Live-broadcast monitoring processing method and system, and device and medium

Abstract

Disclosed in the present invention are a live-broadcast monitoring processing method and system, and a device and a medium. The method comprises: performing data collection and aggregation processing of a whole live-broadcast process, so as to obtain monitoring source data, wherein the monitoring source data comprises monitoring data and historical data; using the monitoring source data as a training data set for model training processing, so as to obtain a large language model; performing anomaly detection processing on the monitoring source data by means of the large language model, so as to obtain a fault result; and performing real-time search and analysis processing on the fault result by means of combining the large language model and an internal search engine, so as to obtain an alarm handling scheme, wherein the alarm handling scheme comprises alarm trend information and a solution. By means of applying a large voice model and an internal search engine to a real-time monitoring system, and quickly processing and querying data, the embodiments of the present invention realize timely discovery and processing of anomalous data, and can be widely applied to the technical field of operation and maintenance monitoring.

Description

Live broadcast monitoring processing method, system, device and medium Technical Field

[0001] The present invention relates to the field of operation and maintenance monitoring technology, and in particular to a live broadcast monitoring processing method, system, equipment and medium. Background Art

[0002] As live streaming services continue to develop, the complexity and scale of systems continue to increase. The process for handling system anomalies has become a major constraint on stable system operation. Various anomalies can occur during live streaming, such as server failures, user experience issues, and platform operation anomalies, negatively impacting user experience and service quality. Therefore, the technical issues currently existing in related technologies urgently need to be addressed. Summary of the Invention

[0003] In view of this, embodiments of the present invention provide a live broadcast monitoring processing method, system, device and medium to improve the efficiency of detecting and processing abnormal situations.

[0004] In one aspect, the present invention provides a live broadcast monitoring processing method, the method comprising:

[0005] Collect and aggregate data from the entire live broadcast process to obtain monitoring source data, which includes monitoring data and historical data;

[0006] Using the monitoring source data as a training data set to perform model training processing to obtain a large language model;

[0007] Performing anomaly detection processing on the monitoring source data using the large language model to obtain a fault result;

[0008] The large language model is combined with an internal search engine to perform real-time search and analysis on the fault results to obtain an alarm processing solution, which includes alarm trend information and solutions.

[0009] Optionally, the method further includes:

[0010] Record and archive the alarm handling solutions to establish a historical database and knowledge sharing library;

[0011] Updating the training data set according to the historical database and the knowledge sharing library;

[0012] The updated training data set is input into the large language model for iterative updating.

[0013] Optionally, the data collection and aggregation process for the entire live broadcast process to obtain monitoring source data includes:

[0014] Collect and process data from the server node to obtain service monitoring data;

[0015] Collect and process data from client nodes to obtain real-time perception data;

[0016] Synchronizing the service monitoring data and the real-time sensing data to a convergence node to obtain monitoring data;

[0017] Acquire historical data, aggregate the historical data with the monitoring data, and obtain monitoring source data.

[0018] Optionally, the using the monitoring source data as a training data set to perform model training processing to obtain a large language model includes:

[0019] Performing data cleaning on the monitoring data in the monitoring source data and increasing the weight of the monitoring data to obtain a training data set;

[0020] Parameters of the pre-trained model are adjusted according to the training data set to obtain a large language model.

[0021] Optionally, performing anomaly detection processing on the monitoring source data using the large language model to obtain a fault result includes:

[0022] Acquiring real-time monitoring data from the monitoring source data;

[0023] Converting the monitoring data to obtain a natural language text description;

[0024] The large language model is used to analyze and process the working status of the natural language text description, and a fault judgment is performed in combination with the scenario to obtain a fault result.

[0025] Optionally, combining the large language model with an internal search engine to perform real-time search and analysis on the fault results to obtain an alarm processing solution includes:

[0026] Perform trend query processing on the fault results through the internal search engine to obtain alarm trend information;

[0027] The large language model is used to perform root cause analysis and suggest processing on the fault result to obtain a solution.

[0028] On the other hand, an embodiment of the present invention further provides a live broadcast monitoring processing system, the system comprising:

[0029] The data acquisition, transmission and storage module is used to collect and aggregate data for the entire live broadcast process to obtain monitoring source data, which includes monitoring data and historical data;

[0030] A large model training module is used to perform model training processing on the monitoring source data as a training data set to obtain a large language model;

[0031] A monitoring anomaly detection module, configured to perform anomaly detection processing on the monitoring source data using the large language model to obtain a fault result;

[0032] The alarm analysis and repair module is used to combine the large language model with the internal search engine to perform real-time search and analysis on the fault results to obtain an alarm processing solution, which includes alarm trend information and solutions.

[0033] Optionally, the system further comprises:

[0034] The exception handling recording module is used to record and archive the alarm handling solution and establish a historical database and a knowledge sharing library; the historical database and the knowledge sharing library are used to iteratively update the large language model.

[0035] Optionally, the data acquisition, transmission and storage module is used to collect and aggregate data for the entire live broadcast process to obtain monitoring source data, including:

[0036] Collect and process data from the server node to obtain service monitoring data;

[0037] Collect and process data from client nodes to obtain real-time perception data;

[0038] Synchronizing the service monitoring data and the real-time sensing data to a convergence node to obtain monitoring data;

[0039] Acquire historical data, aggregate the historical data with the monitoring data, and obtain monitoring source data.

[0040] Optionally, the large model training module is configured to perform model training processing on the monitoring source data as a training data set to obtain a large language model, including:

[0041] Performing data cleaning on the monitoring data in the monitoring source data and increasing the weight of the monitoring data to obtain a training data set;

[0042] Parameters of the pre-trained model are adjusted according to the training data set to obtain a large language model.

[0043] Optionally, the monitoring anomaly detection module is configured to perform anomaly detection processing on the monitoring source data using the large language model to obtain a fault result, including:

[0044] Acquiring real-time monitoring data from the monitoring source data;

[0045] Converting the monitoring data to obtain a natural language text description;

[0046] The large language model is used to analyze and process the working status of the natural language text description, and a fault judgment is performed in combination with the scenario to obtain a fault result.

[0047] Optionally, the alarm analysis and repair module is configured to combine the large language model with an internal search engine to perform real-time search and analysis on the fault results to obtain an alarm handling solution, including:

[0048] Perform trend query processing on the fault results through the internal search engine to obtain alarm trend information;

[0049] The large language model is used to perform root cause analysis and suggest processing on the fault result to obtain a solution.

[0050] On the other hand, an embodiment of the present invention further discloses an electronic device, including a processor and a memory;

[0051] The memory is used to store programs;

[0052] The processor executes the program to implement the method described above.

[0053] On the other hand, an embodiment of the present invention further discloses a computer-readable storage medium, wherein the storage medium stores a program, and the program is executed by a processor to implement the method described above.

[0054] In another aspect, embodiments of the present invention further disclose a computer program product or computer program, which includes computer instructions stored in a computer-readable storage medium. A processor of a computer device can read the computer instructions from the computer-readable storage medium and execute the computer instructions, causing the computer device to perform the aforementioned method.

[0055] Compared with the prior art, the present invention adopts the above technical solution and has the following technical effects:

[0056] The present invention provides a live broadcast monitoring and processing method, including: collecting and aggregating data for the entire live broadcast process to obtain monitoring source data, which includes monitoring data and historical data; using the monitoring source data as a training data set to perform model training processing to obtain a large language model; performing anomaly detection processing on the monitoring source data through the large language model to obtain fault results; combining the large language model with an internal search engine to perform real-time search and analysis processing on the fault results to obtain an alarm processing solution, which includes alarm trend information and solutions.

[0057] The present invention provides a live broadcast monitoring and processing method, which performs abnormal monitoring and processing on the entire live broadcast process based on a large language model and an internal search engine. It can monitor the system operation in real time by utilizing artificial intelligence and big data technology, quickly discover abnormalities and accurately locate the problem, and obtain historical trends of data through an internal search engine. At the same time, combined with the advantages of a large language model, it is possible to quickly derive current alarm trends, related solutions and repair methods, and realize automated processing and repair of system abnormalities. The present invention can greatly improve the accuracy and efficiency of system monitoring exception processing, reduce system maintenance costs, and enhance the performance and availability of end-to-end live broadcast services in the visual network, thereby ensuring the stable operation and high availability of the visual network. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0059] FIG1 is a flow chart of a live broadcast monitoring processing method provided by an embodiment of the present invention;

[0060] FIG2 is a flow chart of an implementation of step S101 in FIG1 ;

[0061] FIG3 is a schematic diagram of a node of an end-to-end live broadcast provided by an embodiment of the present invention;

[0062] FIG4 is a flow chart of an implementation of step S102 in FIG1 ;

[0063] FIG5 is a flow chart of an implementation of step S103 in FIG1 ;

[0064] FIG6 is a flow chart of an implementation of step S104 in FIG1 ;

[0065] 7 is a schematic structural diagram of a live broadcast monitoring processing system provided by an embodiment of the present invention;

[0066] FIG8 is a processing flow chart of a live broadcast monitoring processing system provided by an embodiment of the present invention;

[0067] FIG9 is a schematic structural diagram of an electronic device provided by an embodiment of the present invention.

[0068] FIG10 is a schematic structural diagram of a storage medium provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0069] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0070] First, some terms involved in this invention are analyzed:

[0071] Artificial Intelligence (AI) is a field of technology and science that simulates human intelligence. It encompasses many different subfields, including machine learning, natural language processing, computer vision, and expert systems. The goal of AI is to enable computers to make intelligent decisions, learn, and solve problems in a human-like manner. Its technical foundations include learning and model training using large amounts of data, automated reasoning and decision-making, pattern recognition, and semantic understanding. Through these technologies, AI can accomplish a variety of tasks, including image recognition, speech recognition, natural language processing, and intelligent recommendations.

[0072] Training a model (TM) involves using an existing dataset to train a machine learning model, enabling it to learn features and patterns from the data. The model training process typically involves inputting data into the model, calculating the model's output, comparing it with the actual labels, and adjusting the model's weights and parameters based on the comparison results to gradually optimize the model. Generally, model training requires a large amount of labeled data and repeated training in an iterative process to improve the model's accuracy and generalization capabilities.

[0073] Large language models (LLMs), also known as large language models or large models, are artificial intelligence models designed to understand and generate human language. They are trained on large amounts of text data and can perform a wide range of tasks, including text summarization, translation, sentiment analysis, and more. Large language models are characterized by their massive size, containing billions of parameters, which help them learn complex patterns in language data. These models are often based on deep learning architectures such as transformers, which contributes to their impressive performance on various natural language processing (NLP) tasks.

[0074] The live broadcast service of the Visual Internet is a live broadcast service based on the Visual Internet. The Visual Internet is a comprehensive video service network that integrates cloud-network resources, AI and other capabilities to achieve multi-brand video terminal access, video scheduling, cloud storage and AI application services.

[0075] In related technologies, various abnormal situations may occur during live broadcasts, such as server failures, user experience issues, and platform operation abnormalities, which have a negative impact on user experience and service quality.

[0076] In view of this, an embodiment of the present invention provides a method for processing live broadcast monitoring. The processing method in the embodiment of the present invention can be applied to a terminal or a server, or can be software running on a terminal or a server. The terminal can be a tablet computer, a laptop computer, a desktop computer, etc., but is not limited to such. The server can be an independent physical server, a server cluster or a distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms.

[0077] 1 , an embodiment of the present invention provides a live broadcast monitoring processing method, the method comprising:

[0078] S101, collecting and aggregating data for the entire live broadcast process to obtain monitoring source data, wherein the monitoring source data includes monitoring data and historical data;

[0079] S102: Using the monitoring source data as a training data set to perform model training to obtain a large language model;

[0080] S103, performing anomaly detection processing on the monitoring source data using the large language model to obtain a fault result;

[0081] S104: Combine the large language model with an internal search engine to perform real-time search and analysis on the fault results to obtain an alarm processing solution, which includes alarm trend information and solutions.

[0082] In an embodiment of the present invention, data is collected for the entire live broadcast process to obtain monitoring data and historical data, which are then aggregated to obtain monitoring source data. The monitoring source data is then input into the model as a training data set for training to obtain a large language model. Then, the large language model is used to perform anomaly detection processing on the real-time monitoring data in the monitoring source data to obtain the fault result. Finally, the large language model is combined with the internal search engine to search and analyze the fault results to obtain an alarm processing solution. The embodiment of the present invention utilizes artificial intelligence and big data technology to monitor the system operation in real time, quickly discover anomalies and accurately locate the problem, and obtain the historical trend of the data through the internal search engine. The embodiment of the present invention can greatly improve the accuracy and efficiency of system monitoring exception processing, reduce system maintenance costs, and improve the performance and availability of the end-to-end live broadcast service of the visual network, thereby ensuring the stable operation and high availability of the visual network.

[0083] It should be noted that, in various specific embodiments of the present invention, when it comes to the need to perform relevant processing based on the target object's information, the target object's behavioral data, the target object's historical data, the target object's location information, and other data related to the target object's identity or characteristics, the target object's permission or consent will be obtained first, and the collection, use, and processing of these data will comply with relevant laws, regulations, and standards. In addition, when the embodiment of the present invention needs to obtain sensitive information of the target object, it will obtain the target object's separate permission or consent through a pop-up window or jump to a confirmation page. After clearly obtaining the target object's separate permission or consent, the necessary target object-related data for the embodiment of the present invention to enable normal operation will be obtained.

[0084] 1 , as an optional embodiment, the method further includes:

[0085] S105: Record and archive the alarm processing plan to establish a historical database and knowledge sharing library;

[0086] S106, updating the training data set according to the historical database and the knowledge sharing library;

[0087] S107: Input the updated training data set into the large language model for iterative updating.

[0088] In an embodiment of the present invention, after detecting an anomaly at a certain moment or a certain time, the generated alarm processing solution is recorded and archived, thereby establishing a historical database and a knowledge sharing library, wherein the historical database includes historical alarm information, and the knowledge sharing library includes historical solutions. The historical database and the knowledge sharing library are used to iteratively update the large language model. After the historical database and the knowledge sharing library for exception processing are established, the training data set can be updated, and the updated training data set can be input into the large language model for iterative update processing. By establishing a historical database and a knowledge sharing library, the embodiment of the present invention can obtain historical alarm information through the historical database to train the large language model, and can also obtain previous solutions to the same or similar fault results through the knowledge sharing library, which can provide a reference basis for subsequent exception processing and further improve the processing efficiency of live broadcast monitoring.

[0089] 2 , as an optional embodiment, in step S101 above, the data collection and aggregation process for the entire live broadcast process to obtain monitoring source data includes:

[0090] S201, collect and process data from the server node to obtain service monitoring data;

[0091] S202: Collect and process data from the client node to obtain real-time perception data;

[0092] S203, synchronizing the service monitoring data and the real-time sensing data to a sink node to obtain monitoring data;

[0093] S204: Acquire historical data, and aggregate the historical data with the monitoring data to obtain monitoring source data.

[0094] In an embodiment of the present invention, referring to FIG3 , the node distribution of the end-to-end live broadcast service based on the visual network includes a server node, a client node and an aggregation node. The server node is deployed in the cloud and may include a server, network equipment and (Platform as a Service) PAAS components, (Software as a Service) SAAS services, etc. The server node is used to monitor and collect data on servers, network equipment, components, services, applications, etc. in the server. The client node is deployed on a terminal, which may be a tablet computer, a laptop computer, a desktop computer, etc. The client node is used to collect and acquire real-time perception data from client-side buried points, and then synchronize the data collected from the server and the client to the aggregation node of the monitoring system to obtain monitoring data. Historical data is also collected, and the historical data is aggregated with the monitoring data to obtain monitoring source data. Among them, historical data includes historical alarms and fault information. In an embodiment of the present invention, the source data of the live broadcast monitoring anomaly detection discovery and processing process is aggregated by performing data cleaning on operation logs and historical work orders, as well as knowledge base construction and fault reports, to obtain monitoring source data. The embodiment of the present invention aggregates and processes the monitoring data and historical data of the server and client, thereby providing a data basis for subsequent abnormal monitoring and detection.

[0095] 4 , as an optional embodiment, in step S102 , the monitoring source data is used as a training data set to perform model training to obtain a large language model, including:

[0096] S401, performing data cleaning processing on the monitoring data in the monitoring source data, and increasing the weight of the monitoring data to obtain a training data set;

[0097] S402: Adjust parameters of the pre-trained model according to the training data set to obtain a large language model.

[0098] In an embodiment of the present invention, the monitoring data of the convergence nodes in the monitoring system used for training are cleaned, and the weight of the monitoring data in the fault time period is increased, so that the large language model pays more attention to the analysis and processing of the monitoring data during the data training process. In an embodiment of the present invention, a pre-trained GPT2-xlarge model is used as a large language model for data processing, wherein the GPT2-xlarge model is a large language model for Chinese text processing, which is pre-trained by the multimodal pre-training framework TencentPretrain. In an embodiment of the present invention, the model is trained by using the monitoring source data as a training data set, wherein the monitoring source data includes monitoring data, historical alarm information, alarm processing flow, knowledge base, alarm dispatch information, fault report and other data. The large language model is verified by adjusting the learning rate, number of layers, number of hidden units, number of attention heads, and batch size, and then released to the production environment to obtain a fine-tuned large language model.

[0099] 5 , as an optional embodiment, in step S103 , performing anomaly detection processing on the monitoring source data using the large language model to obtain a fault result includes:

[0100] S501, obtaining real-time monitoring data from the monitoring source data;

[0101] S502: Convert the monitoring data to obtain a natural language text description;

[0102] S503: Perform working status analysis and processing on the natural language text description using the large language model, and perform fault judgment based on the scenario to obtain a fault result.

[0103] In an embodiment of the present invention, real-time monitoring data is obtained from monitoring source data. This monitoring data includes real-time stream pull counts, successful stream pull counts, stream forwarding information, hole punching success rates, CPU and load information, memory usage, and other information. This embodiment of the present invention converts the monitoring data into text to obtain a natural language text description. This data is then analyzed using a large language model to determine the root cause, impact, and recommended handling steps for the fault. The large language model can analyze whether the current state of the device or service is operating normally and, based on the current scenario, determine whether a fault exists. The fault result is obtained. If a fault exists, the next step is taken. Otherwise, data without anomalies is added to the training dataset to train the large language model. This embodiment of the present invention applies the large language model to the monitoring and alarm field, utilizing a large language model trained using a deep learning neural network to train and learn monitoring data to extract data features and patterns, thereby enabling detection of abnormal data. This embodiment of the present invention can more accurately identify abnormal data, reduce false positives and false negatives, and improve the accuracy of anomaly detection.

[0104] 6 , as an optional embodiment, in step S104 , the large language model is combined with an internal search engine to perform real-time search and analysis on the fault results to obtain an alarm processing solution, including:

[0105] S601, performing trend query processing on the fault result through the internal search engine to obtain alarm trend information;

[0106] S602: Analyze the root cause of the fault and provide suggested processing for the fault result through the large language model to obtain a solution.

[0107] In an embodiment of the present invention, an internal search engine and a large language model are combined, and the internal search engine is used to query real-time data and historical trends. For example, when a server memory alarm occurs, the memory usage trend will be queried and a chart will be drawn for display. Based on the analysis conclusions of the large language model, the address of similar fault handling materials in the internal knowledge base is attached. The embodiment of the present invention has the ability to actively query real-time and historical monitoring data, and can synchronously push alarm trend information to form a visual chart when an alarm is generated, and call the large language model to generate an alarm processing method and basis, and provide solutions to operation and maintenance personnel to handle the alarm. By combining the large model with the internal search engine, the embodiment of the present invention utilizes the efficient real-time query capability and index structure of the search engine to actively query real-time monitoring data and historical trends for chart display and auxiliary analysis, thereby improving the reliability of alarm analysis.

[0108] On the other hand, referring to FIG. 7 , an embodiment of the present invention further provides a live broadcast monitoring processing system, the system comprising:

[0109] The data acquisition, transmission and storage module 701 is used to collect and aggregate data for the entire live broadcast process to obtain monitoring source data, which includes monitoring data and historical data;

[0110] A large model training module 702 is configured to perform model training processing on the monitoring source data as a training data set to obtain a large language model;

[0111] A monitoring anomaly detection module 703 is configured to perform anomaly detection processing on the monitoring source data using the large language model to obtain a fault result;

[0112] The alarm analysis and repair module 704 is used to combine the large language model with the internal search engine to perform real-time search and analysis on the fault results to obtain an alarm processing solution, which includes alarm trend information and solutions.

[0113] 7 , as a further preferred embodiment, the system further includes:

[0114] The exception handling recording module 705 is used to record and archive the alarm handling solution and establish a historical database and a knowledge sharing library; the historical database and the knowledge sharing library are used to iteratively update the large language model.

[0115] As an optional embodiment, the data acquisition, transmission and storage module is used to collect and aggregate data for the entire live broadcast process to obtain monitoring source data, including:

[0116] Collect and process data from the server node to obtain service monitoring data;

[0117] Collect and process data from client nodes to obtain real-time perception data;

[0118] Synchronizing the service monitoring data and the real-time sensing data to a convergence node to obtain monitoring data;

[0119] Acquire historical data, aggregate the historical data with the monitoring data, and obtain monitoring source data.

[0120] As a further optional implementation, the large model training module is used to perform model training processing on the monitoring source data as a training data set to obtain a large language model, including:

[0121] Performing data cleaning on the monitoring data in the monitoring source data and increasing the weight of the monitoring data to obtain a training data set;

[0122] Parameters of the pre-trained model are adjusted according to the training data set to obtain a large language model.

[0123] As a further optional implementation, the monitoring anomaly detection module is configured to perform anomaly detection processing on the monitoring source data using the large language model to obtain a fault result, including:

[0124] Acquiring real-time monitoring data from the monitoring source data;

[0125] Converting the monitoring data to obtain a natural language text description;

[0126] The large language model is used to analyze and process the working status of the natural language text description, and a fault judgment is performed in combination with the scenario to obtain a fault result.

[0127] As a further optional implementation, the alarm analysis and repair module is used to combine the large language model with an internal search engine to perform real-time search and analysis on the fault results to obtain an alarm handling solution, including:

[0128] Perform trend query processing on the fault results through the internal search engine to obtain alarm trend information;

[0129] The large language model is used to perform root cause analysis and suggest processing on the fault result to obtain a solution.

[0130] It can be understood that the contents of the above method embodiments are all applicable to the present system embodiments, the functions specifically implemented by the present system embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0131] Referring to Figure 8 , the process of the present invention specifically includes: collecting and aggregating data to obtain monitoring source data through the monitoring data acquisition, transmission, and storage module; inputting the monitoring source data into the large model training model to train the large language model; adding the trained large language model to the monitoring anomaly detection module; and inputting the monitoring source data into the monitoring anomaly detection module through the monitoring data acquisition, transmission, and storage module for anomaly detection. The monitoring anomaly detection module determines whether an anomaly exists. If no anomaly exists, the data is added to the dataset to update the large language model; if an anomaly exists, the alarm analysis and repair module is invoked. The alarm analysis and repair module combines the large language model with an internal search engine to analyze the anomaly and obtain an alarm handling solution. The alarm handling solution for this detection is recorded by the anomaly handling recording module, and the recorded data is added to the training set to iteratively update the large language model. This embodiment of the present invention applies the large model and internal search engine to a real-time monitoring system to rapidly process and query data, enabling the timely detection and handling of abnormal data. This embodiment of the present invention, by combining the large speech model with the internal search engine, enables intelligent anomaly detection, automatically discovers abnormal patterns and trends, reduces manual intervention, and improves detection accuracy and efficiency. The embodiments of the present invention can quickly process large amounts of monitoring data, improve monitoring efficiency, and reduce resource waste. By fully utilizing the parallel computing capabilities of large models and internal search engines, it can also process large-scale monitoring data. It is suitable for large-scale systems and complex network environments, including business scenarios such as live broadcasts.

[0132] 9 , an embodiment of the present invention further provides an electronic device, including a processor 901 and a memory 902 ; the memory 902 is used to store programs; and the processor 901 executes the programs to implement the method described above.

[0133] 10 , an embodiment of the present invention further provides a computer-readable storage medium 1001 , wherein the storage medium 1001 stores a program 1002 , and the program 1002 is executed by a processor to implement the method described above.

[0134] Embodiments of the present invention further disclose a computer program product or computer program, which includes computer instructions stored in a computer-readable storage medium. A processor of a computer device can read the computer instructions from the computer-readable storage medium and execute the computer instructions, causing the computer device to perform the method shown in FIG1 .

[0135] In some optional embodiments, the function / operation mentioned in the block diagram may not occur in the order mentioned in the operation diagram. For example, depending on the function / operation involved, the two boxes shown in succession can actually be executed substantially simultaneously or the boxes can sometimes be executed in reverse order. In addition, the embodiment presented and described in the flow chart of the present invention is provided in an exemplary manner for the purpose of providing a more comprehensive understanding of the technology. The disclosed method is not limited to the operation and logic flow presented herein. Optional embodiments are contemplated in which the order of the various operations is changed and the sub-operations described as a part of a larger operation are performed independently.

[0136] Furthermore, although the present invention is described in the context of functional modules, it should be understood that, unless otherwise indicated, one or more of the functions and / or features described may be integrated into a single physical device and / or software module, or one or more functions and / or features may be implemented in separate physical devices or software modules. It will also be understood that a detailed discussion of the actual implementation of each module is not necessary for understanding the present invention. More specifically, given the properties, functions, and internal relationships of the various functional modules in the devices disclosed herein, the actual implementation of the module will be understood within the ordinary skill of an engineer. Therefore, a person skilled in the art using ordinary skill will be able to implement the present invention set forth in the claims without undue experimentation. It will also be understood that the specific concepts disclosed are merely illustrative and are not intended to limit the scope of the present invention, which is determined by the full scope of the appended claims and their equivalents.

[0137] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0138] The logic and / or steps represented in the flowcharts or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing the logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (e.g., a computer-based system, a system including a processor, or other system that can fetch and execute instructions from an instruction execution system, apparatus, or device). For purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0139] More specific examples (a non-exhaustive list) of computer-readable media include the following: an electrical connection with one or more wires (electronic devices), a portable computer disk cartridge (magnetic devices), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), a fiber optic device, and a portable compact disc read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering, or processing in another suitable manner as necessary, and then stored in a computer memory.

[0140] It should be understood that various parts of the present invention can be implemented using hardware, software, firmware, or a combination thereof. In the above-described embodiments, multiple steps or methods can be implemented using software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented using hardware, as in another embodiment, any one of the following technologies known in the art or a combination thereof can be used: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application-specific integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0141] Throughout this specification, reference to terms such as "one embodiment," "some embodiments," "examples," "specific examples," or "some examples" means that a specific feature, structure, material, or characteristic described in conjunction with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, schematic representations of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.

[0142] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to the embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the claims and their equivalents.

[0143] The above is a specific description of the preferred implementation of the present invention, but the present invention is not limited to the embodiments. Those skilled in the art can make various equivalent modifications or substitutions without violating the spirit of the present invention. These equivalent modifications or substitutions are all included in the scope defined by the claims of the present invention.

Claims

1. A live broadcast monitoring processing method, characterized in that: The method comprises: Collect and aggregate data for the entire live broadcast process to obtain monitoring source data, which includes monitoring data and historical data; Using the monitoring source data as a training data set to perform model training processing to obtain a large language model; Performing anomaly detection processing on the monitoring source data through the large language model to obtain a fault result; The large language model is combined with an internal search engine to perform real-time search and analysis on the fault results to obtain an alarm processing solution, which includes alarm trend information and solutions.

2. The method according to claim 1, characterized in that: The method further comprises: Record and archive the alarm processing plan to establish a historical database and knowledge sharing library; Updating the training data set according to the historical database and the knowledge sharing library; The updated training data set is input into the large language model for iterative updating.

3. The method according to claim 1, characterized in that The data collection and aggregation process of the whole live broadcast process is performed to obtain monitoring source data, including: Collect and process data from the server node to obtain service monitoring data; Collect and process data from client nodes to obtain real-time perception data; Synchronize the service monitoring data and the real-time sensing data to a convergence node to obtain monitoring data; Acquire historical data, aggregate the historical data with the monitoring data, and obtain monitoring source data.

4. The method according to claim 1, characterized in that The method of using the monitoring source data as a training data set to perform model training processing to obtain a large language model includes: Performing data cleaning processing on the monitoring data in the monitoring source data and increasing the weight of the monitoring data to obtain a training data set; The parameters of the pre-trained model are adjusted according to the training data set to obtain a large language model.

5. The method according to claim 1, characterized in that: The performing anomaly detection processing on the monitoring source data by using the large language model to obtain a fault result includes: Acquiring real-time monitoring data from the monitoring source data; Converting the monitoring data to obtain a natural language text description; The large language model is used to analyze and process the working status of the natural language text description, and a fault judgment is performed in combination with the scenario to obtain a fault result.

6. The method according to claim 1, characterized in that The large language model is combined with an internal search engine to perform real-time search and analysis on the fault results to obtain an alarm processing solution, including: Performing trend query processing on the fault results through the internal search engine to obtain alarm trend information; The large language model is used to perform root cause analysis and suggest processing on the fault result to obtain a solution.

7. A live broadcast monitoring and processing system, characterized in that: The system comprises: The data acquisition, transmission and storage module is used to collect and aggregate data for the entire live broadcast process to obtain monitoring source data, which includes monitoring data and historical data; A large model training module is used to perform model training processing on the monitoring source data as a training data set to obtain a large language model; A monitoring anomaly detection module, used to perform anomaly detection processing on the monitoring source data through the large language model to obtain a fault result; The alarm analysis and repair module is used to combine the large language model with the internal search engine to perform real-time search and analysis on the fault results to obtain an alarm processing solution, which includes alarm trend information and solutions.

8. The system according to claim 7, characterized in that The system further comprises: The exception handling recording module is used to record and archive the alarm handling solution and establish a historical database and a knowledge sharing library; the historical database and the knowledge sharing library are used to iteratively update the large language model.

9. An electronic device, characterized in that: The electronic device comprises a memory and a processor; The memory is used to store programs; The processor executes the program to implement the method according to any one of claims 1 to 6.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 6 is implemented.

Images