submodule for: according to the user tone features and the user speech speed features, obtaining the user age through a calculation layer of the age identification model by using a first calculation formula;
a first output sub-module to: according to the age of the user, obtaining the age group of the user, and outputting through an output layer of the age identification model;
wherein the first calculation formula is:
，
a (user) represents the age of the user, alpha represents a first weight, T represents a tone feature vector of the user, beta represents a second weight, the second weight is smaller than the first weight, and S represents a speech rate feature vector of the user.
Specifically, the intention analysis submodule includes:
a word segmentation submodule for: performing shortest path word segmentation on the content text;
a keyword extraction submodule for: selecting a calling language network graph according to the age group to which the user belongs, and extracting keywords from the segmented content text based on the calling language network graph to obtain keyword characteristics;
a second input submodule for: inputting the keyword features and the user tone features into an input layer of the intention analysis model;
a first analysis submodule for: analyzing by using a multi-head attention mechanism through a first analysis layer of the intention analysis model based on the keyword features to obtain a text intention;
a second analysis submodule for: based on the user tone characteristics, performing three-dimensional emotion value calculation through a second analysis layer of the intention analysis model to obtain user tone types, wherein the user tone categories comprise awakening, joyful and control;
a third analysis submodule to: and according to the text intention and the user tone type, carrying out comprehensive analysis through a third analysis layer of the intention analysis model to obtain the user incoming call intention, and outputting through an output layer of the intention analysis model.
Specifically, the user classifying module 350 includes:
a correlation analysis submodule to: performing relevance analysis on the text intention and a pre-stored text intention to obtain an intention relevance value;
a first classification submodule to: when the intention related value is within an interval (0, 1) and the three-dimensional emotion value is within an interval (4, 9), the user is a problem-resolved user;
a second classification submodule to: when the intention-related value is within an interval (0, 1) and the three-dimensional emotion value is within an interval [1,4], the user is an emotion-to-be-placated user;
a third classification submodule to: when the intention related value is in the interval [ -1,0], the user is the other user.
Specifically, the service suggestion module 360 includes:
a first suggestion submodule to: when the user is a user with a problem to be solved, sending a problem solution to the artificial customer service to assist the artificial customer service user;
a second suggestion submodule to: when the user is an emotion-to-be-pacified user, sending a jargon suggestion for soothing emotion to the manual customer service;
a third suggestion submodule to: when the user is other users, sending a service suggestion for recording newly added questions to the manual customer service;
a prediction sub-module to: and sending a query of the time length of the call to be needed to the manual customer service at regular time, and predicting the ending time of the incoming call based on the query result.
Fig. 3 illustrates a physical structure diagram of an electronic device, which may include, as shown in fig. 3: a processor (processor) 810, a communication interface 820, a memory 830 and a communication bus 840, wherein the processor 810, the communication interface 820 and the memory 830 communicate with each other via the communication bus 840. The processor 810 may invoke the logic instructions in the memory 830 to perform a power grid customer service center intelligent management and control method, the method including:
receiving a manual service instruction sent by a user;
based on the incoming call number, regionally classifying the users, associating the power condition for the users based on the regionally classified result, and carrying out priority classification on the users according to the power condition;
based on regional classification results and priority classification results, matching and mastering the manual customer service of the corresponding language for the user;
after the manual customer service connects the incoming call, judging the incoming call intention of the user and the age bracket of the user according to the tone of the user, the speed of the user and the content of the user;
classifying the intentions of the users based on the incoming call intentions of the users to obtain users with problems to be solved, users with emotions to be pacified and other users;
providing service suggestions for the artificial customer service and predicting the incoming call ending time based on the intention classification result and the age group of the user;
and adjusting a traffic control strategy of the power grid customer service center based on the incoming call ending time.
In addition, the logic instructions in the memory 830 may be implemented in software functional units and stored in a computer readable storage medium when the logic instructions are sold or used as independent products. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-only memory (ROM), a Random Access Memory (RAM), a magnetic disk, an optical disk, or other various media capable of storing program codes.
In another aspect, the present invention further provides a computer program product, where the computer program product includes a computer program, where the computer program may be stored on a non-transitory computer-readable storage medium, and when the computer program is executed by a processor, the computer is capable of executing the method for intelligently managing and controlling a power grid customer service center provided by the foregoing methods, where the method includes:
receiving a manual service instruction sent by a user;
based on the incoming call number, regionally classifying the users, associating the power condition for the users based on the regionally classified result, and carrying out priority classification on the users according to the power condition;
matching and mastering manual customer service of corresponding languages for the user based on the regional classification result and the priority classification result;
after the manual customer service connects the incoming call, judging the incoming call intention of the user and the age bracket of the user according to the tone of the user, the speed of the user and the content of the user;
classifying the intentions of the users based on the incoming call intentions of the users to obtain users with problems to be solved, users with emotions to be pacified and other users;
providing service suggestions for the artificial customer service and predicting the incoming call ending time based on the intention classification result and the age group of the user;
and adjusting a traffic control strategy of the power grid customer service center based on the incoming call ending time.
In yet another aspect, the present invention also provides a non-transitory computer-readable storage medium, on which a computer program is stored, the computer program being implemented by a processor to execute the method for intelligently managing and controlling a power grid customer service center provided by the above methods, the method including:
receiving a manual service instruction sent by a user;
based on the incoming call number, regionally classifying the users, associating the power condition for the users based on the regionally classified result, and carrying out priority classification on the users according to the power condition;
matching and mastering manual customer service of corresponding languages for the user based on the regional classification result and the priority classification result;
after the manual customer service connects the incoming call, judging the incoming call intention of the user and the age bracket of the user according to the tone of the user, the speed of the user and the content of the user;
classifying the intentions of the users based on the incoming call intentions of the users to obtain users with problems to be solved, users with emotions to be pacified and other users;
providing service suggestions for the artificial customer service and predicting the incoming call ending time based on the intention classification result and the age bracket to which the user belongs;
and adjusting a traffic control strategy of the power grid customer service center based on the incoming call ending time.
The above-described embodiments of the apparatus are merely illustrative, and the units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one position, or may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment. One of ordinary skill in the art can understand and implement it without inventive effort.
Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware. Based on the understanding, the above technical solutions substantially or otherwise contributing to the prior art may be embodied in the form of a software product, which may be stored in a computer-readable storage medium, such as ROM/RAM, magnetic disk, optical disk, etc., and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the method according to the various embodiments or some parts of the embodiments.
Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (10)
1. An intelligent management and control method for a power grid customer service center is characterized by comprising the following steps:
receiving a manual service instruction sent by a user;
based on the incoming call number, regionally classifying the users, associating the power condition for the users based on the regionally classified result, and carrying out priority classification on the users according to the power condition;
based on regional classification results and priority classification results, matching and mastering the manual customer service of the corresponding language for the user;
after the manual customer service connects the incoming call, judging the incoming call intention of the user and the age bracket of the user according to the tone of the user, the speed of the user and the content of the user;
classifying intentions of the users based on the incoming call intentions of the users to obtain users with problems to be solved, users with emotions to be pacified and other users;
providing service suggestions for the artificial customer service and predicting the incoming call ending time based on the intention classification result and the age group of the user;
and adjusting a telephone traffic control strategy of the power grid customer service center based on the incoming call end time.
2. The intelligent management and control method for the power grid customer service center according to claim 1, wherein the classifying users regionally based on incoming call numbers, associating power conditions with the users based on regional classification results, and classifying the users according to priority according to the power conditions comprises:
based on the attribution information of the incoming call number, regionally classifying the users;
acquiring the power situation of the corresponding area based on the regional classification result, and associating the power situation label to the user;
clustering the users according to the power condition labels, and dividing the users into primary users, secondary users or tertiary users, wherein the processing priority of the primary users is prior to the secondary users, and the processing priority of the secondary users is prior to the tertiary users;
and the manual customer service for matching and mastering the corresponding language for the user based on the regional classification result and the priority classification result comprises the following steps:
calibrating the processing serial number of the user according to the priority classification result;
providing Chinese artificial customer service options and the artificial customer service options for mastering the language of the attribution to the user according to the processing sequence number of the user for the user to select;
and matching and mastering the manual customer service of the corresponding language for the user based on a user selection result.
3. The intelligent management and control method for the power grid customer service center according to claim 2, wherein after the manual customer service is connected to the incoming call, the user's incoming call intention and the age group of the user are judged according to the user's tone, the user's voice tone, the user's speed of speech, and the user's incoming call content, and the method comprises the following steps:
after the manual customer service connects the incoming call, obtaining user tone characteristics, user speech speed characteristics and user incoming call content in a preset time period;
obtaining the age group to which the user belongs through an age identification model according to the tone features of the user and the speech speed features of the user, wherein the age group to which the user belongs comprises a teenager age group, a youth age group, a middle age group and an old age group, and the age identification model is obtained by training according to voice sample data and voice sample label data;
converting the content of the incoming call of the user into a content text;
and obtaining the incoming call intention of the user through an intention analysis model according to the age group of the user, the content text and the tone characteristics of the user, wherein the intention analysis model is obtained through training according to user intention sample data and user intention sample label data.
4. The intelligent management and control method for the power grid customer service center according to claim 3, wherein the step of obtaining the age bracket of the user through an age identification model according to the tone features of the user and the speech speed features of the user comprises the following steps:
inputting the user tone features and the user speech rate features into an input layer of the age identification model;
according to the user tone features and the user speech speed features, obtaining the user age through a calculation layer of the age identification model by using a first calculation formula;
according to the age of the user, obtaining the age group of the user, and outputting through an output layer of the age identification model;
wherein the first calculation formula is:
，
a (user) represents the age of the user, alpha represents a first weight, T represents a tone feature vector of the user, beta represents a second weight, the second weight is smaller than the first weight, and S represents a speech rate feature vector of the user.
5. The intelligent management and control method for the power grid customer service center according to claim 3, wherein the step of obtaining the incoming call intention of the user through an intention analysis model according to the age group of the user, the content text and the user tone characteristics comprises the steps of:
performing shortest path word segmentation on the content text;
selecting a calling language network graph according to the age group to which the user belongs, and extracting keywords from the segmented content text based on the calling language network graph to obtain keyword characteristics;
inputting the keyword features and the user tone features into an input layer of the intention analysis model;
analyzing by a first analysis layer of the intention analysis model by using a multi-head attention mechanism based on the keyword characteristics to obtain a text intention;
based on the user tone characteristics, three-dimensional emotion value calculation is carried out through a second analysis layer of the intention analysis model, and a user tone type is obtained;
and according to the text intention and the user tone type, carrying out comprehensive analysis through a third analysis layer of the intention analysis model to obtain the user incoming call intention, and outputting through an output layer of the intention analysis model.
6. The intelligent management and control method for the power grid customer service center according to claim 5, wherein the intention classification is performed on the users based on the incoming call intention of the users to obtain users with problems to be solved, users with emotions to be soothed, and other users, and comprises the following steps:
performing relevance analysis on the text intention and a pre-stored text intention to obtain an intention relevance value;
when the intention-related value is within an interval (0, 1) and the three-dimensional emotion value is within an interval (4, 9), the user is a problem-resolved user;
when the intention-related value is within an interval (0, 1) and the three-dimensional emotion value is within an interval [1,4], the user is an emotion-to-be-placated user;
when the intention-related value is located in the interval [ -1,0], the user is the other user.
7. The intelligent management and control method for the power grid customer service center according to claim 6, wherein the providing service suggestions for the artificial customer service and predicting the incoming call ending time based on the intention classification result and the age group of the user comprises:
when the user is a user with a problem to be solved, sending a problem solution to the artificial customer service to assist the artificial customer service user;
when the user is an emotion to-be-placated user, sending a speaking suggestion for mood placation to the manual customer service;
when the user is other users, sending a service suggestion for recording newly added questions to the manual customer service;
and sending a query of the time length of the call to be needed to the manual customer service at regular time, and predicting the ending time of the incoming call based on the query result.
8. The utility model provides a power grid customer service center intelligence management and control system which characterized in that includes:
an instruction receiving module to: receiving a manual service instruction sent by a user;
a first user classification module to: based on the incoming call number, regionally classifying the users, associating the power condition for the users based on the regionally classified result, and carrying out priority classification on the users according to the power condition;
the artificial customer service matching module is used for: matching and mastering manual customer service of corresponding languages for the user based on the regional classification result and the priority classification result;
an analysis module to: after the artificial customer service connects the incoming call, judging the intention of the incoming call of the user and the age bracket of the user according to the tone of the user, the speed of the user and the content of the incoming call of the user;
a second user classification module to: classifying the intentions of the users based on the incoming call intentions of the users to obtain users with problems to be solved, users with emotions to be pacified and other users;
a service suggestion module to: providing service suggestions for the artificial customer service and predicting the incoming call ending time based on the intention classification result and the age group of the user;
a policy adjustment module to: and adjusting a telephone traffic control strategy of the power grid customer service center based on the incoming call end time.
9. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the method for intelligently managing a power grid customer service center according to any one of claims 1 to 7 when executing the program.
10. A non-transitory computer-readable storage medium, on which a computer program is stored, wherein the computer program, when executed by a processor, implements the method for intelligently managing a power grid customer service center according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310029432.4A CN115766947B (en) | 2023-01-09 | 2023-01-09 | Intelligent management and control method and system for power grid customer service center |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310029432.4A CN115766947B (en) | 2023-01-09 | 2023-01-09 | Intelligent management and control method and system for power grid customer service center |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115766947A true CN115766947A (en) | 2023-03-07 |
| CN115766947B CN115766947B (en) | 2023-05-09 |
Family
ID=85348767
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310029432.4A Active CN115766947B (en) | 2023-01-09 | 2023-01-09 | Intelligent management and control method and system for power grid customer service center |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115766947B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116737914A (en) * | 2023-08-15 | 2023-09-12 | 南通话时代信息科技有限公司 | Interactive data processing method and system for cloud customer service center |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107657017A (en) * | 2017-09-26 | 2018-02-02 | 百度在线网络技术（北京）有限公司 | Method and apparatus for providing voice service |
| CN111739516A (en) * | 2020-06-19 | 2020-10-02 | 中国—东盟信息港股份有限公司 | Speech recognition system for intelligent customer service call |
| CN112185385A (en) * | 2020-09-30 | 2021-01-05 | 深圳供电局有限公司 | Intelligent client processing method and system for power supply field |
| CN112600982A (en) * | 2020-12-08 | 2021-04-02 | 深圳供电局有限公司 | Power supply service hotline interactive voice response method, system, equipment and medium |
| CN114492391A (en) * | 2021-12-28 | 2022-05-13 | 航天信息股份有限公司 | Intention reasoning method and device |
| CN114942973A (en) * | 2022-04-19 | 2022-08-26 | 尚特杰电力科技有限公司 | Emotion recognition method and system for electric power intelligent customer service system |
-
2023
- 2023-01-09 CN CN202310029432.4A patent/CN115766947B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107657017A (en) * | 2017-09-26 | 2018-02-02 | 百度在线网络技术（北京）有限公司 | Method and apparatus for providing voice service |
| CN111739516A (en) * | 2020-06-19 | 2020-10-02 | 中国—东盟信息港股份有限公司 | Speech recognition system for intelligent customer service call |
| CN112185385A (en) * | 2020-09-30 | 2021-01-05 | 深圳供电局有限公司 | Intelligent client processing method and system for power supply field |
| CN112600982A (en) * | 2020-12-08 | 2021-04-02 | 深圳供电局有限公司 | Power supply service hotline interactive voice response method, system, equipment and medium |
| CN114492391A (en) * | 2021-12-28 | 2022-05-13 | 航天信息股份有限公司 | Intention reasoning method and device |
| CN114942973A (en) * | 2022-04-19 | 2022-08-26 | 尚特杰电力科技有限公司 | Emotion recognition method and system for electric power intelligent customer service system |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116737914A (en) * | 2023-08-15 | 2023-09-12 | 南通话时代信息科技有限公司 | Interactive data processing method and system for cloud customer service center |
| CN116737914B (en) * | 2023-08-15 | 2023-11-17 | 南通话时代信息科技有限公司 | Interactive data processing method and system for cloud customer service center |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115766947B (en) | 2023-05-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111651609A (en) | Multi-turn dialogue method and system integrating knowledge graph and emotion supervision | |
| KR20190004495A (en) | Method, Apparatus and System for processing task using chatbot | |
| CN107957992B (en) | Automatic processing method and system for user feedback information | |
| CN108228764A (en) | A kind of single-wheel dialogue and the fusion method of more wheel dialogues | |
| US10387410B2 (en) | Method and system of classification in a natural language user interface | |
| EP3145127B1 (en) | Short text messaging in digital mobile telecommunication networks | |
| WO2020056621A1 (en) | Learning method and apparatus for intention recognition model, and device | |
| CN107222865A (en) | The communication swindle real-time detection method and system recognized based on suspicious actions | |
| CN108416375B (en) | Work order classification method and device | |
| EP3370230A1 (en) | Voice interaction apparatus, its processing method, and program | |
| CN112885376A (en) | Method and device for improving voice call quality inspection effect | |
| CN115766947A (en) | Intelligent management and control method and system for power grid customer service center | |
| CN115083434B (en) | Emotion recognition method and device, computer equipment and storage medium | |
| CN109993543A (en) | A kind of complaint handling method and system | |
| CN106775665A (en) | The acquisition methods and device of the emotional state change information based on sentiment indicator | |
| CN113988195A (en) | Private domain traffic clue mining method and device, vehicle and readable medium | |
| JP2019041292A (en) | Communication device, program and operator selection method | |
| CN113099043A (en) | Customer service control method, apparatus and computer-readable storage medium | |
| CN113132214A (en) | Conversation method, device, server and storage medium | |
| CN112671984B (en) | Service mode switching method and device, robot customer service and storage medium | |
| CN109787881A (en) | A kind of dialogue method and device with forecast function | |
| CN109002498A (en) | Interactive method, device, equipment and storage medium | |
| WO2021073258A1 (en) | Task follow-up method, apparatus and device based on emotion analysis, and storage medium | |
| CN114756655A (en) | Data query method, device, equipment and storage medium | |
| CN113111157A (en) | Question-answer processing method, device, computer equipment and storage medium |
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 |