Intent translation method, system, medium and interactive interface for information communication network

Abstract

The present invention belongs to the technical field of information communication network, and discloses an intention translation method, system, medium and interactive interface of an information communication network. A user inputs a non-technical intention through a front-end web page, and sends the intention to an intention engine module for entity recognition and translation; the intention engine module feeds back the intention translation result to the front-end web page, and stores the intention confirmed by the user to be accurately translated into an intention library; the experience value of the technical intention can be stored in the intention library through the front-end page, and the experience value is updated through interaction with the intention engine through a learning model; the result of the translation by the intention engine module is fed back to the front-end page. The present invention can be used for non-technical intention translation of ordinary users who do not pay attention to the details of the underlying network, and can also be used for technical intention translation for professionals of specific network technology. Design an intention translation front-end page, provide a text-based intention input page for non-technical intentions, and provide a point-to-point intention input page for technical intentions.

Description

Technical Field

[0001] The present invention belongs to the technical field of information communication networks, and in particular, relates to an intention translation method, system, medium and interactive interface of an information communication network. Background Art

[0002] At present: Due to the rapid development of technologies such as software-defined networks, network function virtualization, natural language processing, and artificial intelligence, the traditional network system is undergoing tremendous changes. The management and operation and maintenance of large enterprise networks are becoming more and more complex. The growing network scale and increasingly complex network strategies have brought unprecedented pressure to network operation and maintenance. In line with the development trend, major organizations and start-ups around the world have proposed intent-based network solutions to cope with different scenarios: Huawei has launched an intent-driven intelligent network, Cisco has proposed the concept of a fully intelligent network, Juniper is committed to building an autonomous network to liberate operation and maintenance personnel, and Yunshan has proposed the concept of an intent-based autonomous driving network. Intent-driven networks include modules such as intent translation, policy mapping, and intent verification. Based on technologies such as natural language processing, formal verification, machine learning, and network telemetry, they will gradually realize automated, intelligent, and convenient network operation and maintenance management, realize automatic and rapid deployment of network applications, improve network agility, shield users from underlying network details, and create a network centered on user experience.

[0003] Intent-driven networking is an emerging network technology that allows end users and network administrators to obtain automated, intelligent and reliable services by simply expressing their network service requirements without having to pay attention to the implementation details of network technology. Intent-driven networking can automatically translate, verify, deploy, configure and optimize according to the intentions of end users and network administrators to achieve the desired network status. At the same time, relying on network holographic perception and feedback optimization closed loop, it can automatically resolve abnormal events and ensure network reliability. Among them, intent is a declarative way to describe the system status. It abstracts the operation objects and functions of the network from the perspective of demand, and can be converted into network policies through policy mapping technology and decentralized to the underlying network for execution.

[0004] Currently, network control personnel rely on low-level languages ​​to formulate network policies. Ordinary users do not have network expertise and programming skills, and cannot control the network conveniently and effectively. The intent representation system provides network control personnel and ordinary users with a method to translate low-level languages ​​into standard intent expressions that can be recognized by the network, which is the first step in realizing the automatic deployment of intent-driven networks.

[0005] The intent translation method solves the problem of diversified intent input, and can be used for non-technical intent translation of terminal users who do not pay attention to the details of the underlying network, such as "dispatch a normal-level video service from Beijing user A to Nanjing user B, the time requirement is 2020-09-12-10:30 to 2020-09-12-12:30." It can also be used for technical intent translation for professionals such as network administrators and service operators for specific network technologies, such as "source node: A (address: 40.0.0.5); destination node: B (address: 54.251.17.123); bandwidth: 100Mbps; delay: 20ms; packet loss rate: 2%." In wireless networks, there have been some studies on network intent translation methods, however, they have not formed a complete theoretical system and engineering implementation method, and cannot take into account both technical and non-technical intent translation. The intention translation system and interactive interface of the information communication network are designed with interactive interfaces for end users and professional technicians respectively, and can solve the problems of non-technical intention and technical intention translation at the same time based on natural language processing, etc.

[0006] Prior art 1 proposes a high-level model for representing network policies, namely, a policy abstraction graph, which can abstract the underlying network infrastructure into endpoints and use simple and intuitive graphics to represent the network policies between endpoints. By using the SDN programming language Pyretic to characterize the middleware functions and analyze the service chain, the intent translation of routing, traffic monitoring, access control, etc. can be realized. The disadvantage of this method is that it does not support the translation of QoS intent and time-varying intent, and can only complete the translation of static intents such as access control, and it is difficult to realize dynamic intent translation.

[0007] The second prior art provides a voice assistant page for large and complex networks to obtain the network administrator's configuration, fault detection and other intentions. The intention engine module can translate the input intention into a network configuration strategy and an automated network management interactive interface. The shortcomings of this method are that the provided intention translation method does not take into account the current state of the network, cannot realize dynamic mapping of intentions in combination with network resources, does not design intention translation result feedback and verification methods, and cannot ensure the accuracy and executability of the intention translation results.

[0008] Through the above analysis, the problems and defects of the prior art are as follows:

[0009] (1) The existing technology only designs a policy graph mapping model for static access control scenarios, but does not involve a dynamic intent translation model, and cannot solve the translation of QoS intent and time-varying intent.

[0010] (2) The front-end interface provided by the second prior art does not feed back the intent translation result to the user for confirmation, and does not complete the translation of the input intent into the network strategy in combination with the current state of the network, and therefore lacks accuracy and executability.

[0011] The difficulty in solving the above problems and defects is as follows: the information and communication network structure is complex, the scale is growing, and more and more applications are deployed, which makes manual operation and maintenance prone to errors. For complex network management, a simple and flexible northbound interface technology is urgently needed to abstract network intent and shield the underlying details of the network. On the other hand, the current northbound interface language of information and communication networks is difficult to unify, which brings inconvenience to the implementation of cross-domain network intent.

[0012] The significance of solving the above problems and defects is that intent translation is the first step to realize intent-driven network. Solving the intent translation technology problem of information communication network can simplify the interaction between upper-layer applications and underlying controllers, shield users from specific network technologies such as MPLS and routing protocols, and realize flexible, intelligent, efficient and convenient network management. The purpose of studying intent translation method is to provide a declarative intent description method for both professional operation and maintenance personnel and ordinary network users, and realize cross-domain intent translation and cross-platform portability through application layer requirements separated from underlying details. Summary of the invention

[0013] In view of the problems existing in the prior art, the present invention provides an intention translation method, system, medium and interactive interface for an information communication network.

[0014] The present invention is implemented as follows: a method for translating the intention of an information communication network, characterized in that the method for translating the intention of an information communication network comprises:

[0015] Users input non-technical intents through the front-end web page, and the intents are sent to the intent engine module for entity recognition and translation processing;

[0016] The intent engine module feeds back the intent translation result to the front-end web page, and stores the intent confirmed by the user to be accurately translated into the intent library;

[0017] The experience value of technical intent can be sent to the intent library through the front-end page, and the experience value is updated through the interaction between the learning model and the intent engine;

[0018] The results of the translation by the intent engine module are fed back to the front-end page.

[0019] Furthermore, the front-end web page includes an input example and an input intent text box. The user enters the original intent in the text box based on the input example, and the input intent text is sent to the intent translation engine in the form of an HTTP request.

[0020] Furthermore, the entity recognition of the intent translation system of the information communication network is to mark the text in the input intent corpus with BMEO tags, and then accurately identify the key information of the intent through named entity recognition;

[0021] Define network state behavior labels, characterize network states, generate state transition diagrams with the help of finite state machine training, and realize intent translation.

[0022] Furthermore, the intent translation result feedback page of the intent translation system of the information communication network is displayed as an intent name-value pair, and the intent translation result is stored in the intent library constructed by the intent translation engine, using the relational database PostgreSQL for storage.

[0023] Furthermore, the technical intent of the intent translation system of the information communication network can be input through a point-and-click page, and the key parameters of the intent are stored in the intent library;

[0024] Build a learning model to train the input technical intent and transmit it to the intent engine for translation confirmation.

[0025] Furthermore, the technical intent translation result of the intent translation system of the information communication network is sent to the feedback page and displayed as an intent name-value pair.

[0026] Another object of the present invention is to provide a computer device, the computer device comprising a memory and a processor, the memory storing a computer program, and when the computer program is executed by the processor, the processor performs the following steps:

[0027] Users input non-technical intents through the front-end web page, and the intents are sent to the intent engine module for entity recognition and translation processing;

[0028] The intent engine module feeds back the intent translation result to the front-end web page, and stores the intent confirmed by the user to be accurately translated into the intent library;

[0029] The experience value of technical intent can be sent to the intent library through the front-end page, and the experience value is updated through the interaction between the learning model and the intent engine;

[0030] The results of the translation by the intent engine module are fed back to the front-end page.

[0031] Another object of the present invention is to provide a computer-readable storage medium storing a computer program, wherein when the computer program is executed by a processor, the processor executes the following steps:

[0032] Users input non-technical intents through the front-end web page, and the intents are sent to the intent engine module for entity recognition and translation processing;

[0033] The intent engine module feeds back the intent translation result to the front-end web page, and stores the intent confirmed by the user to be accurately translated into the intent library;

[0034] The experience value of technical intent can be sent to the intent library through the front-end page, and the experience value is updated through the interaction between the learning model and the intent engine;

[0035] The results of the translation by the intent engine module are fed back to the front-end page.

[0036] Another object of the present invention is to provide an interactive interface information data processing terminal, which is used to implement the intention translation method of the information communication network.

[0037] Another object of the present invention is to provide an information communication network intention translation system for implementing the information communication network intention translation method, the information communication network intention translation system comprising:

[0038] The user intent input module is used to implement non-technical intents. The original intent is input through the intent input text box on the front-end page, and transmitted to the intent engine. The translation result confirmed by the user is stored in the intent library.

[0039] The intent engine module is used to extract key information from the input text intent through natural language processing such as named entity recognition, translate the intent into intent tuples through technologies such as finite state machines, and feed the translation results back to the front-end page to form intent name-value pairs for user confirmation;

[0040] The intention experience value input module is used to input technical intentions through the point-and-click front-end page;

[0041] The intent library is used to store non-technical intents verified by feedback and non-technical intents input by the front-end page;

[0042] The front-end page module is used to provide the intent input page, intent network topology, and intent translation results.

[0043] Combining all the above technical solutions, the advantages and positive effects of the present invention are as follows: The present invention creates an intent translation system for information communication networks, which can be used for non-technical intent translation for end users who do not pay attention to the details of the underlying network, and can also be used for technical intent translation for professionals such as network administrators and service operators for specific network technologies. The intent translation front-end page is designed to provide a text-based intent input page for non-technical intents and a point-and-click intent input page for technical intents.

[0044] The comparison with the prior art is shown in Table 1.

[0045] Table 1

[0046] BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following is a brief introduction to the drawings required for use in the embodiments of the present application. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0048] Figure 1 It is a flow chart of the method for translating the intention of an information communication network provided by an embodiment of the present invention.

[0049] Figure 2 It is a structural diagram of an intention translation system of an information communication network provided by an embodiment of the present invention;

[0050] Figure 2 In: 1. User intention input module; 2. Intent engine module; 3. Intent experience value input module; 4. Intent library; 5. Front-end page module.

[0051] Figure 3 It is a structural diagram of an intention translation system for an information communication network provided by an embodiment of the present invention.

[0052] Figure 4 It is a schematic diagram of the framework of the intent-driven intelligent translation system of the information communication network provided by an embodiment of the present invention.

[0053] Figure 5 It is a schematic diagram of the main page of the intent-driven intelligent translation system for the information communication network provided by an embodiment of the present invention.

[0054] Figure 6 It is a schematic diagram of a new intent page of the intent-driven intelligent translation system of the information communication network provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0055] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0056] In view of the problems existing in the prior art, the present invention provides an information communication network intent translation method, system, medium and interactive interface. The present invention is described in detail below with reference to the accompanying drawings.

[0057] like Figure 1 As shown, the method for translating the intention of the information communication network provided by the present invention comprises the following steps:

[0058] S101: Obtain non-technical intent input through the front-end page, send it to the intent engine for translation, and send the translation result to the feedback page for confirmation;

[0059] S102: Obtain technical intent input through the front-end page and store it in the intent library. At the same time, interact with the intent engine through the learning model to improve the real-time, adaptability and accuracy of the intent experience value.

[0060] S103: The intent engine feeds back the intent translation result to the front-end page and supports transmission to the strategy mapping module in JSON format.

[0061] The method for translating the intention of the information communication network provided by the present invention can also be implemented by other steps by ordinary technicians in the industry. Figure 1 The information communication network intention translation method provided by the present invention is only a specific embodiment.

[0062] like Figure 2 As shown, the information communication network intention translation system provided by the present invention includes:

[0063] User intention input module 1, used to implement non-technical intentions, input the original intention through the input intention text box of the front-end page, transmit it to the intention engine, and store the translation result confirmed by the user in the intention library;

[0064] Intent engine module 2 is used to extract key information of the input text intent through natural language processing such as named entity recognition, translate the intent into an intent tuple through technologies such as finite state machines, and feed back the translation results to the front-end page to form an intent name-value pair for user confirmation;

[0065] Intention experience value input module 3, used for technical intention to input intention through the point-and-click front-end page;

[0066] Intent library 4, used to store non-technical intents verified by feedback and non-technical intents input by the front-end page;

[0067] The front-end page module 5 is used to provide the intent input page, the intent network topology structure and the intent translation result.

[0068] The technical solution of the present invention is further described below in conjunction with the accompanying drawings.

[0069] like Figure 3 As shown, the information communication network intention translation system provided by the embodiment of the present invention includes:

[0070] User input intent, for non-technical intents, the original intent is entered through the input intent text box on the front-end page, and can be transmitted to the intent engine through the "Submit" button, and the user-confirmed translation result is stored in the intent library;

[0071] The intent engine is used to extract key information from the input text intent through natural language processing such as named entity recognition, translate the intent into intent tuples through technologies such as finite state machines, and feed the translation results back to the front-end page to form intent name-value pairs for user confirmation;

[0072] Intent experience value input, used for technical intent to input intent through the point-and-click front-end page;

[0073] The intent library is used to store non-technical intents verified by feedback and non-technical intents input by the front-end page;

[0074] The front-end page is used to provide the intent input page, intent network topology, and intent translation results.

[0075] The application principle of the present invention is further described below in conjunction with the accompanying drawings.

[0076] like Figure 4 As shown, the intention-driven intelligent translation system for an information communication network provided by an embodiment of the present invention includes the following steps:

[0077] Step 1: Front-end input

[0078] Enter the intention expressed in natural language in the text box through the front-end webpage, for example, "From Beijing user A to Nanjing user B, an important voice service will be transferred from 20:00 on October 21, 2020 to 20:36 on October 21, 2020.", "From Room 427 of Xi'an Science and Technology Building and No. 432 of Zhongshan West Road, Shijiazhuang, an ordinary data service will be transferred from 19:13 on July 4, 2034 to 21:30 on July 4, 2034.", "From No. 2 Taibai South Road, Xi'an to No. 12 of Section A, Shijiazhuang, an important video service will be built from 19:13 on July 4, 2034 to 21:30 on July 4, 2034." That is, users can enter various forms of irregular intentions on the front end.

[0079] The input intent is sent to the intent engine in the form of an HTTP request, and the front-end web page is connected to the intent engine with the help of the Django framework, and the input intent is stored in the database.

[0080] Step 2: Intent engine processing

[0081] Constructing intention Figure 5The tuple <domain, attribute, object, operation, result> stores the key indicators of intent respectively. For example, in a multi-domain network scenario, <domain> includes: chain service, virtual network service, etc.; <attribute> includes: bandwidth, latency, priority, time requirement, etc.; <object> includes: source node, destination node, link, etc.; <operation> includes: create link, adjust link, create virtual network link, virtual network expansion, etc.; <result> includes: success, deletion, etc.

[0082] Through named entity recognition technology, valid entities in the intent described in natural language are extracted, and finite state machines are used to define network state behavior labels. Figure 5 Tuples, forming standard expressions.

[0083] Step 3: Translate the intent to store

[0084] The translated intent tuple is stored in the database so that the data information can be called on the front-end page, and intents can be added, deleted, modified, and viewed in an orderly manner in the database, supporting the addition, deletion, modification, and query of intents.

[0085] Step 4: Send translation intent

[0086] The translated intent is sent to the integrated control system in the form of an intent tuple, guiding the network to generate executable strategies and sending the generated specific device configuration information to the network element devices of the underlying network for execution.

[0087] like Figure 5 As shown, the main page design of the intent-driven intelligent translation system for the information communication network provided by the embodiment of the present invention is as follows:

[0088] Click the "Create Intent" button to enter the Create Intent page; select an intent and click the "Modify Intent" button to enter the Modify Intent page; select an intent and click the "Delete Intent" button to delete the intent. Click the "Translation Result" button to display the translation result, which is a list of name-value pairs in a floating table frame or other ways. Click the "Orchestration Result" button, "Unsubmitted" is gray, "Success" is green, and "Failure" is red. Click the green button to enter the comprehensive situation single business presentation page.

[0089] like Figure 6 As shown, the new intent design of the intent-driven intelligent translation system of the information communication network provided by the embodiment of the present invention is as follows:

[0090] Click the "Intent Translation" button to display the translation results. After the translation is successful, the "Submit Orchestration" button is enabled. After clicking the "Submit Orchestration" button, switch to the "Business Orchestration" page.

[0091] For example, in response to the slow opening of cross-domain chained services and inefficient resource allocation, we designed an intent-driven inter-domain service resource allocation process and method to obtain network intent and current network status, and fairly allocate network resources while ensuring service priority. The designed intent-driven intelligent translation system can realize the intention of establishing cross-domain chained services and adjusting the intention of chained services, and can be extended to other service types.

[0092] For the establishment of a multi-domain network chain intent "dispatch an important level data service from Beijing user C to Nanjing user D, the time requirement is 2020-09-12-10:30 to 2020-09-12-12:30.", after being processed by the intent translation system, the intent tuple is fed back to the front-end web page, and the intent tuple is displayed to the user in the form of a name-value pair, as shown in Figure x, and can be converted into a JSON data format and sent to other modules for intent implementation.

[0093]

[0094]

[0095] For the intention of adjusting the chain business "adjust an important-level data business from Beijing user C to Nanjing user D, the time requirement is 2020-09-12-10:30 to 2020-09-12-12:30. Adjust the importance level to normal", the intention translation system processes the adjusted intent data and feeds it back to the front-end web page. The intent tuple is displayed to the user in the form of a name-value pair, and can be converted into a JSON data format and sent to other modules for intent implementation.

[0096]

[0097] In addition, the technical intent input interface allows operation and maintenance personnel with professional knowledge to store expert experience values ​​in the intent library through a point-and-click page. The intent-driven network topology established by the current intent can be created and displayed on the front-end topology page through the current open source SDN controller platform Opendaylight.

[0098] It should be noted that the embodiments of the present invention can be implemented by hardware, software, or a combination of software and hardware. The hardware part can be implemented using dedicated logic; the software part can be stored in a memory and executed by an appropriate instruction execution system, such as a microprocessor or dedicated design hardware. It can be understood by a person of ordinary skill in the art that the above-mentioned devices and methods can be implemented using computer executable instructions and / or contained in a processor control code, such as a carrier medium such as a disk, CD or DVD-ROM, a programmable memory such as a read-only memory (firmware), or a data carrier such as an optical or electronic signal carrier. Such code is provided on the carrier medium. The device and its modules of the present invention can be implemented by hardware circuits such as very large-scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, etc., or programmable hardware devices such as field programmable gate arrays, programmable logic devices, etc., can also be implemented by software executed by various types of processors, and can also be implemented by a combination of the above-mentioned hardware circuits and software, such as firmware.

[0099] The above description is only a specific implementation mode of the present invention, but the protection scope of the present invention is not limited thereto. Any modifications, equivalent substitutions and improvements made by any technician familiar with the technical field within the technical scope disclosed by the present invention and within the spirit and principle of the present invention should be covered by the protection scope of the present invention.

Claims

1. A method for translating intent in an information communication network, characterized in that: The information communication network intention translation method comprises: Users input non-technical intents in natural language through the front-end web page, and send the intents to the intent engine module for entity recognition and translation processing; The intent engine module feeds back the intent translation result to the front-end web page, and stores the intent confirmed by the user to be accurately translated into the intent library; The experience value of technical intent can be sent to the intent library through the front-end page, and the experience value is updated through the interaction between the learning model and the intent engine; The translation results of the intent engine module are fed back to the front-end page; The front-end web page contains input examples and input intent text boxes. The user enters the original intent in the text box based on the input example, and the input intent text is sent to the intent translation engine in the form of an HTTP request. The entity recognition of the intent translation system of the information communication network is to mark the text in the input intent corpus with BMEO tags, and then accurately identify the key information of the intent through named entity recognition; Define network state behavior labels, characterize network states, generate state transition diagrams with the help of finite state machine training, and realize intent translation.

2. The method for translating the intention of an information communication network as claimed in claim 1, wherein: The intent translation result feedback page of the intent translation system of the information communication network is displayed as an intent name-value pair, and the intent translation result is stored in the intent library constructed by the intent translation engine, using the relational database PostgreSQL for storage.

3. The method for translating the intention of an information communication network as claimed in claim 1, wherein: The technical intent of the intent translation system of the information communication network can be input through a point-and-click page, and the key parameters of the intent are stored in the intent library; Build a learning model to train the input technical intent and transmit it to the intent engine for translation confirmation.

4. The method for translating the intention of an information communication network as claimed in claim 1, wherein: The technical intent translation result of the intent translation system of the information communication network is sent to the feedback page and displayed as an intent name-value pair.

5. A computer device, characterized in that: The computer device includes a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor performs the following steps: Users input non-technical intents through the front-end web page, and the intents are sent to the intent engine module for entity recognition and translation processing; The intent engine module feeds back the intent translation result to the front-end web page, and stores the intent confirmed by the user to be accurately translated into the intent library; The experience value of technical intent can be sent to the intent library through the front-end page, and the experience value is updated through the interaction between the learning model and the intent engine; The results of the translation by the intent engine module are fed back to the front-end page.

6. A computer-readable storage medium storing a computer program, wherein when the computer program is executed by a processor, the processor executes the following steps: Users input non-technical intents through the front-end web page, and the intents are sent to the intent engine module for entity recognition and translation processing; The intent engine module feeds back the intent translation result to the front-end web page, and stores the intent confirmed by the user to be accurately translated into the intent library; The experience value of technical intent can be sent to the intent library through the front-end page, and the experience value is updated through the interaction between the learning model and the intent engine; The results of the translation by the intent engine module are fed back to the front-end page.

7. An interactive interface information data processing terminal, characterized in that: The interactive interface information data processing terminal is used to implement the information communication network intention translation method described in any one of claims 1 to 4.

8. An information communication network intention translation system implementing the information communication network intention translation method according to any one of claims 1 to 4, characterized in that: The information communication network intention translation system includes: The user intent input module is used to implement non-technical intents. The original intent can be input through the input intent text box on the front-end page. The original intent can be described in an irregular natural language form and transmitted to the intent engine. The translation result confirmed by the user is stored in the intent library. The intent engine module is used to extract key information from the input text intent through natural language processing such as named entity recognition, translate the intent into intent tuples through technologies such as finite state machines, and feed the translation results back to the front-end page to form intent name-value pairs for user confirmation; The intention experience value input module is used to input technical intentions through the point-and-click front-end page; The intent library is used to store non-technical intents verified by feedback and non-technical intents input by the front-end page; The front-end page module is used to provide the intent input page, intent network topology, and intent translation results.

Images