A method and system for controlling IoT devices based on multiple protocols

Abstract

This application provides a multi-protocol-based IoT device control method and system that combines correlation coefficients to determine a global device state quality expression. It calls the global device state quality expression of the business session topic in the auxiliary device operation event to determine the operation feedback data. Based on the operation feedback data, the communication protocol of the target IoT device corresponding to the operation log of the IoT device to be analyzed is adjusted. This allows for a more accurate acquisition of the global device state quality expression, thereby improving the accuracy and reliability of the obtained operation feedback data. Subsequently, the communication protocol of the target IoT device is continuously adjusted based on the operation feedback data. By adjusting the communication protocol of the target IoT device, control of the target IoT device can be achieved at the data communication level, enabling the target IoT device to perform complete and accurate data transmission based on the optimized communication protocol, thereby improving the working efficiency of the IoT device.

Description

Technical Field

[0001] This application relates to the field of data control technology, and more specifically, to a method and system for controlling Internet of Things (IoT) devices based on multiple protocols. Background Technology

[0002] The Internet of Things (IoT) can collect real-time data on any object or process that needs to be monitored, connected, and interacted with. Corresponding IoT devices can be specifically applied to the field of industrial production technology. However, in practical operation, inventors have discovered through continuous research that industrial production communication protocols may have anomalies, potentially leading to incomplete and inaccurate data transmission, which could reduce the efficiency of IoT devices. Therefore, a technology is urgently needed to solve these technical problems. Summary of the Invention

[0003] In view of this, this application provides a method and system for controlling IoT devices based on multiple protocols.

[0004] Firstly, a multi-protocol-based IoT device control method is provided, applied to an IoT device control system, the method comprising at least:

[0005] The system identifies the IoT device operation logs to be analyzed and the auxiliary device operation events bound to the IoT device operation logs; it then performs event description mining operations on the IoT device operation logs and the auxiliary device operation events in sequence to obtain the bound operation log event description set and the auxiliary operation event description set; the auxiliary operation event descriptions in the auxiliary operation event description set are associated with the business session topics in the auxiliary device operation events;

[0006] For each auxiliary operation event description in the auxiliary operation event description set, a correlation coefficient is determined between it and each operation log event description in the operation log event description set. Based on the correlation coefficient, a global device status quality expression for the business session topic corresponding to each auxiliary operation event description is determined. The global device status quality expression for the business session topic in the auxiliary device operation event is invoked to determine the operation feedback data of the IoT device operation log to be analyzed. The communication protocol of the target IoT device corresponding to the IoT device operation log to be analyzed is adjusted based on the operation feedback data.

[0007] In an alternative embodiment, determining the global device state quality representation of the service session topic corresponding to each auxiliary operation event description, in conjunction with the correlation coefficient, includes:

[0008] By combining the correlation coefficient between the current auxiliary operation event description and each operation log event description in the operation log event description set, the local focus index queue bound to each operation log event description is determined.

[0009] By combining the local focus index queue bound to each of the operation log event descriptions, the operation log event description set, and the current auxiliary operation event description, a global device status quality expression for the business session topic bound to the current auxiliary operation event description is determined.

[0010] In an alternative embodiment, the step of invoking the global device state quality expression of the business session topic in the auxiliary device operation event to determine the operation feedback data of the IoT device operation log to be analyzed includes:

[0011] Invoke the global device status quality expression bound to each service session topic in the auxiliary device operation event to determine the service session topic description value bound to each service session topic in the auxiliary device operation event;

[0012] By combining the global device status quality expression and the business session topic description value, the operation feedback data of the IoT device operation log to be analyzed is determined.

[0013] In an alternative embodiment, determining the operational feedback data of the IoT device operation log to be analyzed by combining the global device state quality expression and the service session topic description value includes:

[0014] Based on the global device status quality expression bound to each business session topic in the auxiliary device operation event, determine the device operation event splicing description corresponding to the auxiliary device operation event;

[0015] By combining the business session topic description value bound to each business session topic in the auxiliary device operation event, a global description value of the business session topic corresponding to the auxiliary device operation event is determined;

[0016] By combining the global description value of the business session topic and the concatenated description of the device operation event, the operation feedback data of the IoT device operation log to be analyzed is obtained.

[0017] In an alternative embodiment, the multi-protocol-based IoT device control method is executed by an AI neural network, which is debugged in the following manner:

[0018] Determine the first example IoT device operation log, the first example device operation event corresponding to the first example IoT device operation log, and the business session topic keywords bound to each business session topic in the first example device operation event;

[0019] The pre-debugged operation log feature extraction unit in the AI ​​neural network is invoked to extract IoT device operation log features from the operation log of the first example IoT device, thereby obtaining the bound candidate operation log event description set;

[0020] The event feature extraction unit to be debugged in the AI ​​neural network is invoked to extract device operation event features from the first example device operation event to obtain the bound candidate auxiliary operation event description set; the candidate auxiliary operation event descriptions in the candidate auxiliary operation event description set correspond to the business session topics in the first example device operation event;

[0021] The parallel local focusing unit in the AI ​​neural network is invoked to determine the correlation coefficient between each candidate auxiliary operation event description in the candidate auxiliary operation event description set and each candidate operation log event description in the candidate operation log event description set. Based on the correlation coefficient, the candidate global device status quality expression of the business session topic corresponding to each candidate auxiliary operation event description is determined.

[0022] The topic translation unit in the AI ​​neural network is invoked to perform topic translation on each of the candidate global device state quality expressions to obtain the first candidate business session topic description value bound to each business session topic in the first example device operation event;

[0023] The AI ​​neural network is first debugged by calling the first candidate business session topic description value and the business session topic keywords until the first debugging completion indicator is met.

[0024] In an alternative embodiment, determining the first example IoT device operation log, the first example device operation event corresponding to the first example IoT device operation log, and the business session topic keywords bound to each business session topic in the first example device operation event includes:

[0025] Identify the operation logs of the first-example IoT device and the basic device operation events that are related to the operation logs of the first-example IoT device; the basic device operation events include at least one basic service session topic.

[0026] Based on the set of backup service session topics, at least one basic service session topic in the basic equipment operation event is updated to obtain the bound first example equipment operation event;

[0027] The keywords of the business session topic bound to the update business session topic in the first example device operation event are adjusted to the first semantic word vector, and the keywords of the business session topic bound to the basic business session topic in the first example device operation event are adjusted to the second semantic word vector; wherein, the first semantic word vector and the second semantic word vector are different.

[0028] In an alternative embodiment, the step of invoking the first candidate business session topic description value and the business session topic keywords to perform a first debugging of the AI ​​neural network until a first debugging completion indicator is met includes:

[0029] By combining the first candidate business session topic description value and the bound business session topic keywords, a performance evaluation of the event feature extraction unit is generated.

[0030] The event feature extraction unit performance evaluation is invoked to perform the first debugging of the event feature extraction unit to be debugged in the AI ​​neural network until the first debugging completion index is met.

[0031] In an alternative embodiment, after the first debugging is completed, the method further includes:

[0032] Determine the second example IoT device operation log, the second example device operation event corresponding to the second example IoT device operation log, the operation feedback data semantic word vector corresponding to the second example IoT device operation log, and the business session topic description value semantic word vector bound to each business session topic in the second example device operation event;

[0033] The AI ​​neural network, which has been pre-tuned, outputs candidate operation feedback data of the second example IoT device operation log and second candidate business session topic description value of each business session topic in the second example device operation event.

[0034] By combining the second candidate business session topic description value and the semantic word vector of the bound business session topic description value, a first performance indicator of the AI ​​neural network is determined; by combining the candidate running feedback data and the semantic word vector of the running feedback data, a second performance indicator of the AI ​​neural network is determined; by combining the first performance indicator and the second performance indicator, a performance evaluation of the AI ​​neural network is generated.

[0035] The performance evaluation of the AI ​​neural network is invoked to perform a second debugging on the pre-debugged AI neural network until the second debugging completion index is met.

[0036] In an alternative embodiment, the runtime log feature extraction unit is obtained through a runtime log feature extraction unit debugging step, which includes:

[0037] The operation log analysis network to be debugged, the operation log of the third-paradigm IoT device, and the log semantic analysis results corresponding to the operation log of the third-paradigm IoT device are identified.

[0038] The operation log feature extraction unit in the operation log analysis network is invoked to extract IoT device operation log features from the operation log of the third example IoT device, and the bound feature extraction result set is obtained.

[0039] The topic translation unit in the runtime log analysis network is invoked to perform topic translation on the feature extraction result set to obtain the bound topic text expression;

[0040] Combining the thematic text expression and the log semantic analysis results, the runtime log feature extraction unit in the runtime log analysis network is subjected to a third debugging process until the third debugging completion indicator is met.

[0041] The running log feature extraction unit in the running log analysis network after the third debugging is completed is used as the pre-debugged running log feature extraction unit in the AI ​​neural network.

[0042] Secondly, a multi-protocol IoT device control system is provided, including a processor and a memory that communicate with each other, wherein the processor is used to read a computer program from the memory and execute it to implement the above-described method.

[0043] This application provides a multi-protocol-based IoT device control method and system. It sequentially performs event description mining operations on the IoT device's operation logs and auxiliary device operation events to obtain an operation log event description set and an auxiliary operation event description set. For each auxiliary operation event description in the auxiliary operation event description set, a correlation coefficient is determined between it and each operation log event description. Based on the correlation coefficient, a global device status quality expression is determined. The global device status quality expression of the business session topic in the auxiliary device operation events is invoked to determine operation feedback data. The communication protocol of the target IoT device corresponding to the IoT device's operation logs to be analyzed is adjusted based on the operation feedback data. This allows for a more accurate acquisition of the global device status quality expression, thereby improving the accuracy and reliability of the obtained operation feedback data. Continuous adjustment of the target IoT device's communication protocol based on the operation feedback data enables control of the target IoT device at the data communication level, allowing the target IoT device to transmit data completely and accurately based on the optimized communication protocol, thus improving the IoT device's operating efficiency. Attached Figure Description

[0044] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0045] Figure 1 This is a flowchart illustrating a multi-protocol-based IoT device control method provided in an embodiment of this application.

[0046] Figure 2 This is a block diagram of a multi-protocol IoT device control device provided in an embodiment of this application.

[0047] Figure 3 This is an architecture diagram of a multi-protocol IoT device control system provided in an embodiment of this application. Detailed Implementation

[0048] To better understand the above technical solutions, the technical solutions of this application will be described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the embodiments of this application and the specific features in the embodiments are detailed descriptions of the technical solutions of this application, rather than limitations on the technical solutions of this application. In the absence of conflict, the embodiments of this application and the technical features in the embodiments can be combined with each other.

[0049] Please see Figure 1 This paper illustrates a multi-protocol-based IoT device control method, which may include the technical solutions described in steps 201-205.

[0050] Step 201: Determine the IoT device operation log to be analyzed and the auxiliary device operation events bound to the IoT device operation log to be analyzed.

[0051] For example, an IoT device control system running a target protocol can represent protocol content attributes within the IoT device control system.

[0052] It is understandable that, based on the aforementioned auxiliary equipment operation events, there can be at least one business session topic. For example, an auxiliary equipment operation event can be a single business session topic, a set of business session topics, etc.

[0053] In an alternative embodiment, the target protocol can be understood as a network information transmission channel, and the IoT device control system operates with the network information transmission channel. When it is determined that "information transmission begins", the IoT device control system obtains the IoT device operation logs in real time while mining auxiliary device operation events, until it is determined that "information transmission ends", and uses the IoT device operation logs obtained between the start and end of information transmission as the IoT device operation logs to be analyzed.

[0054] Step 202: Perform event description mining operations on the IoT device operation logs and auxiliary device operation events to be analyzed in sequence to obtain the bound operation log event description set and auxiliary operation event description set; the auxiliary operation event descriptions in the auxiliary operation event description set correspond to the business session topics in the auxiliary device operation events.

[0055] For example, performing event description mining on the operational logs of IoT devices to be analyzed is a process of translating these logs. The operational log event description set is the feature extraction result set obtained after extracting IoT device operational log features from the log set of the IoT devices to be analyzed. Similarly, performing event description mining on auxiliary device operational events is a process of translating these events. The auxiliary operational event description set is the feature extraction result set obtained after extracting features from the business session topic set of auxiliary device operational events.

[0056] Furthermore, the event description mining operation is performed by an AI neural network. This AI neural network is a pre-tuned artificial intelligence network that includes a feature extraction unit and a topic translation unit. The feature extraction unit comprises an event feature extraction unit and a runtime log feature extraction unit. The event feature extraction unit translates auxiliary device runtime events into an auxiliary runtime event description set; the runtime log feature extraction unit translates the runtime logs of the IoT devices to be analyzed into a runtime log event description set; and the topic translation unit performs topic translation processing to obtain the runtime feedback data from the IoT device runtime logs to be analyzed.

[0057] For example, when an auxiliary device operation event is determined, the IoT device can process the auxiliary device operation event into a business session topic, obtain a set of related business session topics, and load the business session topic set into an AI neural network. Through several feature sub-extraction units of the event feature extraction unit in the AI ​​neural network, features are extracted from the business session topic set one by one to obtain an auxiliary operation event description set. The auxiliary operation event description set combines the key features extracted from each feature sub-extraction unit; the auxiliary operation event descriptions in the auxiliary operation event description set correspond to the business session topics in the auxiliary device operation event.

[0058] Furthermore, once the operational logs of the IoT device to be analyzed are determined, the IoT device can classify these logs to obtain a log set containing several content attribute logs. This log set is then loaded into an AI neural network. Through multiple feature extraction sub-units within the operational log feature extraction unit of the AI ​​neural network, IoT device operational log features are extracted from the log set one by one, resulting in a bound operational log event description set. The operational log event descriptions in this set correspond to the content attribute logs in the operational logs of the IoT device to be analyzed.

[0059] By using multiple feature extraction sub-units in the feature extraction unit of the AI ​​neural network, features are extracted one by one from the business session topic set or log set.

[0060] In an alternative embodiment, for the sub-service session topic processing of auxiliary device operation events that are Chinese device operation events, the auxiliary device operation events can be classified into service session topics by using a service session topic set or by using a determined service session topic classification method to obtain the bound service session topic set.

[0061] In an alternative embodiment, before classifying the IoT device operation logs to be analyzed, the IoT device operation logs to be analyzed may be preprocessed with noise and other methods to remove external influences from the IoT device operation logs to be analyzed.

[0062] Step 203: For each auxiliary operation event description in the auxiliary operation event description set, determine the correlation coefficient between it and each operation log event description in the operation log event description set.

[0063] For example, the IoT device determines each auxiliary operation event description in the auxiliary operation event description set and each operation log event description in the operation log event description set, and compares each auxiliary operation event description with each operation log event description for each auxiliary operation event description in the auxiliary operation event description set to determine the correlation coefficient between each auxiliary operation event description and each operation log event description set.

[0064] Understandably, IoT devices can perform correlation processing on each auxiliary operation event description and each operation log event description in the operation log event description set, based on the content described above, to obtain the correlation coefficient between each auxiliary operation event description and each operation log event description.

[0065] In an alternative embodiment, the IoT device may determine the correlation coefficient between the current auxiliary operation event description and the current operation log event description one by one, or it may determine the correlation coefficient between each auxiliary operation event description and each operation log event description in parallel.

[0066] In an alternative embodiment, the AI ​​neural network may also include a parallel local focusing unit. After obtaining the auxiliary operation event description set and the operation log event description set, the IoT device can determine the correlation coefficient between each auxiliary operation event description and each auxiliary operation event description through the parallel local focusing unit in the AI ​​neural network.

[0067] Step 204: Combine the correlation coefficient to determine the global device status quality expression of the business session topic corresponding to each auxiliary operation event description.

[0068] For example, when determining the correlation coefficient between each auxiliary operation event description in the auxiliary operation event description set and each operation log event description in the operation log event description set, the IoT device can combine the correlation coefficient to determine the global device state quality expression of the business session topic to which each auxiliary operation event description is bound. Here, the global device state quality expression is key descriptive data used to characterize the business session topic corresponding to the auxiliary operation event description.

[0069] To better explain this embodiment, the following steps can determine the operation of the global device status quality expression of the service session topic corresponding to the current auxiliary operation event description.

[0070] The IoT device combines the correlation coefficient between the current auxiliary operation event description in the auxiliary operation event description set and each operation log event description in the operation log event description set to determine the local focus index bound to each operation log event description in the operation log event description set. Then, by combining the local focus index bound to each operation log event description and the operation log event description set, it determines the global device status quality expression of the business session topic bound to the current auxiliary operation event description.

[0071] In an alternative embodiment, after determining the correlation coefficient between each auxiliary operation event description and each operation log event description, the IoT device can determine the global device state quality expression of the business session topic corresponding to each auxiliary operation event description through the parallel local focusing unit in the AI ​​neural network.

[0072] In an alternative embodiment, for the auxiliary operation event description set and the operation log event description set, the IoT device can load the auxiliary operation event description set and the operation log event description set into the parallel local focusing unit in the AI ​​neural network, through the parallel local focusing unit.

[0073] Step 205: Determine the IoT device operation to be analyzed by using the global device status quality expression of the business session topic in the auxiliary device operation event; adjust the communication protocol of the target IoT device corresponding to the operation log of the IoT device to be analyzed by combining the operation feedback data.

[0074] For example, operational feedback data refers to mining quantification metrics designed to represent the operational logs of IoT devices to be analyzed.

[0075] For example, when the global device state quality expression bound to each business session topic is determined, the IoT device can classify the global device state quality expression to obtain the business session topic description value of the business session topic corresponding to the global device state quality expression, and combine it with the business session topic description value of each business session topic in the auxiliary device operation event to determine the operation feedback data of the IoT device operation log to be analyzed.

[0076] In an alternative embodiment, the IoT device can depolarize the description value of the business session topic bound to each business session topic to obtain the operation feedback data of the IoT device operation log to be analyzed.

[0077] In an alternative embodiment, when it is determined that the operational feedback data of the IoT device's operational log to be analyzed is obtained, the IoT device can be evaluated by the IoT device control system through the network information transmission channel, and the operational feedback data can be displayed according to the evaluation results.

[0078] According to the above-mentioned IoT device control method based on multiple protocols, by determining the IoT device operation log and auxiliary device operation events to be analyzed, event description mining operations can be performed on the IoT device operation log and auxiliary device operation events to be analyzed in sequence to obtain the bound operation log event description set and auxiliary operation event description set.

[0079] By defining the operational log event description set and the auxiliary operational event description set, the operational log event descriptions can be correlated with the auxiliary operational event descriptions to determine the correlation coefficient between each auxiliary operational event description in the auxiliary operational event description set and each operational log event description in the operational log event description set. Thus, the correlation coefficient can be used to determine the globally relevant device status quality expression bound to each business session topic, and the operational feedback data of the IoT device operational log to be analyzed can be determined using the globally relevant device status quality expression. Since the operational feedback data of the IoT device operational log to be analyzed is determined by combining the correlation coefficient between each auxiliary operational event description and each operational log event description, this application can improve the stability of the multi-protocol-based IoT device control method.

[0080] In an alternative embodiment, the global device state quality representation of the business session topic corresponding to each auxiliary operation event description is determined by combining the correlation coefficient, including: determining the local focus index queue bound to each operation log event description by combining the correlation coefficient between the current auxiliary operation event description and each operation log event description in the operation log event description set; and determining the global device state quality representation of the business session topic bound to the current auxiliary operation event description by combining the local focus index queue bound to each operation log event description, the operation log event description set, and the current auxiliary operation event description.

[0081] For example, the IoT device combines the correlation coefficient between the current auxiliary operation event description and each operation log event description to determine the local focus index queue bound to each operation log event description. For instance, the weighted processing result between the current auxiliary operation event description and the current operation log event description is used as the local focus index queue bound to the current operation log event description. Further, the IoT device combines the local focus index queue bound to each set of operation log event descriptions with the current auxiliary operation event descriptions to perform concatenation processing on the operation log event description set and the current auxiliary operation event description, obtaining a global device status quality expression for the business session topic bound to the current auxiliary operation event description.

[0082] In an alternative embodiment, the operational feedback data of the IoT device operation log to be analyzed is determined by using the global device status quality expression of the business session topic in the auxiliary device operation event. This includes: determining the business session topic description value bound to each business session topic in the auxiliary device operation event by using the global device status quality expression bound to each business session topic in the auxiliary device operation event; and combining the global device status quality expression and the business session topic description value to determine the operational feedback data of the IoT device operation log to be analyzed.

[0083] For example, when the global device state quality expression bound to each business session topic is obtained, the IoT device loads each global device state quality expression into the topic translation unit in the AI ​​neural network. The topic translation unit classifies the global device state quality expression bound to each business session topic in the auxiliary device operation event, determines the business session topic description value bound to each business session topic in the auxiliary device operation event, and performs topic translation processing on the global device state quality expression in combination with the business session topic description value to obtain the operation feedback data corresponding to the IoT device operation log to be analyzed.

[0084] In this embodiment, the operational feedback data is determined by combining the measurement description values ​​of each service session topic with the global device status quality expression, making the determined operational feedback data more accurate.

[0085] In an alternative embodiment, the operational feedback data of the IoT device operation log to be analyzed is determined by combining the global device status quality expression and the business session topic description value. This includes: determining the device operation event splicing description corresponding to the auxiliary device operation event by combining the global device status quality expression bound to each business session topic in the auxiliary device operation event; determining the business session topic global description value corresponding to the auxiliary device operation event by combining the business session topic description value bound to each business session topic in the auxiliary device operation event; and obtaining the operational feedback data of the IoT device operation log to be analyzed by combining the business session topic global description value and the device operation event splicing description.

[0086] For example, the IoT device simplifies the global device status quality representation bound to each business session topic in the auxiliary device operation event to obtain a device operation event concatenation description corresponding to the auxiliary device operation event. It also merges the business session topic description values ​​bound to each business session topic in the auxiliary device operation event to obtain a global description value for the business session topic corresponding to the auxiliary device operation event. Further, the IoT device corrects the global description value of the business session topic and the device operation event concatenation description to obtain a transition situation, and optimizes the transition situation to obtain the operation feedback data of the IoT device operation log to be analyzed.

[0087] In this embodiment, by combining the global description value of the business session topic and the description of the device operation event, a further description of the auxiliary device operation event can be obtained, thereby making the operation feedback data obtained through the further description of the auxiliary device operation event more accurate.

[0088] In an alternative embodiment, an event description mining operation is performed on the IoT device operation logs and auxiliary device operation events to be analyzed, resulting in a bound set of operation log event descriptions and an auxiliary operation event description set. This includes: extracting IoT device operation log features from the IoT device operation logs to be analyzed using an operation log feature extraction unit in an AI neural network to obtain a bound set of operation log event descriptions; extracting device operation event features from the auxiliary device operation events using an event feature extraction unit in an AI neural network to obtain a bound set of auxiliary operation event descriptions; and, for each auxiliary operation event description in the auxiliary operation event description set, determining its association with each operation log event description in the operation log event description set. The correlation coefficients between them include: determining the correlation coefficient between each auxiliary operation event description in the auxiliary operation event description set and each operation log event description in the operation log event description set through the parallel local focusing unit in the AI ​​neural network; combining the correlation coefficients, determining the global device status quality expression of the business session topic corresponding to each auxiliary operation event description, including: determining the global device status quality expression of the business session topic corresponding to each auxiliary operation event description through the parallel local focusing unit in the AI ​​neural network and the correlation coefficients; determining the operation feedback data of the IoT device operation log to be analyzed through the global device status quality expression of the business session topic in the auxiliary device operation event, including:

[0089] By using the topic translation unit in the AI ​​neural network, the global device status quality expression of the business session topic in the auxiliary device operation event is translated into a topic, and the operation feedback data of the IoT device operation log to be analyzed is obtained.

[0090] For example, the AI ​​neural network includes a feature extraction unit and a topic translation unit. The feature extraction unit includes an operation log feature extraction unit, an event feature extraction unit, and a parallel local focusing unit. The operation log feature extraction unit performs IoT device operation log event description mining on the operation logs of the IoT device to be analyzed, obtaining an operation log event description set. The event feature extraction unit performs event description mining on auxiliary device operation events, obtaining an auxiliary operation event description set. The parallel local focusing unit determines the correlation coefficient between each auxiliary operation event description in the auxiliary operation event description set and each operation log event description in the operation log event description set, and combines the correlation coefficient to determine the global device status quality expression of the business session topic corresponding to each auxiliary operation event description. The topic translation unit performs topic translation processing on the global device status quality expression of the business session topic in the auxiliary device operation events, obtaining the operation feedback data of the IoT device operation logs to be analyzed.

[0091] In one possible embodiment, when the IoT device operation logs and auxiliary device operation events to be analyzed are determined, the IoT device can classify the IoT device operation logs to obtain a log set, and process the auxiliary device operation events by business session topic to obtain a business session topic set. The log set is then loaded into the operation log feature extraction unit, and the business session topic set is loaded into the event feature extraction unit to obtain the bound operation log event description set and auxiliary operation event description set. Further, the IoT device, through a parallel local focusing unit, determines the correlation coefficient between each auxiliary operation event description in the auxiliary operation event description set and each operation log event description in the operation log event description set. Based on the correlation coefficient, it determines the global device state quality expression for the business session topic corresponding to each auxiliary operation event description.

[0092] The IoT device loads the global device state quality representation of the business session topic corresponding to each auxiliary operation event description into the computation unit of the AI ​​neural network's topic translation unit. The computation unit outputs the business session topic description value bound to each business session topic in the auxiliary device operation event, and combines this with the business session topic description value bound to each business session topic in the auxiliary device operation event to determine the global description value of the business session topic corresponding to the auxiliary device operation event. Furthermore, the simplification processing unit in the topic translation unit determines the device operation event concatenation description corresponding to the auxiliary device operation event. Further, the IoT device, through multiple units in the topic translation unit, combines the global description value of the business session topic and the device operation event concatenation description to obtain the operation feedback data of the IoT device operation log to be analyzed.

[0093] In this embodiment, the operation logs of the IoT devices to be analyzed are evaluated by each network unit in the AI ​​neural network, so that the operation feedback data of the IoT devices to be analyzed is more accurate.

[0094] In an alternative embodiment, the multi-protocol-based IoT device control method is executed by an AI neural network, and the debugging steps of the AI ​​neural network may specifically include the following.

[0095] Step 701: Determine the first example IoT device operation log, the first example device operation event corresponding to the first example IoT device operation log, and the business session topic keywords bound to each business session topic in the first example device operation event.

[0096] For example, the first-example IoT device operation log and the corresponding first-example device operation events are debugging information required for network debugging. Business session topic keywords are semantic word vector information required for network debugging; the AI ​​neural network can adjust network variables accordingly using these keywords. Several first-example device operation events and their associated first-example IoT device operation logs can be determined, and semantic word vector annotations can be performed on each business session topic within the first-example device operation events to obtain the business session topic keywords associated with each topic. The first-example IoT device operation logs associated with the first-example device operation events can characterize the first-example IoT device operation logs related to those events, for example, the first-example IoT device operation logs determined through mining of the first-example device operation events.

[0097] Furthermore, the IoT device can use the first example device operation event and the corresponding first example IoT device operation log as a debug tuple in the debug set. In this way, the AI ​​neural network can be debugged through several debug tuples in the debug set.

[0098] Step 702: Using the pre-tuned operation log feature extraction unit in the AI ​​neural network, perform IoT device operation log feature extraction on the operation log of the first example IoT device to obtain the bound candidate operation log event description set.

[0099] Step 703: Through the event feature extraction unit to be debugged in the AI ​​neural network, the device operation event features of the first example device operation event are extracted to obtain the bound candidate auxiliary operation event description set; the candidate auxiliary operation event descriptions in the candidate auxiliary operation event description set correspond to the business session topics in the first example device operation event.

[0100] For example, an IoT device can load the first example IoT device operation log into a pre-debugged operation log feature extraction unit in an AI neural network, load the first example device operation event into an event feature extraction unit to be debugged in the AI ​​neural network, perform IoT device operation log event description mining operation on the first example IoT device operation log through the pre-debugged operation log feature extraction unit to obtain the bound candidate operation log event description set, and perform device operation event description mining operation on the first example device operation event through the event feature extraction unit to be debugged to obtain the bound candidate auxiliary operation event description set.

[0101] Step 704: Through the parallel local focusing unit in the AI ​​neural network, determine the correlation coefficient between each candidate auxiliary operation event description in the candidate auxiliary operation event description set and each candidate operation log event description in the candidate operation log event description set, and combine the correlation coefficient to determine the candidate global device status quality expression of the business session topic corresponding to each candidate auxiliary operation event description.

[0102] Step 705: The topic translation unit in the AI ​​neural network performs topic translation on each candidate global device state quality expression to obtain the first candidate business session topic description value bound to each business session topic in the first example device operation event.

[0103] For example, for each candidate auxiliary operation event description in the candidate auxiliary operation event description set, the IoT device can use the parallel local focusing unit in the AI ​​neural network to compare each candidate auxiliary operation event description with each candidate operation log event description to determine the correlation coefficient between each candidate auxiliary operation event description and each candidate operation log event description set. Based on the correlation coefficient, the device can determine the candidate global device state quality expression of the business session topic to which each candidate auxiliary operation event description is bound. Further, the topic translation unit in the AI ​​neural network can classify the candidate global device state quality expressions to obtain the first candidate business session topic description value of the business session topic corresponding to the candidate global device state quality expression.

[0104] Step 706: Perform the first debugging of the AI ​​neural network using the first candidate business session topic description value and business session topic keywords until the first debugging completion indicator is met.

[0105] In this embodiment, during the debugging operation, since the debugging target references the first business session topic description value and business session topic keywords bound to each business session topic in the first example device operation event, the AI ​​neural network can perform complete training on each business session topic, thereby improving the accuracy of the business session topic description value and thus improving the credibility of the operation feedback data.

[0106] In an alternative embodiment, determining the first example IoT device operation log, the first example device operation event bound to the first example IoT device operation log, and the business session topic keywords bound to each business session topic in the first example device operation event includes: determining the first example IoT device operation log and the basic device operation events related to the first example IoT device operation log; the basic device operation events include at least one basic business session topic; based on a set of backup business session topics, updating at least one basic business session topic in the basic device operation events to obtain the bound first example device operation events; adjusting the business session topic keywords bound to the updated business session topics in the first example device operation events to a first semantic word vector, and adjusting the business session topic keywords bound to the basic business session topics in the first example device operation events to a second semantic word vector, wherein the first semantic word vector and the second semantic word vector are different.

[0107] For example, several first-example IoT device operation logs and basic device operation events mined from the first-example IoT device operation logs are collected. The first-example IoT device operation logs and the bound basic device operation events are loaded into the IoT device, so that the IoT device continuously updates at least one basic business session topic in the basic device operation events through a configured backup business session topic set. Here, a basic business session topic refers to an unupdated sub-business session topic included in the basic device operation event. The backup business session topic set refers to a business session topic library including several update business session topics used to update the basic business session topics. Further, the IoT device determines the update business session topic and the basic business session topic in the first-example device operation events, adjusts the business session topic keywords bound to the update business session topic into a first semantic word vector, and adjusts the business session topic keywords bound to the basic business session topic into a second semantic word vector. Here, the first semantic word vector and the second semantic word vector differ.

[0108] Based on the above description, it can be understood that debugging the AI-trained network mainly relies on several debugging information, thereby improving upon the existing technology that relies on manual semantic word vector annotation of debugging information. This reduces work efficiency and wastes human and time resources. This embodiment can intelligently and continuously update the basic business session topics and intelligently configure business session topic keywords based on the update status. This not only improves the efficiency of semantic word vector annotation but also effectively saves labor and time costs.

[0109] In an alternative embodiment, the AI ​​neural network is first debugged using the first candidate business session topic description value and the business session topic keywords until the first debugging completion indicator is met, including: combining the first candidate business session topic description value and the bound business session topic keywords to generate an event feature extraction unit performance evaluation.

[0110] The event feature extraction unit in the AI ​​neural network is first debugged by evaluating its performance, and the process ends when the first debugging completion index is met.

[0111] For example, the IoT device determines the first candidate business session topic description value and business session topic keyword bound to the same business session topic in the first debugging device operation event, and generates a device operation event feature extraction performance evaluation by combining the first candidate business session topic description value and business session topic keyword in the same business session topic in the first debugging device operation event. That is, it generates an event feature extraction unit performance evaluation by combining the first candidate business session topic description value and the bound business session topic keyword. Further, the IoT device fixes the network variables of the operation log feature extraction unit in the AI ​​neural network, and debugs the event feature extraction unit to be debugged through the event feature extraction unit performance evaluation until the first debugging completion index is met. For example, until the global error between each first candidate business session topic description value and the bound business session topic keyword in the first debugging device operation event meets a preset standard.

[0112] In this embodiment, the performance evaluation of the event feature extraction unit is determined by the first candidate service session topic description value and the bound service session topic keywords, so that the auxiliary running event description set output by the event feature extraction unit obtained by debugging through the performance evaluation of the event feature extraction unit is more complete.

[0113] In an alternative embodiment, after the first debugging is completed, the network debugging method further includes: determining the second example IoT device operation log, the second example device operation event corresponding to the second example IoT device operation log, the semantic word vector of the operation feedback data corresponding to the second example IoT device operation log, and the semantic word vector of the business session topic description value bound to each business session topic in the second example device operation event; outputting candidate operation feedback data of the second example IoT device operation log and the second candidate business session topic description value of each business session topic in the second example device operation event through the pre-debugged AI neural network; determining the first performance index of the AI ​​neural network by combining the second candidate business session topic description value and the bound business session topic description value semantic word vector; determining the second performance index of the AI ​​neural network by combining the candidate operation feedback data and the operation feedback data semantic word vector; generating the performance evaluation of the AI ​​neural network by combining the first performance index and the second performance index; and performing a second debugging on the pre-debugged AI neural network through the performance evaluation of the AI ​​neural network until the second debugging completion index is met.

[0114] For example, after the first debugging is completed, a second debugging can be performed on the AI ​​neural network to further improve the evaluation accuracy. The IoT device determines the second-paradigm IoT device operation log, the second-paradigm device operation events corresponding to the second-paradigm IoT device operation log, the operation feedback data semantic word vectors corresponding to the second-paradigm IoT device operation log, and the business session topic description value semantic word vectors bound to each business session topic in the second-paradigm device operation events. The operation feedback data semantic word vectors are intended to represent the mining results of the second-paradigm IoT device operation logs; they can be understood as describing the mining results of the second-paradigm IoT device operation logs. The business session topic description value semantic word vectors are used to characterize the mining results of the business session topics in the second-paradigm IoT device operation logs; they can be understood as describing the mining results of the business session topics.

[0115] Furthermore, the IoT device loads the second-example device operation events and the bound second-example IoT device operation logs into a pre-debugged AI neural network, i.e., into the first debugged AI neural network. The pre-debugged AI neural network outputs candidate operation feedback data corresponding to the second-example IoT device operation logs and second candidate business session topic description values ​​for each business session topic in the second-example device operation events. The IoT device combines the second candidate business session topic description values ​​and semantic word vectors of the business session topic description values ​​for the same business session topic in the second-example device operation events to determine the first performance index of the AI ​​neural network. It then combines the candidate operation feedback data from the second-example IoT device operation logs and the bound operation feedback data semantic word vectors to determine the second performance index of the AI ​​neural network. Finally, it combines the first and second performance indices to generate a performance evaluation of the AI ​​neural network.

[0116] Furthermore, the IoT device performs a second debugging of the pre-debugged AI neural network by evaluating the performance of the generated AI neural network. This involves jointly debugging the event feature extraction unit, the operation log feature extraction unit, and the topic translation unit in the debugged AI neural network until the second debugging completion index is met.

[0117] In one possible embodiment, multiple people can process the descriptive values ​​of the business session topics in the second example device operation events and the bound second example IoT device operation logs, and perform depolarization processing on the descriptive values ​​of the second example IoT device operation logs to obtain the semantic word vector of the operation feedback data corresponding to the second example IoT device operation logs; and perform depolarization processing on the descriptive values ​​of the same business session topics to obtain the semantic word vector of the business session topic descriptive values ​​of the corresponding business session topics.

[0118] In this embodiment, since the semantic word vectors of the running feedback data and the semantic word vectors of the business session topic description values ​​can determine the real-time status data, the AI ​​neural network that has been debugged in advance is debugged a second time through the real-time status data, so that the running feedback data output by the debugged AI neural network is more accurate.

[0119] In an alternative embodiment, the operation log feature extraction unit is obtained through operation log feature extraction unit debugging. The operation log feature extraction unit debugging steps may include the following: determining the operation log analysis network to be debugged, the operation log of the third example IoT device, and the log semantic analysis results corresponding to the operation log of the third example IoT device; extracting IoT device operation log features from the operation log of the third example IoT device through the operation log feature extraction unit in the operation log analysis network to obtain the bound feature extraction result set; translating the feature extraction result set into a topic through the topic translation unit in the operation log analysis network to obtain the bound topic text expression; combining the topic text expression and the log semantic analysis results, performing a third debugging on the operation log feature extraction unit in the operation log analysis network until the third debugging completion index is met.

[0120] The runtime log feature extraction unit in the runtime log analysis network after the third debugging is completed is used as the runtime log feature extraction unit in the AI ​​neural network after pre-debugging.

[0121] For example, before performing the first debugging of the AI ​​neural network, the runtime log feature extraction unit in the runtime log analysis network can be pre-debugged. Since the runtime log feature extraction unit in the AI ​​neural network mainly performs feature extraction, similar runtime log feature extraction units in the runtime log analysis network can be used as runtime log feature extraction units in the AI ​​neural network.

[0122] The IoT device identifies the operational log analysis network to be debugged, the operational logs of a third-paradigm IoT device, and the corresponding log semantic analysis results. It then loads the operational logs of the third-paradigm IoT device into the operational log feature extraction unit of the operational log analysis network to be debugged. The operational log feature extraction unit of the operational log analysis network extracts features from the operational logs of the third-paradigm IoT device, obtaining a bound feature extraction result set. Further, the IoT device loads the feature extraction result set into the topic translation unit of the operational log analysis network. The topic translation unit of the operational log analysis network translates the feature extraction result set into a topic text expression optimized for the operational logs of the third-paradigm IoT device.

[0123] The IoT device combines the differences between the topic text expression and the bound log semantic analysis results to generate a performance evaluation of the running log analysis network. The running log feature extraction unit of the running log analysis network is then debugged through the performance evaluation of the running log analysis network until the third debugging completion index is met. The running log feature extraction unit in the running log analysis network after the third debugging is completed is used as the pre-debugged running log feature extraction unit in the AI ​​neural network.

[0124] In this embodiment, by performing a third debugging on the runtime log analysis network, the runtime log event description set output by the debugged runtime log feature extraction unit can be more accurate.

[0125] In an alternative embodiment, a multi-protocol-based IoT device control method is provided. This multi-protocol-based IoT device control method specifically includes the following description.

[0126] Step 801 indicates an auxiliary equipment operation event.

[0127] For example, an IoT device control system operates a network information transmission channel, through which auxiliary device operation events can be expressed. These auxiliary device operation events are represented through a webpage.

[0128] Step 802: In response to the mining step performed for the auxiliary device operation event, perform IoT device operation log collection to obtain the IoT device operation log to be analyzed obtained by mining the auxiliary device operation event.

[0129] For example, when it is determined that a user performs a mining step on an auxiliary device operation event, the IoT device control system collects the attribute content determined by the user through mining the auxiliary device operation event, and uses the attribute content as the IoT device operation log to be analyzed. Step 803: Display the operation feedback data of the IoT device operation log to be analyzed.

[0130] For example, when the IoT device operation log to be analyzed is determined, the IoT device control system loads the IoT device operation log to be analyzed and the bound auxiliary device operation events into the AI ​​neural network, outputs the business session topic description value of each business session topic in the auxiliary device operation events through the AI ​​neural network, and determines the operation feedback data of the IoT device operation log to be analyzed by combining the business session topic description value of each business session topic.

[0131] Step 804: In response to the triggering operation of the business session topic in the auxiliary equipment operation event, display the business session topic description value of the business session topic; further, the operation feedback data is determined by combining the business session topic description value of each business session topic in the auxiliary equipment operation event.

[0132] In an alternative embodiment, the IoT device control system may operate a network information transmission channel. This channel can display auxiliary device operation events and collect IoT device operation logs for analysis, which users can then mine and analyze. The auxiliary device operation events and the collected IoT device operation logs are then sent to the IoT device control system. When the auxiliary device operation events and the IoT device operation logs are determined, the IoT device control system loads them into an AI neural network. The AI ​​neural network determines the correlation coefficient between the auxiliary device operation events and the IoT device operation logs, and outputs a business session topic description value and operational feedback data based on the correlation coefficient. Further, the IoT device control system returns the business session topic description value and operational feedback data to the network information transmission channel, providing explanations of these data.

[0133] According to the above-described multi-protocol-based IoT device control method, by displaying auxiliary device operation events, the IoT device operation logs to be analyzed can be collected in response to the mining steps performed on the auxiliary device operation events. By collecting the IoT device operation logs to be analyzed, the business session topic description value of each business session topic in the auxiliary device operation events can be determined through the IoT device operation logs to be analyzed and the auxiliary device operation events. Combined with the business session topic description value of each business session topic, the operation feedback data of the IoT device operation logs to be analyzed can be determined. By determining the business session topic description value of each business session topic and the operation feedback data of the IoT device operation logs to be analyzed, the description of the business session topic description value can be output in conjunction with the execution steps of the business session topics in the auxiliary device operation events while outputting the operation feedback data.

[0134] In one possible embodiment, the multi-protocol-based IoT device control method may specifically include the following steps.

[0135] Step 1001: Determine the IoT device operation log to be analyzed and the auxiliary device operation events bound to the IoT device operation log to be analyzed.

[0136] Step 1002: Perform event description mining operations on the IoT device operation logs and auxiliary device operation events to be analyzed in sequence to obtain the bound operation log event description set and auxiliary operation event description set; the auxiliary operation event descriptions in the auxiliary operation event description set correspond to the business session topics in the auxiliary device operation events.

[0137] Step 1003: For each auxiliary operation event description in the auxiliary operation event description set, determine the correlation coefficient between it and each operation log event description in the operation log event description set; combine the correlation coefficient between the current auxiliary operation event description and each operation log event description in the operation log event description set to determine the local focus index queue bound to each operation log event description.

[0138] Step 1004: Combine the local focus index queue, the set of operation log event descriptions, and the current auxiliary operation event description bound to each operation log event description to determine the global device status quality expression of the business session topic bound to the current auxiliary operation event description.

[0139] Step 1005: Determine the business session topic description value bound to each business session topic in the auxiliary device operation event by using the global device status quality expression bound to each business session topic in the auxiliary device operation event; combine the global device status quality expression bound to each business session topic in the auxiliary device operation event to determine the device operation event splicing description corresponding to the auxiliary device operation event.

[0140] Step 1006: Combine the business session topic description values ​​bound to each business session topic in the auxiliary equipment operation event to determine the global description value of the business session topic corresponding to the auxiliary equipment operation event.

[0141] Step 1007: Combine the global description value of the business session topic and the concatenated description of the device operation event to obtain the operation feedback data of the IoT device operation log to be analyzed.

[0142] In one possible embodiment, the steps described in the AI ​​neural network debugging method specifically include the following.

[0143] Step 1101: Determine the operation log analysis network to be debugged, the operation log of the third-example IoT device, and the log semantic analysis results corresponding to the operation log of the third-example IoT device.

[0144] Step 1102: Through the operation log feature extraction unit in the operation log analysis network, the operation log of the third example IoT device is subjected to IoT device operation log feature extraction to obtain the bound feature extraction result set; through the topic translation unit in the operation log analysis network, the feature extraction result set is translated to obtain the bound topic text expression.

[0145] Step 1103: Combining the topic text expression and log semantic analysis results, perform a third debugging on the running log feature extraction unit in the running log analysis network until the third debugging completion index is met; use the running log feature extraction unit in the running log analysis network after the third debugging as the pre-debugged running log feature extraction unit in the AI ​​neural network.

[0146] Step 1104: Determine the first example IoT device operation log and the basic device operation events related to the first example IoT device operation log; the basic device operation events include at least one basic business session topic; based on the set backup business session topic set, update at least one basic business session topic in the basic device operation events to obtain the bound first example device operation events.

[0147] Step 1111: Adjust the business session topic keywords bound to the update business session topic in the first example device operation event to the first semantic word vector, and adjust the business session topic keywords bound to the basic business session topic in the first example device operation event to the second semantic word vector; wherein, there is a difference between the first semantic word vector and the second semantic word vector.

[0148] Step 1112: Using the pre-tuned operation log feature extraction unit in the AI ​​neural network, extract IoT device operation log features from the operation log of the first example IoT device to obtain the bound candidate operation log event description set; Using the event feature extraction unit to be debugged in the AI ​​neural network, extract device operation event features from the operation events of the first example device to obtain the bound candidate auxiliary operation event description set.

[0149] Step 1113: Through the parallel local focusing unit in the AI ​​neural network, determine the correlation coefficient between each candidate auxiliary operation event description in the candidate auxiliary operation event description set and each candidate operation log event description in the candidate operation log event description set, and combine the correlation coefficient to determine the candidate global device status quality expression of the business session topic corresponding to each candidate auxiliary operation event description.

[0150] Step 1114: The topic translation unit in the AI ​​neural network performs topic translation on each candidate global device state quality expression to obtain the first candidate business session topic description value bound to each business session topic in the first example device operation event.

[0151] Step 1115: Combine the first candidate business session topic description value and the bound business session topic keywords to generate an event feature extraction unit performance evaluation; perform the first debugging of the event feature extraction unit to be debugged in the AI ​​neural network through the event feature extraction unit performance evaluation until the first debugging completion index is met.

[0152] Step 1116: Determine the second-example IoT device operation log, the second-example device operation event corresponding to the second-example IoT device operation log, the semantic word vector of the operation feedback data corresponding to the second-example IoT device operation log, and the semantic word vector of the business session topic description value bound to each business session topic in the second-example device operation event.

[0153] Step 1117: Using a pre-tuned AI neural network, output candidate operation feedback data of the second example IoT device operation log, and second candidate business session topic description values ​​for each business session topic in the second example device operation event.

[0154] Step 1118: Combine the second candidate business session topic description value and the semantic word vector of the bound business session topic description value to determine the first performance index of the AI ​​neural network; combine the candidate running feedback data and the semantic word vector of the running feedback data to determine the second performance index of the AI ​​neural network.

[0155] Step 1119: Combine the first performance index and the second performance index to generate a performance evaluation of the AI ​​neural network; perform a second debugging on the pre-debugged AI neural network based on the performance evaluation of the AI ​​neural network until the second debugging completion index is met.

[0156] Based on the above, please refer to the following: Figure 2 A multi-protocol-based IoT device control device 200 is provided, applied to a multi-protocol-based IoT device control system, the device comprising:

[0157] The description binding module 210 is used to determine the IoT device operation log to be analyzed and the auxiliary device operation events bound to the IoT device operation log to be analyzed; and to perform event description mining operations on the IoT device operation log to be analyzed and the auxiliary device operation events in sequence to obtain the bound operation log event description set and the auxiliary operation event description set; the auxiliary operation event descriptions in the auxiliary operation event description set are associated with the business session topics in the auxiliary device operation events;

[0158] The protocol adjustment module 220 is used to determine, for each auxiliary operation event description in the auxiliary operation event description set, a correlation coefficient between it and each operation log event description in the operation log event description set; combine the correlation coefficient to determine the global device status quality expression of the business session topic corresponding to each auxiliary operation event description; call the global device status quality expression of the business session topic in the auxiliary device operation event to determine the operation feedback data of the IoT device operation log to be analyzed; and adjust the communication protocol of the target IoT device corresponding to the IoT device operation log to be analyzed based on the operation feedback data.

[0159] Based on the above, please refer to the following: Figure 3 The present invention illustrates a multi-protocol IoT device control system 300, including a processor 310 and a memory 320 that communicate with each other. The processor 310 is used to read computer programs from the memory 320 and execute them to implement the above-described method.

[0160] Based on the above, a computer-readable storage medium is also provided, on which a computer program stored implements the above method during runtime.

[0161] In summary, based on the above scheme, event description mining operations are performed sequentially on the operational logs of the IoT devices to be analyzed and the operational events of auxiliary devices to obtain the operational log event description set and the auxiliary operational event description set. For each auxiliary operational event description in the auxiliary operational event description set, the correlation coefficient between it and each operational log event description is determined. Based on the correlation coefficient, a global device status quality expression is determined. The global device status quality expression of the business session topic in the auxiliary device operational events is called to determine the operational feedback data. The communication protocol of the target IoT device corresponding to the operational logs of the IoT devices to be analyzed is adjusted based on the operational feedback data. This allows for a more accurate acquisition of the global device status quality expression, thereby improving the accuracy and reliability of the obtained operational feedback data. Subsequently, the communication protocol of the target IoT device is continuously adjusted based on the operational feedback data. By adjusting the communication protocol of the target IoT device, control over the target IoT device can be achieved at the data communication level, enabling the target IoT device to transmit data completely and accurately based on the optimized communication protocol, thereby improving the working efficiency of the IoT device.

[0162] It should be understood that the systems and modules described above can be implemented in various ways. For example, in some embodiments, the systems and modules can be implemented by hardware, software, or a combination of both. The hardware portion can be implemented using dedicated logic; the software portion can be stored in memory and executed by an appropriate instruction execution system, such as a microprocessor or dedicated-design hardware. Those skilled in the art will understand that the methods and systems described above can be implemented using computer-executable instructions and / or included in processor control code, for example, such code provided on a carrier medium such as a disk, CD, or DVD-ROM, a programmable memory such as read-only memory (firmware), or a data carrier such as an optical or electronic signal carrier. The systems and modules of this application can be implemented not only by hardware circuits such as very large-scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, or programmable hardware devices such as field-programmable gate arrays, programmable logic devices, etc., but also by software executed by various types of processors, or by a combination of the aforementioned hardware circuits and software (e.g., firmware).

[0163] It should be noted that different embodiments may produce different beneficial effects. In different embodiments, the beneficial effects may be any one or a combination of the above, or any other possible beneficial effects.

[0164] The basic concepts have been described above. Obviously, for those skilled in the art, the detailed disclosure above is merely illustrative and does not constitute a limitation of this application. Although not explicitly stated herein, those skilled in the art may make various modifications, improvements, and corrections to this application. Such modifications, improvements, and corrections are suggested in this application, and therefore remain within the spirit and scope of the exemplary embodiments of this application.

[0165] Furthermore, this application uses specific terms to describe embodiments of the application. For example, "an embodiment," "one embodiment," and / or "some embodiments" refer to a particular feature, structure, or characteristic associated with at least one embodiment of the application. Therefore, it should be emphasized and noted that "an embodiment," "one embodiment," or "an alternative embodiment" mentioned twice or more in different locations in this specification do not necessarily refer to the same embodiment. In addition, certain features, structures, or characteristics in one or more embodiments of the application can be appropriately combined.

[0166] Furthermore, those skilled in the art will understand that aspects of this application can be described and illustrated through several patentable types or situations, including any new and useful combination of processes, machines, products, or substances, or any new and useful improvements thereof. Accordingly, aspects of this application can be implemented entirely by hardware, entirely by software (including firmware, resident software, microcode, etc.), or by a combination of hardware and software. All of the above hardware or software may be referred to as a “data block,” “module,” “engine,” “unit,” “component,” or “system.” Furthermore, aspects of this application may manifest as a computer product located on one or more computer-readable media, the product including computer-readable program code.

[0167] Computer storage media may contain a propagated data signal containing computer program code, for example, on baseband or as part of a carrier wave. This propagated signal may take various forms, including electromagnetic, optical, and suitable combinations thereof. Computer storage media can be any computer-readable medium other than a computer-readable storage medium, which can be connected to an instruction execution system, apparatus, or device to enable communication, propagation, or transmission of a program for use. The program code located on the computer storage medium can be propagated through any suitable medium, including radio, cable, fiber optic cable, RF, or similar media, or any combination of the above media.

[0168] The computer program code required for the operation of each part of this application can be written in any one or more programming languages, including object-oriented programming languages ​​such as Java, Scala, Smalltalk, Eiffel, JADE, Emerald, C++, C#, VB.NET, Python, etc., conventional procedural programming languages ​​such as C, Visual Basic, Fortran 2003, Perl, COBOL 2002, PHP, ABAP, dynamic programming languages ​​such as Python, Ruby, and Groovy, or other programming languages. This program code can run entirely on the user's computer, or as a standalone software package on the user's computer, or partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the latter case, the remote computer can be connected to the user's computer through any network, such as a local area network (LAN) or wide area network (WAN), or connected to an external computer (e.g., via the Internet), or in a cloud computing environment, or used as a service such as Software as a Service (SaaS).

[0169] Furthermore, unless expressly stated in the claims, the order of processing elements and sequences, the use of numbers and letters, or other names described in this application are not intended to limit the order of the processes and methods of this application. Although the foregoing disclosure has discussed some currently considered useful embodiments of the invention through various examples, it should be understood that such details are for illustrative purposes only, and the appended claims are not limited to the disclosed embodiments; rather, the claims are intended to cover all modifications and equivalent combinations that conform to the substance and scope of the embodiments of this application. For example, while the system components described above can be implemented using hardware devices, they can also be implemented solely through software solutions, such as installing the described system on existing servers or mobile devices.

[0170] Similarly, it should be noted that, in order to simplify the description of the present application and thus aid in the understanding of one or more embodiments of the invention, the foregoing description of the embodiments of the present application sometimes combines multiple features into a single embodiment, drawing, or description thereof. However, this disclosure method does not imply that the subject matter of the application requires more features than those mentioned in the claims. In fact, the embodiments contain fewer features than all the features of the single embodiments disclosed above.

[0171] In some embodiments, numbers describing the quantity of components and attributes are used. It should be understood that such numbers used in the description of embodiments are modified in some examples with the terms "approximately," "approximately," or "generally." Unless otherwise stated, "approximately," "approximately," or "generally" indicates that the numbers are open to adaptive variation. Accordingly, in some embodiments, the numerical parameters used in the specification and claims are approximate values, which may be changed depending on the characteristics required by individual embodiments. In some embodiments, numerical parameters are taken into account a specified number of significant digits and employ a general method of digit reservation. Although the numerical ranges and parameters used to confirm their breadth of application in some embodiments of this application are approximate values, in specific embodiments, such values ​​are set as precisely as feasible.

[0172] For each patent, patent application, patent application publication, and other material such as articles, books, specifications, publications, and documents referenced in this application, the entire contents of that patent are incorporated herein by reference. This excludes historical application documents that are inconsistent with or conflict with the content of this application, as well as documents that limit the broadest scope of the claims in this application (currently or subsequently appended to this application). It should be noted that if there are any inconsistencies or conflicts between the descriptions, definitions, and / or terminology used in the supplementary materials of this application and the content of this application, the descriptions, definitions, and / or terminology used in this application shall prevail.

[0173] Finally, it should be understood that the embodiments described in this application are merely illustrative of the principles of the embodiments of this application. Other modifications may also fall within the scope of this application. Therefore, alternative configurations of the embodiments of this application are considered as examples and not limitations, and are regarded as consistent with the teachings of this application. Accordingly, the embodiments of this application are not limited to the embodiments explicitly described and illustrated in this application.

[0174] The above are merely embodiments of this application and are not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A method for controlling IoT devices based on multiple protocols, characterized in that, Applied to an IoT device control system, the method includes at least: The system identifies the IoT device operation logs to be analyzed and the auxiliary device operation events bound to the IoT device operation logs; it then performs event description mining operations on the IoT device operation logs and the auxiliary device operation events in sequence to obtain the bound operation log event description set and the auxiliary operation event description set; the auxiliary operation event descriptions in the auxiliary operation event description set are associated with the business session topics in the auxiliary device operation events; For each auxiliary operation event description in the auxiliary operation event description set, a correlation coefficient is determined between it and each operation log event description in the operation log event description set. Based on the correlation coefficient, a global device status quality expression for the business session topic corresponding to each auxiliary operation event description is determined. The global device status quality expression for the business session topic in the auxiliary device operation event is invoked to determine the operation feedback data of the IoT device operation log to be analyzed. The communication protocol of the target IoT device corresponding to the IoT device operation log to be analyzed is adjusted based on the operation feedback data. The step of determining the global device status quality representation of the service session topic corresponding to each auxiliary operation event description by combining the correlation coefficient includes: By combining the correlation coefficient between the current auxiliary operation event description and each operation log event description in the operation log event description set, the local focus index queue bound to each operation log event description is determined. By combining the local focus index queue bound to each of the operation log event descriptions, the operation log event description set, and the current auxiliary operation event description, a global device status quality expression for the business session topic bound to the current auxiliary operation event description is determined.

2. The method as described in claim 1, characterized in that, The step of calling the global device status quality expression of the business session topic in the auxiliary device operation event to determine the operation feedback data of the IoT device operation log to be analyzed includes: Invoke the global device status quality expression bound to each service session topic in the auxiliary device operation event to determine the service session topic description value bound to each service session topic in the auxiliary device operation event; By combining the global device status quality expression and the business session topic description value, the operation feedback data of the IoT device operation log to be analyzed is determined.

3. The method as described in claim 2, characterized in that, The step of combining the global device status quality expression and the business session topic description value to determine the operation feedback data of the IoT device operation log to be analyzed includes: Based on the global device status quality expression bound to each business session topic in the auxiliary device operation event, determine the device operation event splicing description corresponding to the auxiliary device operation event; By combining the business session topic description value bound to each business session topic in the auxiliary device operation event, a global description value of the business session topic corresponding to the auxiliary device operation event is determined; By combining the global description value of the business session topic and the concatenated description of the device operation event, the operation feedback data of the IoT device operation log to be analyzed is obtained.

4. The method according to any one of claims 1 to 3, characterized in that, The multi-protocol-based IoT device control method is executed by an AI neural network, which is obtained through debugging in the following way: Determine the first example IoT device operation log, the first example device operation event corresponding to the first example IoT device operation log, and the business session topic keywords bound to each business session topic in the first example device operation event; The pre-debugged operation log feature extraction unit in the AI ​​neural network is invoked to extract IoT device operation log features from the operation log of the first example IoT device, thereby obtaining the bound candidate operation log event description set; The event feature extraction unit to be debugged in the AI ​​neural network is invoked to extract device operation event features from the first example device operation event to obtain the bound candidate auxiliary operation event description set; the candidate auxiliary operation event descriptions in the candidate auxiliary operation event description set correspond to the business session topics in the first example device operation event; The parallel local focusing unit in the AI ​​neural network is invoked to determine the correlation coefficient between each candidate auxiliary operation event description in the candidate auxiliary operation event description set and each candidate operation log event description in the candidate operation log event description set. Based on the correlation coefficient, the candidate global device status quality expression of the business session topic corresponding to each candidate auxiliary operation event description is determined. The topic translation unit in the AI ​​neural network is invoked to perform topic translation on each of the candidate global device state quality expressions to obtain the first candidate business session topic description value bound to each business session topic in the first example device operation event; The AI ​​neural network is first debugged by calling the first candidate business session topic description value and the business session topic keywords until the first debugging completion indicator is met.

5. The method as described in claim 4, characterized in that, The determination of the first example IoT device operation log, the first example device operation event corresponding to the first example IoT device operation log, and the business session topic keywords bound to each business session topic in the first example device operation event includes: Identify the operation logs of the first-example IoT device and the basic device operation events that are related to the operation logs of the first-example IoT device; the basic device operation events include at least one basic service session topic. Based on the set of backup service session topics, at least one basic service session topic in the basic equipment operation event is updated to obtain the bound first example equipment operation event; The keywords of the business session topic bound to the update business session topic in the first example device operation event are adjusted to the first semantic word vector, and the keywords of the business session topic bound to the basic business session topic in the first example device operation event are adjusted to the second semantic word vector; wherein, the first semantic word vector and the second semantic word vector are different.

6. The method as described in claim 4, characterized in that, The step of calling the first candidate business session topic description value and the business session topic keywords to perform a first debugging of the AI ​​neural network until the first debugging completion indicator is met includes: By combining the first candidate business session topic description value and the bound business session topic keywords, a performance evaluation of the event feature extraction unit is generated. The event feature extraction unit performance evaluation is invoked to perform the first debugging of the event feature extraction unit to be debugged in the AI ​​neural network until the first debugging completion index is met.

7. The method as described in claim 4, characterized in that, After the first debugging is completed, the method further includes: Determine the second example IoT device operation log, the second example device operation event corresponding to the second example IoT device operation log, the operation feedback data semantic word vector corresponding to the second example IoT device operation log, and the business session topic description value semantic word vector bound to each business session topic in the second example device operation event; The AI ​​neural network, which has been pre-tuned, outputs candidate operation feedback data of the second example IoT device operation log and second candidate business session topic description value of each business session topic in the second example device operation event. By combining the second candidate business session topic description value and the semantic word vector of the bound business session topic description value, a first performance indicator of the AI ​​neural network is determined; by combining the candidate running feedback data and the semantic word vector of the running feedback data, a second performance indicator of the AI ​​neural network is determined; by combining the first performance indicator and the second performance indicator, a performance evaluation of the AI ​​neural network is generated. The performance evaluation of the AI ​​neural network is invoked to perform a second debugging on the pre-debugged AI neural network until the second debugging completion index is met.

8. The method as described in claim 4, characterized in that, The runtime log feature extraction unit is obtained through debugging steps, which include: The operation log analysis network to be debugged, the operation log of the third-paradigm IoT device, and the log semantic analysis results corresponding to the operation log of the third-paradigm IoT device are identified. The operation log feature extraction unit in the operation log analysis network is invoked to extract IoT device operation log features from the operation log of the third example IoT device, and the bound feature extraction result set is obtained. The topic translation unit in the runtime log analysis network is invoked to perform topic translation on the feature extraction result set to obtain the bound topic text expression; Combining the thematic text expression and the log semantic analysis results, the runtime log feature extraction unit in the runtime log analysis network is subjected to a third debugging process until the third debugging completion indicator is met. The running log feature extraction unit in the running log analysis network after the third debugging is completed is used as the pre-debugged running log feature extraction unit in the AI ​​neural network.

9. A multi-protocol-based Internet of Things (IoT) device control system, characterized in that, The method includes a processor and a memory that communicate with each other, the processor being configured to read a computer program from the memory and execute it to implement the method of any one of claims 1-8.

Images