CN113298494A

CN113298494A - Intelligent response process engine system for intelligent dialogue system

Info

Publication number: CN113298494A
Application number: CN202110563984.4A
Authority: CN
Inventors: 解思语
Original assignee: Shaanxi Heyou Network Technology Co ltd
Current assignee: Shaanxi Heyou Network Technology Co ltd
Priority date: 2021-05-24
Filing date: 2021-05-24
Publication date: 2021-08-24
Anticipated expiration: 2041-05-24
Also published as: CN113298494B

Abstract

The invention provides an intelligent response process engine system for an intelligent dialog system, which comprises a system embedding module, a process optimizing module, a process executing engine, a manual processing module and an engine upgrading module, wherein the system embedding module is used for embedding the intelligent response process engine system into the intelligent dialog system in an expanding way, the intelligent answer flow engine system is used for expanding and upgrading the intelligent answer system, the flow optimization module is used for optimizing the intelligent flow unit according to the service flow of the original intelligent answer system, the flow execution engine is used for operating a flow frame formed by the flow optimization module, the manual processing module is used for manually processing the abnormal conditions in the operation process of the intelligent answer flow engine system, and the engine upgrading module is used for updating and upgrading the intelligent answer flow engine system.

Description

Intelligent response process engine system for intelligent dialogue system

Technical Field

The invention belongs to the technical field of artificial intelligence, and particularly relates to an intelligent response process engine system for an intelligent dialog system.

Background

The intelligent dialogue system is developed on the basis of large-scale knowledge processing, is applied to the industry, is a large-scale knowledge processing technology, a natural language understanding technology, a knowledge management technology, an automatic question-answering system, an inference technology and the like, has industrial universality, not only provides a fine-grained knowledge management technology for enterprises, but also establishes a quick and effective technical means based on natural language for communication between the enterprises and mass users; meanwhile, statistical analysis information required by fine management can be provided for enterprises.

The intelligent response system orderly and scientifically arranges the accumulated unordered corpus information and establishes a knowledge-based classification model; the classification models can guide the newly added corpus consultation and service information, save human resources, improve the automation of information processing and reduce the operation cost of the website. The method is characterized in that common questions and answers about the basic conditions of governments and enterprises accumulated for many years on the basis of websites are organized into a standard question-answer library form so as to support intelligent question answering of various types of questions. The method is convenient for users, improves the working efficiency and improves the enterprise image.

Due to the uniqueness of enterprise business or the unique management idea of the high level of the enterprise, a plurality of enterprises can customize an exclusive intelligent dialogue system according to the business conditions of the enterprises, so that the unique competitiveness of the enterprises is realized. The intelligent dialogue system is operated for many years, is continuously optimized along with the development of enterprises, and is matched with the business process of the enterprises.

However, the technology of the current intelligent dialogue system is rapidly developed, the previous system is limited by development technology due to early development time, and the system cannot be applied to the current enterprises gradually, and the system is customized again, so that the cost is high, the matching degree of the new system and the enterprises is not high, and more problems are caused. Therefore, the best method for solving the problem is to perform extension upgrading on the original system in an extension embedding mode, and the upgrading of the original system is realized on the premise that the functions of the original system are adjusted as little as possible. Therefore, the invention designs an intelligent answer flow engine system for an intelligent dialogue system.

Disclosure of Invention

In view of the above problems, the present invention provides an intelligent answer flow engine system for an intelligent dialog system.

The technical scheme of the invention is as follows: an intelligent answering process engine system for an intelligent dialogue system comprises a system embedding module, a process optimizing module, a process executing engine, a manual processing module and an engine upgrading module for updating and upgrading the intelligent answering process engine system,

the system embedding module is used for embedding the intelligent response process engine system into an intelligent dialog system in an expanding way and is used for expanding and upgrading the intelligent dialog system, and the system embedding module comprises an expanding interface unit, a process identification unit and a function leading-in unit; the hardware provided with the intelligent response process engine system is inserted into the intelligent dialogue system load terminal, or the intelligent response process engine system data packet is downloaded to the intelligent dialogue system load terminal, the expansion interface unit identifies the intelligent dialogue system interface and carries out serial connection, on the premise of compatibility, the function leading-in unit leads each module unit of the intelligent response process engine system into the corresponding program of the intelligent dialogue system and carries out encapsulation, the process identification unit carries out detection and identification on the service process of the intelligent dialogue system,

the flow optimization module is used for optimizing the intelligent flow unit according to the service flow of the original intelligent dialogue system, and comprises a service logic evaluation unit, a flow increasing unit, a flow deleting unit, a flow changing unit and a closed loop detection unit; the service logic evaluation unit carries out logic detection analysis on the detected and identified service flow, the service flow is added, deleted or changed by the flow adding unit, the flow deleting unit and the flow changing unit, the service flow is backed up before deletion and change, then the optimized service flow is subjected to closed-loop detection by the closed-loop detection unit to ensure that the service flow normally runs in a closed loop,

the flow execution engine is used for operating the flow framework formed by the flow optimization module and comprises a parameter setting unit, an aging limiting unit, a flow operation unit, an automatic transaction unit and a flow backup unit; the optimized business process is operated by the process operation unit, the operation parameters of each link in the business process can be set by the parameter setting unit, the processing aging range of each link in the business process is limited by the aging limiting unit, the business process exceeding the processing aging range is processed preferentially in the running process of the business process and is automatically finished by the automatic finishing unit, the whole business process is backed up by the process backup unit after the business process is operated in a closed loop,

the manual processing module is used for manually processing the abnormal conditions of the running process of the intelligent response process engine system.

Furthermore, the intelligent response process engine system further comprises backup databases respectively connected with the system embedding module, the process optimizing module, the process executing engine, the manual processing module and the engine upgrading module, and used for storing backup data of each module, wherein the backup databases comprise backup restoring units used for restoring the backup data to the intelligent dialogue system or the intelligent response process engine system, and the backup data can avoid that the system data before the abnormality occurs cannot be restored when the system is abnormal.

Furthermore, the system embedding module also comprises a data backup unit and a compatibility adapting unit; before the system embedding module embeds the intelligent response process engine system extension into the intelligent dialog system, the method further comprises the following steps:

the data backup unit stores the original data backup of the intelligent dialogue system in the backup database, the original data backup can ensure that the original normal running system can be restored when the system has a problem of embedding and expansion, the damage or the loss of the original data can be avoided,

the data backup unit stores the original data backup of the intelligent dialogue system in the backup database, the original data backup can ensure that the system which is initially normally operated can be restored when the system is in a problem of embedding and expanding, the original data is prevented from being damaged or lost, the compatibility adaptation unit carries out algorithm detection on the compatibility of the intelligent dialogue system and the intelligent response process engine system, the compatibility of the two systems can ensure the normal operation of the program algorithm of the module unit, and the system operation fault caused by the incompatibility or partial incompatibility of the systems is avoided.

Further, the backup data in the backup database executes an auto-delete algorithm:

after the backup data is uploaded to the backup database, writing a backup date character string in a log of the backup data, if the backup date is 2021, 04, 29, writing upload20210429 in the log,

the backup database runs an algorithm program for calculating the backup storage time at 00:00 every day: t0 is t2-t1, wherein t0 is the number of days the backup data has been stored in the backup database, t2 is the date of day, t1 is the date of the backup day, when t0 is larger than or equal to 90d, the backup data is deleted,

and if the backup data is subjected to a restoring operation during the saving period, updating the backup date character string in the backup data log to the date of the current date of restoration.

Furthermore, the abnormal conditions of the intelligent response flow engine system in the operation process include flow abnormal conditions caused by abnormal system operation, business flow abnormal conditions caused by the failure of intelligent response to solve the conversation problem, and manual customer service intervention conditions actively initiated by the client in the intelligent conversation process, and different types of abnormalities can be matched with different customer service personnel for solving.

Furthermore, the manual processing module also comprises a service transfer-over unit and a service auxiliary unit,

the different processing algorithms of the manual processing module when different abnormal conditions occur comprise:

when the abnormal flow condition caused by the abnormal system operation occurs, the abnormal condition is transferred to the abnormal system processing customer service by the service transfer unit, and common customer service personnel can not rapidly solve the problem and need professional system personnel to process the problem when the system is in the abnormal operation;

when the abnormal condition of the business process caused by the failure of intelligent response to solve the conversation problem and the intervention condition of the artificial customer service actively initiated by the client in the intelligent conversation process occur, the abnormal condition is forwarded to the business process abnormal handling customer service by the business forwarding unit, and when the abnormal handling customer service of the business process handles the abnormal business process, the business auxiliary unit provides the relevant information data of the business process to the artificial customer service.

Furthermore, the service flow related information data provided by the service auxiliary unit comprises intelligent dialogue historical dialogue information, response information matched with keywords in the intelligent dialogue, intelligent response common-use-technology information and an intelligent response flow chart, and information acquisition of manual processing response flows of manual customer service is optimized.

Further, the manual processing module processes the algorithm strategy of the business process after the abnormal business process is processed:

the result detection unit detects the processing result, the service rotation unit rotates the service flow to the flow execution engine for normal operation after the processing result is detected, the flow execution engine can automatically operate the recovered normal service flow without occupying human customer service resources,

and when the processing result does not pass, returning to the manual processing module for continuous processing until the processing result is detected to pass, and ensuring that the abnormal business flow is processed.

Further, the engine upgrading module comprises a system backup unit, a networking downloading unit, a decryption decompression unit and an updating upgrading unit, and the process of updating and upgrading the intelligent response process engine system by the engine upgrading module comprises the following steps:

the system backup unit backs up the whole system, the backup before updating and upgrading can avoid the loss or damage of the original system data caused by updating and upgrading,

the updating and upgrading data packet is downloaded through the networking downloading unit, the updating and upgrading data packet is decrypted and decompressed by the decryption and decompression unit, so that external viruses are prevented from entering the original system through the updating and upgrading data packet,

and the updating and upgrading program in the updating and upgrading data packet is covered into the original intelligent response process engine system by the updating and upgrading unit.

Furthermore, the engine upgrading module further includes an update detection unit, and after the engine upgrading module performs update and upgrade on the intelligent response process engine system, the engine upgrading module further includes:

the update detection unit is used for detecting the running condition of the intelligent response process engine system after update and upgrade,

the update data packet is deleted after the normal operation, the system space is saved,

and when the operation is abnormal, the backup restoring unit of the backup database restores the system to the state before updating and upgrading.

The working method of the invention comprises the following steps:

s1: embedded extension of intelligent answer flow engine system

Inserting hardware provided with an intelligent response process engine system on an intelligent dialogue system load terminal, or downloading an intelligent response process engine system data packet to the intelligent dialogue system load terminal, identifying and connecting an intelligent dialogue system interface by an expansion interface unit of a system embedding module, then storing original data backup of the intelligent dialogue system in a backup database by a data backup unit, then detecting the compatibility of the intelligent dialogue system and the intelligent response process engine system, leading each module unit of the intelligent response process engine system into a corresponding program of the intelligent dialogue system by a function leading-in unit on the premise of compatibility, and then detecting and identifying the service process of the intelligent dialogue system by a process identification unit;

s2: optimized upgrade of intelligent response flow

The service logic evaluation unit of the process optimization module performs logic detection and analysis on the service process detected and identified in the step S1, the service process is added, deleted and changed through the process adding unit, the process deleting unit and the process changing unit, the service process is backed up before deletion and change, and then the optimized service process is subjected to closed-loop detection through the closed-loop detection unit, so that the service process is ensured to normally operate in a closed loop;

s3: intelligent conversational business process operations

The optimized business process is operated through the process execution engine, the operation parameters of each link in the business process can be set through the parameter setting unit, the processing aging range of each link in the business process is limited by the aging limiting unit, the business process exceeding the processing aging range is processed preferentially in the running process of the business process, the business process is automatically finished by the automatic finishing unit, and the whole business process is backed up by the process backup unit after the business process is operated in a closed loop;

s4: abnormal business process handling

When a process abnormal condition caused by system operation abnormity occurs, a service transfer unit of a manual processing module transfers the abnormal condition to a system abnormal processing customer service, when an intelligent response fails to solve the service process abnormal condition caused by a conversation problem and a manual customer service intervention condition actively initiated by a client in an intelligent conversation process occurs, a service transfer unit of the manual processing module transfers the abnormal condition to the service process abnormal processing customer service, when the service process abnormal processing customer service processes the abnormal service process, a service auxiliary unit provides relevant information data of the service process to the manual customer service, after the abnormal service process is processed, a result detection unit detects a processing result, after the processing result is detected to pass, a service rotary unit rotates the service process to a process execution engine to normally operate, and when the processing result is not passed, the result is returned to the manual processing module to continue processing, until the processing result is detected to pass;

s5: update and upgrade of intelligent response process engine system

Before updating and upgrading, a system backup unit backs up the whole system, an updating and upgrading data packet is downloaded through a networking downloading unit, a decryption and decompression unit decrypts and decompresses the updating and upgrading data packet, an updating and upgrading program in the updating and upgrading data packet is covered into an original intelligent response process engine system through an updating and upgrading unit, an updating detection unit is used for detecting the operation condition of the intelligent response process engine system after updating and upgrading technology is carried out, the updating data packet is deleted after normal operation, and a backup restoration unit of a backup database restores the system to the state before updating and upgrading when abnormal operation occurs.

The invention has the beneficial effects that: the invention provides an intelligent response flow engine system for an intelligent dialogue system, which is particularly suitable for upgrading, expanding and upgrading the intelligent dialogue system with laggard technology, can be expanded to a terminal loaded by the intelligent dialogue system in a hardware insertion and software downloading way, has diversified expansion ways, meets the use of the intelligent dialogue system in a networking and offline state, can expand and embed the intelligent response flow engine system into the intelligent dialogue system through a system embedding module on the premise of less influence on the intelligent dialogue system, realizes the updating and upgrading of the intelligent dialogue system, can back up the system before the expansion and upgrading so as to be restored at any time, can carry out compatibility detection before the expansion and embedding, can expand and embed after adaptation, can carry out detection and optimization on the service flow of the original intelligent dialogue system after expansion, and the optimized service flow is operated by an execution engine, the timeliness of the treatment is ensured, and the intelligent response process engine system can also be updated and upgraded. In a word, the invention has the advantages of perfect system, convenient use, large upgrading space and the like.

Drawings

FIG. 1 is a block diagram of the system architecture of the present invention.

Wherein, 1-system embedding module, 11-expansion interface unit, 12-data backup unit, 13-compatibility adapting unit, 14-process identifying unit, 15-function leading-in unit, 2-process optimizing module, 21-service logic evaluating unit, 22-process increasing unit, 23-process deleting unit, 24-process changing unit, 25-closed loop detecting unit, 3-process executing engine, 31-parameter setting unit, 32-time limit unit, 33-process operating unit, 34-automatic transaction unit, 35-process backup unit, 4-manual processing module, 41-service transfer unit, 42-service auxiliary unit, 43-result detecting unit, 44-service rotary unit, 13-service rotary unit, 5-engine upgrading module, 51-system backup unit, 52-networking downloading unit, 53-decryption decompression unit, 54-update upgrading unit, 55-update detection unit, 6-backup database and 61-backup restoring unit.

Detailed Description

For the convenience of understanding the technical solution of the present invention, the following description is made with reference to fig. 1 and the specific embodiments, which are not intended to limit the scope of the present invention.

Example 1: an intelligent response process engine system for an intelligent dialogue system comprises a system embedding module 1, a process optimizing module 2, a process executing engine 3, a manual processing module 4 and an engine upgrading module 5 for updating and upgrading the intelligent response process engine system,

the system embedding module 1 is used for embedding the intelligent response process engine system into an intelligent dialog system in an expanding way and is used for expanding and upgrading the intelligent dialog system, and the system embedding module 1 comprises an expanding interface unit 11, a process identification unit 14 and a function importing unit 15; the hardware provided with the intelligent response process engine system is inserted into the intelligent dialogue system load terminal, or the intelligent response process engine system data packet is downloaded to the intelligent dialogue system load terminal, the expansion interface unit 11 of the system embedded module 1 identifies the intelligent dialogue system interface and carries out serial port connection, on the premise of compatibility, the function leading-in unit 15 leads each module unit of the intelligent response process engine system into the corresponding program of the intelligent dialogue system and carries out encapsulation, the process identification unit 14 carries out detection and identification on the service process of the intelligent dialogue system,

the process optimization module 2 is used for optimizing the intelligent process unit according to the service process of the original intelligent dialogue system, and the process optimization module 2 comprises a service logic evaluation unit 21, a process adding unit 22, a process deleting unit 23, a process changing unit 24 and a closed-loop detection unit 25; the service logic evaluation unit 21 performs logic detection analysis on the detected and identified service flow, the service flow is added, deleted or changed by the flow adding unit 22, the flow deleting unit 23 and the flow changing unit 24, the service flow is backed up before deletion and change, then the optimized service flow is subjected to closed-loop detection by the closed-loop detection unit 25 to ensure that the service flow normally operates in a closed loop,

the process execution engine 3 is used for operating the process framework formed by the process optimization module 2, and the process execution engine 3 comprises a parameter setting unit 31, an aging limiting unit 32, a process operation unit 33, an automatic transaction unit 34 and a process backup unit 35; the optimized business process is operated by the process operation unit 33, the operation parameters of each link in the business process can be set by the parameter setting unit 31, the processing aging range of each link in the business process is limited by the aging limiting unit 32, the business process exceeding the processing aging range is processed preferentially in the running process of the business process and is automatically finished by the automatic finishing unit 34, the business process is backed up integrally by the process backup unit 35 after the business process is operated in a closed loop,

the manual processing module 4 is used for manually processing the abnormal conditions of the operation process of the intelligent response flow engine system,

the intelligent response process engine system further comprises a backup database 6 respectively connected with the system embedding module 1, the process optimizing module 2, the process executing engine 3, the manual processing module 4 and the engine upgrading module 5, and used for storing backup data of each module, wherein the backup database 6 comprises a backup restoring unit 61 used for restoring the backup data to the intelligent dialogue system or the intelligent response process engine system.

Example 2: an intelligent response process engine system for an intelligent dialogue system comprises a system embedding module 1, a process optimizing module 2, a process executing engine 3, a manual processing module 4 and an engine upgrading module 5 for updating and upgrading the intelligent response process engine system,

the system embedding module 1 is used for embedding the intelligent response process engine system into an intelligent dialog system in an expanding way and is used for expanding and upgrading the intelligent dialog system, and the system embedding module 1 comprises an expanding interface unit 11, a process identification unit 14 and a function importing unit 15; the hardware provided with the intelligent response process engine system is inserted on the intelligent dialogue system load terminal, or the intelligent response process engine system data packet is downloaded to the intelligent dialogue system load terminal, the expansion interface unit 11 of the system embedding module 1 identifies the intelligent dialogue system interface and carries out serial port connection,

the system embedding module also comprises a data backup unit 12 and a compatibility adapting unit 13, wherein the data backup unit 12 stores the original data backup of the intelligent dialogue system in the backup database 6, the compatibility adapting unit 13 carries out algorithm detection on the compatibility of the intelligent dialogue system and the intelligent response process engine system,

under the premise of compatibility, the function importing unit 15 imports each module unit of the intelligent answering process engine system into the corresponding program of the intelligent dialogue system and packages the module unit, the process identifying unit 14 detects and identifies the business process of the intelligent dialogue system,

Example 3: an intelligent response process engine system for an intelligent dialogue system comprises a system embedding module 1, a process optimizing module 2, a process executing engine 3, a manual processing module 4 and an engine upgrading module 5 for updating and upgrading the intelligent response process engine system,

the intelligent answering process engine system also comprises a backup database 6 respectively connected with the system embedding module 1, the process optimizing module 2, the process executing engine 3, the manual processing module 4 and the engine upgrading module 5, and is used for storing backup data of each module, and the backup data in the backup database 6 executes an automatic deleting algorithm:

after the backup data is uploaded to the backup database 6, writing a backup date character string in the log of the backup data, if the backup date is 2021, 04, 29, writing upload20210429 in the log,

the backup database 6 runs an algorithm program for calculating the backup storage time at 00:00 a day: t0 is t2-t1, where t0 is the number of days the backup data has been stored in the backup database 6, t2 is the date of day, t1 is the date of the backup day, when t0 is greater than or equal to 90d, the backup data is deleted,

if the backup data is subjected to a restore operation during the saving, the backup date character string in the backup data log is updated to the date of the restore,

the backup database 6 includes a backup restoring unit 61 for restoring backup data to the intelligent dialogue system or the intelligent answer flow engine system.

Example 4: an intelligent response process engine system for an intelligent dialogue system comprises a system embedding module 1, a process optimizing module 2, a process executing engine 3, a manual processing module 4 and an engine upgrading module 5 for updating and upgrading the intelligent response process engine system,

the manual processing module 4 is used for manually processing the abnormal conditions of the intelligent response flow engine system in the operation process, the abnormal conditions of the intelligent response flow engine system in the operation process comprise flow abnormal conditions caused by abnormal system operation, service flow abnormal conditions caused by the intelligent response failing to solve the conversation problem, and manual customer service intervention conditions actively initiated by customers in the intelligent conversation process,

the manual processing module 4 comprises a service care-of unit 41 and a service supplementary unit 42,

the different processing algorithms of the manual processing module 4 when different exceptional situations occur include:

when a process abnormal condition caused by system operation abnormity occurs, the abnormal condition is forwarded to a system abnormal handling customer service by the service forwarding unit 41;

when the abnormal situation of the business process caused by the failure of intelligent response to solve the conversation problem and the intervention situation of the artificial customer service actively initiated by the client in the intelligent conversation process occur, the abnormal situation is forwarded to the abnormal handling customer service of the business process by the business forwarding unit 41, the related information data of the business process is provided for the artificial customer service by the business auxiliary unit 42 when the abnormal handling customer service of the business process handles the abnormal business process, the related information data of the business process provided by the business auxiliary unit 42 comprises the historical conversation information of the intelligent conversation, the response information matched with the key words in the intelligent conversation, the common conversation information of the intelligent response and the intelligent response process chart, the information acquisition of the artificial handling response process of the artificial customer service is optimized,

Example 5: an intelligent response process engine system for an intelligent dialogue system comprises a system embedding module 1, a process optimizing module 2, a process executing engine 3, a manual processing module 4 and an engine upgrading module 5 for updating and upgrading the intelligent response process engine system,

the manual processing module 4 further includes a result detection unit 43, and the manual processing module 4 processes the algorithm strategy of the abnormal business process after the abnormal business process is finished:

the processing result is detected by the result detecting unit 43, the service flow is returned to the flow executing engine 3 by the service returning unit 44 after the processing result is detected to be passed for normal operation,

if the processing result does not pass, returning to the manual processing module 4 for continuous processing until the processing result is detected to pass,

Example 6: an intelligent response process engine system for an intelligent dialogue system comprises a system embedding module 1, a process optimizing module 2, a process executing engine 3, a manual processing module 4 and an engine upgrading module 5 for updating and upgrading the intelligent response process engine system,

the process of updating and upgrading the intelligent response process engine system by the engine upgrading module 5 comprises the following steps:

the system as a whole is backed up by the system backup unit 51,

the updating and upgrading data package is downloaded by the networking download unit 52, the updating and upgrading data package is decrypted and decompressed by the decryption and decompression unit 53,

the upgrade program in the upgrade data packet is overwritten into the original intelligent response flow engine system by the upgrade unit 54,

Example 7: an intelligent response process engine system for an intelligent dialogue system comprises a system embedding module 1, a process optimizing module 2, a process executing engine 3, a manual processing module 4 and an engine upgrading module 5 for updating and upgrading the intelligent response process engine system,

the system as a whole is backed up by the system backup unit 51,

the engine upgrading module 5 further comprises the following steps after updating and upgrading the intelligent response process engine system:

the update detection unit 55 is used to detect the operation status of the intelligent response process engine system after update and upgrade,

the update data packet is deleted after the normal operation,

the backup restoring unit 61 of the backup database 6 restores the system to the state before update and upgrade when the operation is abnormal,

The working method of the above embodiment 7 includes the steps of:

s1: embedded extension of intelligent answer flow engine system

Inserting hardware provided with an intelligent response process engine system on an intelligent conversation system load terminal, or downloading an intelligent response process engine system data packet to the intelligent conversation system load terminal, identifying and connecting an intelligent conversation system interface by an expansion interface unit 11 of a system embedding module 1, then backing up and storing original data of the intelligent conversation system in a backup database 6 by a data backup unit 12, then detecting the compatibility of the intelligent conversation system and the intelligent response process engine system, importing each module unit of the intelligent response process engine system into a corresponding program of the intelligent conversation system by a function importing unit 15 on the premise of compatibility, and then detecting and identifying a service process of the intelligent conversation system by a process identifying unit 14;

s2: optimized upgrade of intelligent response flow

The service logic evaluation unit 21 of the process optimization module 2 performs logic detection and analysis on the service flow detected and identified in step S1, the service flow is added, deleted and changed through the flow adding unit 22, the flow deleting unit 23 and the flow changing unit 24, the service flow is backed up before deletion and change, and then the optimized service flow is subjected to closed-loop detection through the closed-loop detection unit 25, so as to ensure that the service flow normally operates in a closed loop;

s3: intelligent conversational business process operations

The optimized business process is operated through the process execution engine 3, the operation parameters of each link in the business process can be set through the parameter setting unit 31, the processing aging range of each link in the business process is limited by the aging limiting unit 32, the business process exceeding the processing aging range is processed preferentially in the running process of the business process and is automatically finished by the automatic finishing unit 34, and the whole business process is backed up by the process backup unit 35 after the business process is closed in loop;

s4: abnormal business process handling

When a process abnormal condition caused by system operation abnormity occurs, the abnormal condition is transferred to a system abnormal processing customer service by a service transfer unit 41, when a service process abnormal condition caused by that intelligent response can not solve a conversation problem and a manual customer service intervention condition actively initiated by a client in an intelligent conversation process occur, the abnormal condition is transferred to the service process abnormal processing customer service by the service transfer unit 41, when the abnormal service process is processed by the service process abnormal processing customer service, the service auxiliary unit 42 provides information data related to the service process to the manual customer service, after the abnormal service process is processed, a processing result is detected by a result detection unit 43, after the processing result is detected, the service process is rotated to a process execution engine 3 by a service rotation unit 44 to be normally operated, and when the processing result is not passed, the processing result is returned to a manual processing module 4 to be continuously processed, until the processing result is detected to pass;

s5: update and upgrade of intelligent response process engine system

Before updating and upgrading, a system backup unit 51 backs up the whole system, an updating and upgrading data packet is downloaded through a networking downloading unit 52, the updating and upgrading data packet is decrypted and decompressed by a decryption and decompression unit 53, an updating and upgrading program in the updating and upgrading data packet is covered into an original intelligent response flow engine system by an updating and upgrading unit 54, the operation condition of the intelligent response flow engine system after updating and upgrading is detected by an updating detection unit 55 after updating and upgrading technology, the updating data packet is deleted after normal operation, and the system is restored to the state before updating and upgrading by a backup restoration unit 61 of a backup database 6 when the operation is abnormal.

Claims

1. An intelligent response process engine system for an intelligent dialogue system is characterized by comprising a system embedding module (1), a process optimizing module (2), a process executing engine (3), a manual processing module (4) and an engine upgrading module (5) for updating and upgrading the intelligent response process engine system,

the system embedding module (1) is used for embedding the intelligent response process engine system into an intelligent dialog system in an expanding way and is used for expanding and upgrading the intelligent dialog system, and the system embedding module (1) comprises an expanding interface unit (11), a process identification unit (14) and a function importing unit (15); the hardware provided with the intelligent response process engine system is inserted into the intelligent dialogue system load terminal, or the intelligent response process engine system data packet is downloaded to the intelligent dialogue system load terminal, the expansion interface unit (11) of the system embedding module (1) identifies the intelligent dialogue system interface and carries out serial connection, each module unit of the intelligent response process engine system is led into the corresponding program of the intelligent dialogue system by the function leading-in unit (15) and is packaged under the premise of compatibility, the process identification unit (14) is used for detecting and identifying the business process of the intelligent dialogue system,

the flow optimization module (2) is used for optimizing the intelligent flow unit according to the service flow of the original intelligent dialogue system, and the flow optimization module (2) comprises a service logic evaluation unit (21), a flow adding unit (22), a flow deleting unit (23), a flow changing unit (24) and a closed loop detection unit (25); the service logic evaluation unit (21) performs logic detection analysis on the detected and identified service flow, the service flow is added, deleted or changed through the flow adding unit (22), the flow deleting unit (23) and the flow changing unit (24), the service flow is backed up before deletion and change, then the optimized service flow is subjected to closed-loop detection through the closed-loop detection unit (25) to ensure that the service flow normally operates in a closed loop,

the flow execution engine (3) is used for operating the flow framework formed by the flow optimization module (2), and the flow execution engine (3) comprises a parameter setting unit (31), an aging limiting unit (32), a flow operation unit (33), an automatic handling unit (34) and a flow backup unit (35); the optimized business process is operated by the process operation unit (33), the operation parameters of each link in the business process can be set by the parameter setting unit (31), the processing aging range of each link in the business process is limited by the aging limiting unit (32), the business process exceeding the processing aging range is processed preferentially in the business process operation process and automatically finished by the automatic finishing unit (34), the business process is backed up integrally by the process backup unit (35) after the business process operates a closed loop,

and the manual processing module (4) is used for manually processing the abnormal conditions of the intelligent response flow engine system in the operation process.

2. The intelligent answer flow engine system for the intelligent dialog system according to claim 1, further comprising a backup database (6) connected to the system embedding module (1), the flow optimizing module (2), the flow executing engine (3), the manual processing module (4), and the engine upgrading module (5), respectively, for storing backup data of each module, wherein the backup database (6) comprises a backup restoring unit (61) for restoring the backup data to the intelligent dialog system or the intelligent answer flow engine system.

3. The intelligent answer flow engine system for intelligent dialog systems according to claim 2, characterized in that the system embedding module (1) further comprises a data backup unit (12), a compatibility adaptation unit (13); the system embedding module (1) is used for embedding the intelligent response process engine system extension into the intelligent dialog system, and the method also comprises the following steps:

the intelligent dialogue system original data backup is stored in a backup database (6) by a data backup unit (12),

the compatibility of the intelligent dialogue system and the intelligent response flow engine system is detected by a compatibility adaptation unit (13) in an algorithm mode.

4. An intelligent answer flow engine system for an intelligent dialog system according to claim 2, characterized in that the backup data in the backup database (6) executes an automatic deletion algorithm:

after the backup data are uploaded to a backup database (6), a backup date character string is written in a log of the backup data, if the backup date is 2021, 04, 29, an upload20210429 is written in the log,

the backup database (6) runs an algorithm program for calculating the backup storage time at 00:00 a day: t0 is t2-t1, wherein t0 is the number of days that the backup data has been stored in the backup database (6), t2 is the date of the day, t1 is the date of the backup day, when t0 is larger than or equal to 90d, the backup data is deleted,

5. The intelligent response flow engine system for the intelligent dialogue system according to claim 1, wherein the abnormal situations of the intelligent response flow engine system during the operation process include abnormal flow situations caused by abnormal system operation, abnormal business flow situations caused by the intelligent response failing to solve the dialogue problem, and manual customer service intervention situations actively initiated by the customer during the intelligent dialogue process.

6. An intelligent answer flow engine system for intelligent dialogue system according to claim 1, characterized by that, the manual processing module (4) further comprises a service handover unit (41), a service auxiliary unit (42);

the different processing algorithms of the manual processing module (4) when different abnormal conditions occur comprise:

when the abnormal flow condition caused by the abnormal system operation occurs, the abnormal condition is transferred to the system abnormal handling customer service by the service transfer unit (41);

when the abnormal condition of the business process caused by the fact that the intelligent response cannot solve the conversation problem and the intervention condition of the artificial customer service actively initiated by the client in the intelligent conversation process occur, the abnormal condition is forwarded to the business process abnormal handling customer service by the business forwarding unit (41), and when the abnormal business process is handled by the business process abnormal handling customer service, the business auxiliary unit (42) provides the relevant information data of the business process to the artificial customer service.

7. The intelligent response flow engine system for the intelligent dialogue system according to claim 6, wherein the business auxiliary unit (42) provides the business flow related information data including intelligent dialogue historical dialogue information, response information matched with keywords in the intelligent dialogue, intelligent response common dialogue information and intelligent response flow chart, and optimizes information acquisition of manual customer service manual processing response flow.

8. The intelligent answer flow engine system for the intelligent dialogue system according to claim 1, wherein the manual processing module (4) processes the business flow after the abnormal business flow is processed:

the processing result is detected by a result detection unit (43), the service process is returned to the process execution engine (3) by a service return unit (44) to normally operate after the processing result is detected,

and when the processing result does not pass, returning to the manual processing module (4) for continuous processing until the processing result is detected to pass.

9. The intelligent answer flow engine system for the intelligent dialogue system according to claim 1, wherein the engine upgrade module (5) comprises a system backup unit (51), a networking download unit (52), a decryption decompression unit (53), and an update upgrade unit (54), and the process of the engine upgrade module (5) for updating and upgrading the intelligent answer flow engine system comprises:

the system backup unit (51) backs up the whole system,

the updating and upgrading data packet is downloaded by a networking download unit (52), the updating and upgrading data packet is decrypted and decompressed by a decryption and decompression unit (53),

and the updating and upgrading program in the updating and upgrading data packet is covered into the original intelligent response process engine system by the updating and upgrading unit (54).

10. The intelligent answer flow engine system for intelligent dialog system according to claim 1, wherein the engine upgrade module (5) further comprises an update detection unit (55), and the engine upgrade module (5) further comprises, after performing update upgrade on the intelligent answer flow engine system:

the operation condition of the intelligent response process engine system after updating and upgrading is detected by an updating detection unit (55),

the update data packet is deleted after the normal operation,

when the operation is abnormal, the backup restoration unit (61) of the backup database (6) restores the system to the state before updating and upgrading.