Label detection method, detection system, electronic device, storage medium and program product for telephone number

Abstract

Embodiments of the present application provide a telephone number label detection method, a detection system, an electronic device, a storage medium and a program product. Among them, the detection server can batch process multiple label detection tasks, assign an adaptive terminal device to each label detection task, trigger the detection client in the terminal device to execute the call record insertion task, and then batch construct multiple telephone number call records and call the call record insertion interface to directly batch insert multiple call records in the call record database locally in the terminal device without SIM card and real call. Subsequently, the detection server interacts with the terminal device using a multi-agent system, and automatically completes the label detection operation for multiple telephone numbers in batches through a knowledge base matching method and an Internet search method. Therefore, the accuracy and reliability of the telephone number label detection are improved.

Description

Technical Field

[0001] This application relates to the field of artificial intelligence technology, and in particular to a method, system, electronic device, storage medium, and program product for detecting phone number tags. Background Technology

[0002] In cloud communication and enterprise-level service scenarios, telephone numbers serve as a crucial communication resource connecting customers and end users. The accuracy and real-time nature of this information directly impacts communication efficiency and service experience. With the maturation of the mobile communication ecosystem, terminal devices have widely adopted number tagging functionality to provide caller identification services. However, due to the dynamic allocation of number resources (such as number recycling and redeployment), if the tagging information stored in the database is inconsistent with the number's current actual ownership (e.g., a number originally belonging to the logistics industry is reassigned to a financial institution, but the tag still displays "delivery" or "express delivery"), information misalignment will occur, affecting the interaction efficiency between the communicating parties. Therefore, establishing an efficient and accurate number tagging status monitoring and governance mechanism is of great significance for maintaining the healthy operation of the communication ecosystem. Summary of the Invention

[0003] This application provides a method, system, electronic device, storage medium, and program product for detecting telephone number tags, in order to provide an efficient and accurate mechanism for monitoring and managing telephone number tagging status.

[0004] This application provides a method for detecting phone number tags, applied to a detection server, comprising: acquiring multiple tag detection tasks, wherein task information of each tag detection task is used to instruct the detection of tags for each phone number among multiple phone numbers; selecting a target terminal device corresponding to a target tag detection task from among the multiple terminal devices based on the task information of the multiple tag detection tasks and device information of multiple terminal devices, wherein the target tag detection task is any one of the multiple tag detection tasks and the target terminal device is any one of the multiple terminal devices; and sending task information of a target call record insertion task to the target terminal device to trigger the detection client in the target terminal device to construct corresponding call records for each of the multiple first phone numbers to be detected by the target tag detection task. The system calls the call log insertion interface to write the call logs corresponding to each of the multiple first phone numbers into the local call log database of the target terminal device. In response to the target terminal device completing the target call log insertion task, the system calls a multi-agent system to interact with the target terminal device to perform a target tag detection task. During the target tag detection task, the multi-agent system obtains screenshot information and / or view hierarchy information of the current call log list interface displayed by the target terminal device. Based on the screenshot information and / or view hierarchy information of the current call log list interface, it obtains the specific text content corresponding to each of the multiple first phone numbers. Based on the specific text content corresponding to each of the multiple first phone numbers, it determines the tag corresponding to each of the multiple first phone numbers sequentially through knowledge base matching and internet search.

[0005] This application also provides an electronic device, including: a memory and a processor; the memory for storing a computer program; and the processor coupled to the memory for executing the computer program to perform steps in a telephone number tag detection method.

[0006] This application also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, enables the processor to implement the steps in the telephone number tag detection method.

[0007] This application also provides a computer program product, including a computer program / instructions, which, when executed by a processor, enable the processor to implement the steps in the telephone number tag detection method.

[0008] The technical solution provided in this application allows the detection server to process multiple tag detection tasks in batches. Each tag detection task is assigned a suitable terminal device. The detection server triggers the detection client on the terminal device to execute a call record insertion task, thereby constructing call records for multiple phone numbers in batches and calling the call record insertion interface to directly insert multiple call records into the local call record database of the terminal device, without requiring a SIM card or relying on real calls. Subsequently, the detection server interacts with the terminal device using a multi-agent system, automatically completing the tag detection operation for multiple phone numbers in batches through knowledge base matching and internet search. Therefore, a phone number tag detection mechanism that does not require a SIM card and does not rely on real calls is provided, thereby establishing an efficient and accurate number tag status monitoring and governance mechanism. This effectively avoids bottlenecks such as signal blind spots, service interruption due to non-payment, and limited concurrency caused by SIM card dependence. It supports batch injection of call records for a single terminal device and parallel detection for multiple terminal devices, improving detection throughput, meeting the high-efficiency detection needs of large batches of phone numbers, and improving the accuracy and reliability of phone number tag detection. Attached Figure Description

[0009] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments of this application and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings: Figure 1 A system architecture diagram of a detection system provided in this application embodiment; Figure 2 This application provides a flowchart of a method for detecting phone number tags; Figure 3 An exemplary interaction process diagram provided for an embodiment of this application; Figure 4 This is an example diagram illustrating device scenario management; Figure 5 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0010] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below in conjunction with specific embodiments and corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0011] The following is a description of some terms used in the embodiments of this application: A language model (LM) is a model that learns general language structures and knowledge through unsupervised or self-supervised pre-training on large-scale text data. Language model architectures include, but are not limited to: bidirectional encoder representation from transformers (BERT), autoregressive language models, and generative pre-trained transformers (GPT). The Transformer module is a neural network structure based on a self-attention mechanism, which significantly improves model performance through parallel processing and self-attention.

[0012] Large Language Models (LLMs), also known as large-scale language models, refer to a class of natural language processing models with an extremely large number of parameters. LLMs are typically based on deep learning architectures, especially the Transformer architecture, which learns the complex structure of language and rich contextual information through pre-training on massive amounts of text data. The Transformer architecture addresses the bottleneck problem of traditional neural network models when processing long sequences by introducing a self-attention mechanism, and its highly parallelizable nature greatly improves training efficiency. The Transformer architecture includes either an encoder or a decoder.

[0013] A knowledge base (KB) is an information system or data collection used to store and organize information. It can save, manage, and access structured or unstructured information.

[0014] A multi-agent system is an architecture in which multiple agents work together, each with independent responsibilities and communicating and collaborating through standardized protocols. It is suitable for handling complex, cross-domain task processes.

[0015] An agent is a core technology component used to extend the capabilities of language models (such as LLMs). Its core idea is to enable language models not only to "answer questions" but also to "perform tasks." By integrating external tools and services, agents can achieve autonomous decision-making, task planning, and execution.

[0016] In cloud communication and enterprise-level service scenarios, telephone numbers serve as a crucial communication resource connecting customers and end users. The accuracy and real-time nature of this information directly impacts communication efficiency and service experience. With the maturation of the mobile communication ecosystem, terminal devices have widely adopted number tagging functionality to provide caller identification services. However, due to the dynamic allocation of number resources (such as number recycling and redeployment), if the tagging information stored in the database is inconsistent with the number's current actual ownership (e.g., a number originally belonging to the logistics industry is reassigned to a financial institution, but the tag still displays "delivery" or "express delivery"), information misalignment will occur, affecting the interaction efficiency between the communicating parties. Therefore, establishing an efficient and accurate number tagging status monitoring and governance mechanism is of great significance for maintaining the healthy operation of the communication ecosystem.

[0017] In cloud communication and enterprise-level service scenarios, phone numbers are critical communication resources, and their connection rate directly affects the reach and service experience of end users. If phone numbers are labeled with tags such as "sales calls," "recruitment calls," or "delivery / takeout calls" by terminal devices such as mobile phones, or are incorrectly associated with unrelated organization names (e.g., a phone number labeled "Company A" is assigned to Company B), it may negatively impact customers. Therefore, it is necessary to manage these number resources.

[0018] Currently, common phone number tag detection schemes rely on real mobile devices with a physical SIM (Subscriber Identity Module) card inserted. The calling device initiates an outbound call by dialing a number, and the called device displays the caller ID interface, which shows the caller's phone number and its tag. A screenshot of the called device's caller ID interface is taken, and OCR (Optical Character Recognition) technology is used to identify the phone number tag from the screenshot.

[0019] However, the above solutions face multiple bottlenecks in actual large-scale applications. For example, SIM cards are susceptible to factors such as cellular signal blind spots, service suspension due to unpaid bills, unilateral account cancellation by operators, and physical damage, resulting in call failures and poor detection reliability. A single mobile device can only perform a call operation on one phone number at a time, resulting in weak horizontal scalability and limited concurrent detection capabilities, which cannot meet the needs of efficient detection of a large number of phone numbers.

[0020] The technical solutions of this application and how they solve the aforementioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The technical solutions provided by each embodiment of this application are described in detail below with reference to the accompanying drawings.

[0021] Figure 1 This is a system architecture diagram of a detection system provided in an embodiment of this application. See also... Figure 1 The detection system may include a detection server and a cluster of terminal devices. The cluster of terminal devices includes multiple terminal devices, such as terminal device 1, terminal device 2, ..., terminal device n, where n is a positive integer.

[0022] The testing server is primarily responsible for device management and task management. Servers can include, but are not limited to, single servers, distributed server clusters consisting of multiple servers, cloud servers, virtual servers, containerized servers, or edge servers.

[0023] Specifically, the detection server can be responsible for device management such as scene management, activity detection management, terminal status management or terminal version management, as well as task management such as device scheduling, task execution priority management or report management. The detection server can also provide a multi-agent system for task management.

[0024] The terminal equipment mainly refers to communication terminals capable of making voice calls based on telephone numbers, including but not limited to smartphones, wearable communication devices (such as smartwatches), and in-vehicle terminals. The terminal equipment has a corresponding testing client installed, which works in conjunction with the testing server to provide testing services.

[0025] It should be noted that, Figure 1 The detection system shown is only an example. In actual applications, there is no limit to the number of detection servers and terminal device clusters.

[0026] Figure 2 This application provides a flowchart of a method for detecting phone number tags. This method is applied to a detection server; see [link to relevant documentation]. Figure 2 The method may include the following steps: 201. Obtain multiple tag detection tasks. The task information for each tag detection task is used to instruct the detection of tag labels for each of the multiple phone numbers.

[0027] Specifically, the detection server can acquire multiple tag detection tasks in batches. The task information for each tag detection task is used to instruct the tag detection of each phone number among multiple phone numbers. The task information for each tag detection task includes, but is not limited to: the list of phone numbers to be detected, the source channel, the required device model, the urgency of the task, etc.

[0028] The list of phone numbers to be detected includes multiple phone numbers, some of which may have been tagged with labels, such as, but not limited to, intermediary calls, sales calls, or delivery notifications.

[0029] The source channel of the tag detection task is used to indicate the triggering reason or initiator of the tag detection task. The source channel of the tag detection task is, for example, but not limited to: new number activation, existing number inspection, customer manual triggering, etc.

[0030] The "new number activation" channel refers to the automatic initiation of a tag detection task when a cloud communication service provider activates a new phone number for a user. This task verifies whether the new phone number has been incorrectly tagged as a brokerage or sales call due to historical reuse or other reasons, thereby ensuring smooth communication for new users.

[0031] The "existing number inspection" channel refers to the periodic batch inspection of phone numbers that have been in use for a period of time. This is used to continuously monitor their markings on various terminal devices and promptly detect mislabeling, omissions, or malicious tampering.

[0032] The "customer-triggered" channel refers to the tag detection task triggered by enterprise customers to verify the accuracy of their phone number tags, ensure that customer service or marketing outbound calls are not intercepted or misunderstood, and improve user reach and brand image.

[0033] Among them, the device requirement information refers to the requirements for the device model of the terminal device when the tag detection task is executed. The device requirement information includes, but is not limited to: device brand, device model, operating system type and version, screen resolution, etc.

[0034] Among them, the task urgency level is used to reflect the urgency of the tag detection task needing to be scheduled and executed as soon as possible.

[0035] In some optional embodiments, the detection server can store triggered tag detection tasks in a task queue and retrieve multiple tag detection tasks from the task queue. The task queue can be implemented based on a message middleware, which can support task buffering, decoupling, asynchronous processing, and reliable delivery.

[0036] In some optional embodiments, the acquisition of multiple tag detection tasks is achieved by: acquiring multiple tag detection tasks with high task execution priority from the task queue; wherein, the task execution priority of the tag detection tasks in the task queue is determined according to the source channel of the tag detection tasks.

[0037] Specifically, the detection server can determine the execution priority of tag detection tasks based on the source channel of the task and the pre-configured mapping relationship between source channels and task execution priorities. This allows the detection server to schedule tasks based on their execution priority. The source channel-based task priority scheduling mechanism ensures that high-value or time-sensitive tasks are processed first, effectively improving the flexibility of task scheduling. For example, tasks ordered from highest to lowest execution priority could be: tag detection tasks originating from "new number activation," tag detection tasks originating from "customer manual triggering," and tag detection tasks originating from "existing number inspection."

[0038] 202. Based on the task information of multiple tag detection tasks and the device information of multiple terminal devices, select the target terminal device corresponding to the target tag detection task from the multiple terminal devices. The target tag detection task can be any one of the multiple tag detection tasks, and the target terminal device can be any one of the multiple terminal devices.

[0039] Specifically, a terminal device cluster can contain multiple terminal devices. The detection server can select the appropriate terminal device for each tag detection task. The task information for each tag detection task includes, but is not limited to: a list of phone numbers to be detected, source channels, device model requirements, task urgency, etc. Terminal device information includes, but is not limited to: device brand, device model, operating system type and version, screen resolution, and current idle status. Based on precise matching and efficient scheduling of task characteristics and device capabilities, the detection server meets the compatibility requirements of the task with the terminal environment, improving the accuracy and response efficiency of the overall detection service.

[0040] In practical applications, when the detection server selects a suitable terminal device for each tag detection task, it can first filter out compatible candidate terminal devices from the terminal device cluster based on the device model requirements and source channels; then, it can select a suitable terminal device for the tag detection task by combining the urgency of the task and the current idle status of the candidate terminal devices (whether they are executing other tasks, resource utilization, etc.).

[0041] For example, suppose there are three tag detection tasks: Task A: originates from "New Number Activation," the number to be detected is A, the device requirement is "must use a phone of brand A," and the task urgency is "high." Task B: originates from "Customer Manual Trigger," the number to be detected is B, the device requirement is "a phone with operating system B," and the task urgency is "medium." Task C: originates from "Existing Number Inspection," the number to be detected is C, there is no specific device requirement, and the task urgency is "low." Meanwhile, the terminal device cluster contains the following four phones: Phone 1: Brand A, Operating System A, currently idle; Phone 2: Brand C, Operating System B, currently busy (performing other tasks); Phone 3: Brand D, Operating System B, currently idle; Phone 4: Brand A, Operating System A, currently idle. Finally, Task A is assigned to Phone 1; Task B is assigned to Phone 3; and Task C is assigned to Phone 4.

[0042] In some optional embodiments, the method of selecting the target terminal device corresponding to the target tag detection task from multiple terminal devices based on the task information of multiple tag detection tasks and the device information of multiple terminal devices is as follows: the target terminal device corresponding to the target tag detection task is selected from multiple terminal devices based on the task information of multiple tag detection tasks, the device information of multiple terminal devices, and the pre-configured mapping relationship between the source channel of the tag detection task and the model of the terminal device.

[0043] Specifically, the detection system can maintain a mapping relationship between the source channels of tag detection tasks and the models of terminal devices. Based on the detection target, it can select the terminal device with the matching model to perform the tag detection task as needed, ensuring the coverage and accuracy of tag detection.

[0044] For example, during task scheduling, the mapping relationship can be queried based on the source channel of the target tag detection task to obtain the terminal devices of candidate models; then, by combining the task information of the target tag detection task and the device information of the terminal devices of the candidate models, a compatible and idle target terminal device can be selected for the target tag detection task.

[0045] 203. Send the task information of the target call record insertion task to the target terminal device to trigger the detection client in the target terminal device to construct the corresponding call record for each of the multiple first phone numbers to be detected by the target tag detection task, and call the call record insertion interface to write the call records corresponding to the multiple first phone numbers into the local call record database of the target terminal device.

[0046] Specifically, the detection server determines the task information for the target call record insertion task corresponding to the target tag detection task based on the task information of the target tag detection task. For ease of understanding and distinction, the phone number to be detected by the target tag detection task is referred to as the first phone number. The task information for the target call record insertion task may include, but is not limited to, multiple first phone numbers, the call start time of each first phone number, the call duration, the contact name, and the call type, which may include, but is not limited to, incoming calls, outgoing calls, or missed calls.

[0047] The target call log insertion task instructs the detection client on the target terminal device to construct corresponding call logs for each of the multiple first phone numbers to be detected by the target tag detection task, and to call the call log insertion interface to write the call logs corresponding to each of the multiple first phone numbers into the local call log database of the target terminal device. The call log insertion interface refers to the program interface provided by the operating system for writing new call logs to the call log database; for example, the Callloginsert interface provided by the operating system.

[0048] In this embodiment, even if the terminal device does not have a SIM card installed, the detection client can still construct call records without generating call records through actual outbound calls. In this way, the detection client can generate call records for multiple phone numbers in batches and use the call record insertion interface to write multiple call records in batches. This significantly improves efficiency compared to generating and writing call records one by one, and increases the number processing throughput of a single round of tasks.

[0049] Optionally, after batch writing multiple call records, the detection client can also perform integrity verification on the writing results. Call records that fail to write can be automatically recorded and reported to the quality inspection server for further processing. This mechanism effectively improves the reliability, observability, and fault tolerance of task execution while ensuring high-throughput batch writing efficiency, thus ensuring the integrity and accuracy of the tag detection results.

[0050] Specifically, when the detection client performs integrity verification on the write results, it compares the expected number of call records to be written with the actual number of call records successfully written. The expected number of call records to be written refers to the number of call records that need to be written to the local call record database of the target terminal device, which is also the number of first phone numbers that the target tag detection task needs to detect; the actual number of call records successfully written refers to the number of call records successfully written to the local call record database of the target terminal device. If the expected number of call records to be written is the same as the actual number of call records successfully written, the write is successful and the integrity verification passes; if the expected number of call records to be written is greater than the actual number of call records successfully written, the write fails and the integrity verification fails. The detection client can automatically report the failure context to the detection server. The failure context includes, but is not limited to, the device information of the terminal device where the write failed, the number of call records that failed to be written, and the phone numbers that failed to be written. Based on the failure context, the detection server retryes the detection task or alarms for the phone numbers that failed to be written on other compatible terminal devices. For example, the detection server can re-detect phone numbers that failed to write in subsequent tag detection tasks.

[0051] Among them, the terminal device that failed to write refers to the terminal device corresponding to the phone number that failed to write; the phone number that failed to write refers to the phone number that was not successfully written to the call record database among the phone numbers that the target tag detection task needs to detect; the number of call records that failed to write is the difference between the expected number of call records to be written and the actual number of call records that were successfully written.

[0052] 204. In response to the target terminal device completing the target call record insertion task, the multi-agent system is invoked to interact with the target terminal device to perform the target tag detection task.

[0053] Specifically, after the detection client on the target terminal device completes the target call record insertion task, it notifies the detection server that the target terminal device has completed the target call record insertion task. After confirming that the target terminal device has completed the target call record insertion task, the detection server invokes the multi-agent system to interact with the target terminal device to perform the target tag detection task.

[0054] In practical applications, if the detection client confirms that all the phone numbers required for the target call record insertion task have been inserted into the local call record database of the target terminal device, the detection client confirms that the target call record insertion task has been completed.

[0055] Optionally, the detection client may confirm that some phone numbers among all the phone numbers required for the target call record insertion task have not been inserted into the local call record database of the target terminal device, and the detection client may also confirm that the target call record insertion task has been completed. However, in this case, the detection client can automatically report a failure context to the detection server. The failure context may include, but is not limited to, the device information of the terminal device where the failed phone numbers were written, the number of failed call records, and the failed phone numbers. Based on the failure context, the detection server can retry the detection task for the failed phone numbers or perform subsequent processing such as alarms on other compatible terminal devices. For example, it can rebuild and insert the call records for the failed phone numbers into the local call record database on other compatible terminal devices and perform a tagging detection task.

[0056] Understandably, after completing the target call record insertion task, the detection client can confirm the completion of the task, regardless of whether all or only some phone numbers are successfully written to the local call record database of the target terminal device. If some numbers fail to be written, the detection client will automatically report the failure context to the detection server. Based on this, the detection server will intelligently retry on other compatible terminal devices, reconstruct and insert the call records of the failed phone numbers, and then trigger the multi-agent system to execute the corresponding tag detection task or perform subsequent processing such as alarms. This mechanism ensures the continuous progress of the task process and improves the adaptability to abnormal scenarios, detection coverage, and overall task reliability through accurate capture of failure data and cross-device retry strategies.

[0057] In this embodiment, the multi-agent system can directly interact with the target terminal device to perform target tag detection tasks. Interaction operations include, but are not limited to: interface operations, screenshot operations, view hierarchy acquisition, pop-up window handling, and swipe page turning.

[0058] Interface operations include, but are not limited to: launching the dialer application, clicking to trigger the display of the call log list interface, and returning to the previous interface. Screenshot operations include, but are not limited to: taking a screenshot of the call log list interface, or taking a screenshot of the call details interface of a specific phone number.

[0059] A dialer application is a system-level application installed on a terminal device (such as a smartphone) to enable core communication functions such as making phone calls, managing call logs, viewing contacts, and handling incoming calls.

[0060] The view hierarchy is a tree-like structure that organizes all UI (User Interface) controls (such as text, buttons, list items, etc.) in the user interface according to the parent-child nesting relationship. With the help of the view hierarchy, the key attribute information of each UI control in the call log list interface or call details interface can be extracted. The key attribute information includes, but is not limited to: text content, layout position, visibility, control type, etc.

[0061] Pop-up handling includes, but is not limited to: automatically identifying and disabling intrusive pop-ups such as permission prompts, advertising overlays, and update reminders to ensure that the main interface content can be collected normally; and swiping page turning is used to automatically trigger interface scrolling when the call log list is too long, so as to cover all the phone numbers to be detected.

[0062] In this embodiment, the multi-agent system acquires specific text content corresponding to multiple first phone numbers during the interaction process; and determines the respective tags of multiple first phone numbers based on the specific text content corresponding to each of the multiple first phone numbers. The specific text content is a text fragment that may include the tags in the text content associated with the first phone number. For example, the specific text content is "real estate agent phone number", "delivery and takeout", "real estate agent", etc.

[0063] In this embodiment, during the target tag detection task, the multi-agent system acquires screenshot information and / or view hierarchy information of the current call log list interface displayed by the target terminal device; based on the screenshot information and / or view hierarchy information of the current call log list interface, it acquires specific text content corresponding to each of the multiple first phone numbers; based on the specific text content corresponding to each of the multiple first phone numbers, it sequentially determines the tag corresponding to each of the multiple first phone numbers through knowledge base matching and Internet search.

[0064] In practical applications, the multi-agent system can trigger the target terminal device to display the current call log list interface and take a screenshot of it, obtaining screenshot information of the current call log list interface; it can also obtain the view hierarchy structure information of the current call log list interface; and perform Optical Character Recognition (OCR) on the screenshot information of the current call log list interface to extract the first phone number displayed in the current call log list interface and its associated specific text content (such as tags like "real estate agent phone number" and "delivery"); it can also extract the specific text content of each first phone number from the view hierarchy structure information of the current call log list interface. Next, the multi-agent system first searches the knowledge base for tags that match the specific text content of the first phone number as the tag corresponding to the first phone number; if no tag matching the specific text content of the first phone number is found in the knowledge base, it searches for tags matching the specific text content of the first phone number on the internet platform.

[0065] In this embodiment, the multi-agent system can automatically trigger and capture the current call log list interface and obtain its view hierarchy structure by coordinating the operation of the target terminal device. It can accurately extract the specific text content of the first phone number by combining the screenshot and the view hierarchy structure. It can first search for matching tags in the knowledge base. If the tags are not found in the knowledge base, it will link the Internet platform for real-time query, thereby efficiently and accurately detecting the tags of the phone number.

[0066] In practical applications, there are no restrictions on the structure of the multi-agent system. Optionally, the multi-agent system includes a master agent and sub-agents. The implementation method for calling the multi-agent system to interact with the target terminal device to perform the target tag detection task is as follows: the master agent interacts with the target terminal device to collect the specific text content corresponding to each of the multiple first phone numbers; the sub-agents determine the tag for each of the multiple first phone numbers based on the specific text content corresponding to each of the multiple first phone numbers.

[0067] Specifically, the multi-agent system, through a collaborative architecture in which the main agent is responsible for "perception and acquisition" and the sub-agents are responsible for "semantic decision-making", can achieve automated, highly accurate and large-scale execution of tag detection tasks.

[0068] In practical applications, multi-agent systems can also perform text extraction integrity verification, telephone number consistency verification, and outlier filtering on information collected during the interaction process, ensuring the accuracy and reliability of tag detection.

[0069] Optional, see Figure 3The method for collecting specific text content corresponding to multiple first phone numbers by interacting with the target terminal device through the main intelligent agent includes the following steps S1-S5: S1. Trigger the target terminal device to display the current call log list interface through the main intelligent agent.

[0070] Specifically, the main intelligent agent launches the dialing application of the target terminal device by calling the operating system interface or UI automation tools of the target terminal device, and automatically navigates to the call log list interface. The call log list interface is an interface that displays the call log list, and the call logs in the call log list are the call logs that have been written to the call log database.

[0071] S2. Trigger the target terminal device through the main intelligent agent to take a screenshot of the current call log list interface and / or obtain the view hierarchy information of the current call log list interface.

[0072] Specifically, the main agent takes a screenshot of the current call log list interface by calling the screenshot interface provided by the target terminal device's operating system or the screenshot function built into the UI automation tool. The main agent can then obtain the view hierarchy information of the current call log list interface by calling the relevant interfaces provided by the target terminal device's operating system.

[0073] In some optional embodiments, before taking a screenshot of the current call log list interface, the main intelligent agent can determine whether there is a pop-up window on the current call log list interface; if there is a pop-up window, the main intelligent agent can trigger the target terminal device to close the pop-up window on the current call log list interface.

[0074] Specifically, after the target terminal device displays the current call log list interface, it identifies and closes pop-ups that may obscure the content. This preprocessing mechanism effectively avoids the problem of missed or incorrect collection of phone numbers or tags due to pop-up obstruction, thereby significantly improving the coverage and accuracy of tag extraction.

[0075] S3. Based on the screenshot information and / or view hierarchy information of the current call log list interface, the main intelligent agent extracts specific text content from at least one second phone number in the current call log list interface to obtain specific text content of at least one third phone number. The second phone number is any one of a plurality of first phone numbers, and the third phone number is any one of at least one second phone number.

[0076] Specifically, the main agent performs Optical Character Recognition (OCR) on the screenshot of the current call log list interface, extracting the phone numbers displayed in the interface and their associated specific text content (such as tags like "real estate agent phone number" or "delivery / takeout"). The main agent can also extract the specific text content of each phone number from the view hierarchy of the current call log list interface.

[0077] S4. The main intelligent agent determines whether the collection of specific text content from multiple first phone numbers has been completed.

[0078] Specifically, if the judgment result is negative, proceed to step S5. If the judgment result is positive, end the entire process. The number of first phone numbers to be detected by the target tag detection task is called the total number, and the number of first phone numbers whose specific text content has been collected is called the collected number. The main agent compares the total number and the collected number. If the total number is greater than the collected number, the collection of specific text content for multiple first phone numbers has not yet been completed. If the total number equals the collected number, the collection of specific text content for multiple first phone numbers is complete.

[0079] S5. Trigger the target terminal device to slide the current call log list interface through the main intelligent agent.

[0080] Specifically, the main agent can simulate user gestures (such as swiping up or scrolling quickly) to trigger the target terminal device to swipe the current call log list interface to load more call logs; after the swiping operation is completed, return to execute step S2 until the specific text content of multiple first phone numbers is collected.

[0081] Optionally, before the main agent determines whether the collection of specific text content from multiple first phone numbers has been completed, the method further includes: for any fourth phone number whose specific text content extraction fails, and the fourth phone number is any one of at least two second phone numbers, the main agent triggers the target terminal device to display the call details interface of the fourth phone number, and takes a screenshot of the call details interface of the fourth phone number and / or obtains the view hierarchy information of the call details interface of the fourth phone number; the main agent extracts specific text content from the fourth phone number based on the screenshot information and / or view hierarchy information of the call details interface of the fourth phone number.

[0082] Specifically, by utilizing the call details interface of phone numbers, the success rate of collecting specific text content can be improved, thereby ensuring high coverage and high accuracy of the tag detection task.

[0083] In this embodiment, when the main agent does not mention the specific text content of the fourth phone number based on the screenshot information and / or view hierarchy information of the current call log list interface, the main agent triggers the target terminal device to enter the call details interface of the fourth phone number. Subsequently, a screenshot of the call details interface and / or its view hierarchy information are obtained, and the specific text content of the fourth phone number is extracted again based on this information. The call details interface refers to the detailed information page displayed after clicking on a call log in the dialing application of the target terminal device. This interface can display more additional information related to the phone number.

[0084] In this embodiment, the main intelligent agent can automatically trigger the target terminal device to display the interface, combine screenshots and view hierarchy information parsing, accurately extract specific text content of multiple phone numbers, and achieve the coverage collection of phone numbers by sliding page turning, effectively ensuring the automation, high accuracy and large-scale execution capability of the tag detection task.

[0085] In this embodiment, the sub-agent is responsible for determining the respective tagging labels of the first phone number based on the specific text content corresponding to the first phone number. For example, the sub-agent driven by a large language model performs semantic understanding on the specific text content corresponding to the first phone number to determine the respective tagging labels of the first phone number.

[0086] In some optional embodiments, determining the tagging labels of multiple first phone numbers based on their respective specific text content by a sub-agent includes: the sub-agent performing the following operations: for any given first phone number, matching its specific text content with multiple tags stored in a knowledge base; if a tag successfully matches the specific text content of the first phone number, using the matched tag as the tagging label of the first phone number; if no tag successfully matches the specific text content of the first phone number, inputting the specific text content of the first phone number into an internet platform for searching and obtaining search results; determining the tagging label of the first phone number based on the search results; if the tagging label of the first phone number cannot be determined based on the search results, prompting for manual determination of the tagging label of the first phone number.

[0087] Specifically, the knowledge base can store multiple tags. Optionally, the knowledge base can categorize and store tags according to semantic attributes, including but not limited to the following categories: negative tags, neutral tags, and positive tags. Negative tags include, but are not limited to, "real estate agent phone number," "advertising and sales," etc.; neutral tags include, but are not limited to, "delivery," "ride-hailing," "food delivery," etc.; positive tags include, but are not limited to, "customer service of xxx company," "customer service of xxx group," etc. By categorizing and storing tags, the knowledge base not only supports sub-agents in efficiently matching specific text content but also assists them in risk level assessment, thereby improving the overall accuracy of tag determination and user experience.

[0088] In this embodiment, the sub-agent first performs semantic or keyword matching between the specific text content of the first phone number and the tags in the knowledge base. If the match fails, it automatically uses the specific text content of the first phone number as a search term to search on an internet platform and infers a reasonable tag based on the search results. If it still cannot be determined, it triggers manual review. In this way, the sub-agent adopts a three-level progressive judgment mechanism of "knowledge base matching - internet search - manual backup," which effectively integrates knowledge information in the knowledge base, open network information, and human experience, improving the accuracy and reliability of tag judgment while ensuring high automation efficiency.

[0089] The technical solution provided in this application allows the detection server to process multiple tag detection tasks in batches. Each tag detection task is assigned a suitable terminal device. The detection server triggers the detection client on the terminal device to execute a call record insertion task, thereby constructing call records for multiple phone numbers in batches and calling the call record insertion interface to directly insert multiple call records into the local call record database of the terminal device, without requiring a SIM card or relying on real calls. Subsequently, the detection server uses a multi-agent system to interact with the terminal device to automatically complete the tag detection operation for multiple phone numbers in batches. This provides a phone number tag detection mechanism that does not require a SIM card or rely on real calls, thus establishing an efficient and accurate number tag status monitoring and management mechanism. It effectively avoids bottlenecks such as signal blind spots, service interruption due to non-payment, and limited concurrency caused by SIM card dependence. It supports batch injection of call records by a single terminal device and parallel detection by multiple terminal devices, improving detection throughput, meeting the high-efficiency detection needs of large batches of phone numbers, and improving the accuracy and reliability of phone number tag detection.

[0090] In some optional embodiments, the detection server is further configured to: detect the task execution status of the target tag detection task; if an abnormality is detected in the execution of the target tag detection task, return to the step of selecting the target terminal device corresponding to the target tag detection task from multiple terminal devices based on the task information of multiple tag detection tasks and the device information of multiple terminal devices, until the maximum number of retries is reached.

[0091] Specifically, the detection server continuously monitors the execution status of the target tag detection task. If it detects a timeout, call log writing failure, or failure to collect phone number-related data, it automatically triggers a retry process. The retry strategy supports configuring a maximum number of retries and can dynamically switch terminal devices during the retry process, effectively avoiding detection distortion caused by single terminal failures, thereby improving the overall fault tolerance, execution stability, and result reliability of the detection task.

[0092] In some optional embodiments, the detection server may further: receive a failure context reported by the detection client, wherein the detection client performs integrity verification on the write result of the target call record insertion task; if the write result of the target call record insertion task fails the integrity verification, it reports a failure context to the detection server; the failure context includes: device information, the number of call records that failed to be written, and the phone numbers that failed to be written; and update the task execution status and task information of the target tag detection task according to the failure context, so as to rewrite the phone numbers that failed to be written into the call record database to continue tag detection.

[0093] In practical applications, upon receiving a failure context, the detection server can update the target tag detection task's execution status to "execution failed" or "partially failed." The updated task information includes: device information of the terminal device that failed to write, the number of call records that failed to write, and the phone numbers that failed to write. The detection server can treat the updated target tag detection task as a new tag detection task. Based on the new task's task information and the device information of multiple terminal devices, it assigns a corresponding terminal device to the new task and sends a call record insertion task to that terminal device. This triggers the detection client on the corresponding terminal device to construct the corresponding call records for the phone numbers to be detected by the new tag detection task and calls the call record insertion interface to write the call records corresponding to the phone numbers into the terminal device's local call record database. In response to the terminal device completing the call record insertion task, the detection server calls a multi-agent system to interact with the terminal device to execute the new tag detection task.

[0094] In practical applications, a new tag detection task regenerated due to a failed write of a phone number corresponding to the target tag detection task can be seen as a retry of the target tag detection task. Each time a new tag detection task is generated, the retry count for the target tag detection task is incremented by 1. Once the maximum retry count is reached, no further retrying is performed. This combination of retry count and maximum retry count control ensures retry opportunities for failed phone number writes while avoiding resource waste caused by infinite retries.

[0095] In this embodiment, by combining the failure context (including device information, number of failures and specific phone numbers) reported by the detection client, the detection server can accurately update the task status and task information, repackage the failure context into a new tagging task and intelligently allocate it to available terminal devices, and re-execute the call record construction and tagging detection, thereby improving the overall fault tolerance, task completion rate and overall detection result reliability of the detection task.

[0096] In some optional embodiments, the detection server is further configured to: perform environmental integrity detection on the new terminal device in response to the addition of a new terminal device to the terminal device cluster; if the new terminal device does not meet the environmental integrity conditions, repair the new terminal device with the goal of making the new terminal device meet the environmental integrity conditions; and bind a purpose tag to the new terminal device.

[0097] Specifically, when a new terminal device connects to the terminal device cluster, the detection server automatically performs an environment integrity check. This check includes, but is not limited to, checking whether the detection client is installed, whether the client version is the latest, whether necessary system permissions (call log read / write, notification access, etc.) have been granted, and whether the client is configured to start automatically. If the detection client is installed and is the latest version, necessary system permissions have been granted, and the client is configured to start automatically, the new terminal device is confirmed to meet the environment integrity requirements. If the detection client is not installed, its version is not the latest, necessary system permissions (call log read / write, notification access, etc.) have not been granted, or the client is not configured to start automatically, the new terminal device does not meet the environment integrity requirements. The server then repairs the new terminal device to ensure it meets these requirements. In this way, the detection server automatically performs repair operations for detected environmental defects without manual intervention, ensuring that new terminal devices meet the unified operating environment requirements when connecting to the terminal device cluster. This provides a stable and consistent guarantee of terminal resources for large-scale, highly reliable automated detection tasks.

[0098] In practical applications, administrators can monitor the entire lifecycle of terminal devices, from access and initialization to maintenance and decommissioning, through the monitoring server. Administrators can also perform task pause / resume operations on single or batches of terminal devices using the monitoring server. Paused terminal devices will no longer receive new task assignments, but tasks already in progress are allowed to run until completion. Upon resumption, the device automatically rejoins the pool of available terminal devices.

[0099] In this embodiment, after a new terminal device joins the terminal device cluster, the detection server can bind a usage tag to the new terminal device. The usage tag can indicate the usage scenario of the terminal device, and usage tags include, but are not limited to, "Number Tagging," "SMS Channel Detection," and "General Scenarios." The terminal device corresponding to the "Number Tagging" tag is used to perform the phone number tag detection task. The terminal device corresponding to the "SMS Channel Detection" tag is used to perform the SMS channel detection task; the terminal device corresponding to the "General Scenarios" tag is used to perform various tasks.

[0100] In practical applications, a device scenario management model is used to manage various information such as purpose tags and task execution status of each terminal device, achieving scenario-based isolation of terminal device resources. Terminal devices bound to purpose tags only respond to task instructions of the scenario type corresponding to that purpose tag, achieving resource isolation for task execution in different scenarios and avoiding cross-interference. Terminal devices not bound to scenarios serve as a general resource pool and can be dynamically requisitioned by tasks of any scenario type.

[0101] by Figure 4 Taking the device scenario management diagram shown as an example, when a device is first connected, that is, a new terminal device joins the terminal device cluster, the new terminal device has not yet been assigned a scenario, that is, it has not yet been bound to a purpose tag. At this time, the new terminal device can be assigned a general scenario by default, that is, the new terminal device is bound to a "general scenario" tag. Alternatively, the administrator can assign the purpose tag of the new terminal device to the "SMS channel detection" tag, that is, the new terminal device is used in the SMS channel detection scenario. Of course, the administrator can also assign the purpose tag of the terminal device to the purpose tag corresponding to other scenarios.

[0102] In practical applications, administrators can assign a new terminal device the "Number Search" tag, meaning the new terminal device can be used in number search scenarios. Of course, administrators can also unbind the new terminal device from the "Number Search" tag, i.e., unbind it from the scenario.

[0103] The new terminal device is labeled "Number Tag Lookup". It can receive tag lookup tasks (i.e., tag detection tasks). The new terminal device begins executing the tag detection task, and the device scenario management model records the task execution status, such as task in progress and task completion. Of course, the administrator can pause and resume task execution on the new terminal device, and the task execution status is maintained in real-time in the device scenario management model.

[0104] Of course, in practical applications, terminal devices labeled "General Scenarios" can be dynamically requisitioned for number lookup scenarios, SMS channel detection scenarios, or other scenarios without restriction.

[0105] It is worth noting that the administrator sends an allocation request to the detection server, which then assigns user tags.

[0106] In some optional embodiments, the detection server and the terminal device can establish a reliable communication link, supporting both wired and wireless channels as complementary mechanisms to ensure real-time perception of device status and reliable transmission of tasks.

[0107] When the wired connection between the detection server and the terminal device is stable, the detection server performs device status detection. When only the wireless connection is stable, the terminal device's detection client performs device status detection and sends the device status to the detection server. In this way, dual-channel device detection ensures that backup tasks can be executed even when the wired connection is unstable.

[0108] The following is a brief introduction to the heartbeat keep-alive mechanism: The detection client periodically sends heartbeat packets to the detection server. The heartbeat packets may contain information including but not limited to the following: a device identifier that uniquely identifies the terminal device, current network connectivity, device battery level, detection client version number, current task execution status, and other multi-dimensional information, so that the detection server can monitor the health status of the terminal device cluster in real time.

[0109] The detection server maintains the last heartbeat timestamp for each terminal device. When no reported heartbeat packet is received for N consecutive heartbeat cycles, the detection server automatically marks the terminal device's status as offline and triggers the automatic migration of tasks to be executed on that terminal device, ensuring that the continuity of task execution is not affected by single-point device failure. Here, N is a positive integer that can be set as needed.

[0110] In some optional embodiments, the detection server can periodically trigger the terminal device to update the local call record database to ensure the timeliness and accuracy of the phone number tagging and avoid missed or false detections due to the expiration of the call record database.

[0111] In some optional embodiments, the detection server can proactively issue a cleanup command to trigger the terminal device to delete call records from its local call log database. Alternatively, the detection client can periodically trigger the terminal device to delete call records from its local call log database. This dual-trigger cleanup mechanism can doubly ensure the timeliness of the cleanup operation.

[0112] In some optional embodiments, the detection server is also used to: periodically detect the number of constructed call records in the local call record database of the target terminal device; if the number of constructed call records is greater than a set threshold, then delete the number of constructed call records in the local call record database of the target terminal device.

[0113] Specifically, the detection server can also periodically detect the number of constructed call records in each target terminal device. If the number exceeds the set threshold (such as 100 records), a cleanup operation will be automatically triggered to delete all or part of the constructed call records. This prevents a large number of constructed call records from accumulating over a long period, occupying system resources, affecting the performance of dialing applications, or interfering with the accuracy of subsequent detection tasks, so that the call record environment on the terminal device is always in a controllable and clean state.

[0114] In the embodiments of this application, "at least one" refers to one or more, and "more than one" refers to two or more. "And / or" describes the access relationship between associated objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone, where A and B can be singular or plural. In the textual description of this application, the character " / " generally indicates that the preceding and following associated objects have an "or" relationship. Furthermore, in the embodiments of this application, "first," "second," "third," etc., are only used to distinguish the content of different objects and have no other special meaning.

[0115] It should be noted that, in the cases involving user information in the embodiments of this application, the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in the embodiments of this application are all information and data authorized by the user or fully authorized by all parties. Furthermore, the collection, use, and processing of related data must comply with the relevant laws, regulations, and standards of the relevant countries and regions, and corresponding operation entry points are provided for users to choose to authorize or refuse. In addition, the various models involved in this application (including but not limited to language models or large models) comply with relevant laws and standards.

[0116] It should be noted that the execution subject of each step of the method provided in the above embodiments can be the same device, or the method can be executed by different devices. For example, the execution subject of steps 201 to 204 can be device A; or the execution subject of steps 201 and 202 can be device A, and the execution subject of steps 203 to 204 can be device B; and so on.

[0117] Furthermore, in some of the processes described in the above embodiments and accompanying drawings, multiple operations appear in a specific order. However, it should be clearly understood that these operations may not be executed in the order they appear herein, or they may be executed in parallel. The operation numbers, such as 201, 202, etc., are merely used to distinguish different operations and do not represent any execution order. Additionally, these processes may include more or fewer operations, and these operations may be executed sequentially or in parallel. It should be noted that the descriptions such as "first" and "second" in this document are used to distinguish different messages, devices, modules, etc., and do not represent a sequential order, nor do they limit "first" and "second" to different types.

[0118] Figure 5 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Figure 5 As shown, the electronic device includes: a memory 51 and a processor 52; Memory 51 is used to store computer programs and can be configured to store various other data to support operation on the computing platform. Examples of this data include instructions for any application or method operating on the computing platform, data structures, contact data, phone book data, messages, pictures, videos, etc.

[0119] Processor 52, coupled to memory 51, is used to execute a computer program in memory 51 for: performing steps in a telephone number tag detection method.

[0120] Optional, such as Figure 5 As shown, the electronic device also includes other components such as a communication component 53, a display 54, a power supply component 55, and an audio component 56. Figure 5 The diagram only shows some components and does not mean that the electronic device includes only these components. Figure 5 The components shown. Additionally... Figure 5The components within the dashed box are optional, not mandatory, and their specific requirements depend on the product form of the electronic device. The electronic device in this embodiment can be a desktop computer, laptop computer, smartphone, or IoT (Internet of Things) device, or a server-side device such as a conventional server, cloud server, or server array. If the electronic device in this embodiment is a desktop computer, laptop computer, or smartphone, it may include... Figure 5 The components within the dashed box; if the electronic device in this embodiment is implemented as a conventional server, cloud server, or server array, etc., it may be omitted. Figure 5 The component within the dashed box.

[0121] The aforementioned memory can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random-Access Memory (SRAM), Electrically Erasable Programmable Read Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Read-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0122] The aforementioned communication component is configured to facilitate wired or wireless communication between the device containing the communication component and other devices. The device containing the communication component can access wireless networks based on communication standards, such as 2G (2nd Generation), 3G (3rd Generation), 4G (4th Generation) / LTE (long Term Evolution), 5G (5th Generation), or combinations thereof. In one exemplary embodiment, the communication component receives broadcast signals or broadcast-related information from an external broadcast management system via a broadcast channel.

[0123] The aforementioned display includes a screen, which may include a Liquid Crystal Display (LCD) and a Touch Panel (TP). If the screen includes a Touch Panel, the screen can be implemented as a touchscreen to receive input signals from the user. The Touch Panel includes one or more touch sensors to sense touches, swipes, and gestures on the Touch Panel. The touch sensors can sense not only the boundaries of touch or swipe actions but also the duration and pressure associated with the touch or swipe operation.

[0124] The aforementioned power supply components provide power to various components within the device in which they reside. These power supply components may include a power management system, one or more power sources, and other components associated with generating, managing, and distributing power to the device in which they reside.

[0125] The aforementioned audio component can be configured to output and / or input audio signals. For example, the audio component includes a microphone (MIC) configured to receive external audio signals when the device containing the audio component is in an operating mode, such as call mode, recording mode, or voice recognition mode. The received audio signals can be further stored in memory or transmitted via a communication component. In some embodiments, the audio component also includes a speaker for outputting audio signals.

[0126] Accordingly, embodiments of this application also provide a computer-readable storage medium storing a computer program, which, when executed by a processor, enables the processor to implement the steps in the above-described method embodiments. The computer-readable storage medium may be volatile, non-volatile, or a combination thereof, and may be removable or non-removable. Examples of computer-readable storage media include, but are not limited to, phase-change random access memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), flash memory or other memory technologies, CD-ROM, digital video disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transfer medium.

[0127] Accordingly, this application also provides a computer program product, which includes a computer program or instructions. When the computer program or instructions are executed by a processor, the processor is able to implement the steps in the above method embodiments. It should be understood that each step or combination of steps in the above method flow can be implemented by the computer program or instructions. In addition, these computer programs or instructions can be applied to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device, so that the processor of the general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing device can be implemented as a means to implement the corresponding functions in the above method embodiments.

[0128] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0129] 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 detecting phone number tags, characterized in that, Applications on the testing server include: Obtain multiple tag detection tasks, and the task information for each tag detection task is used to indicate the tag detection for each phone number among multiple phone numbers; Based on the task information of the plurality of tag detection tasks and the device information of the plurality of terminal devices, a target terminal device corresponding to the target tag detection task is selected from the plurality of terminal devices, wherein the target tag detection task is any one of the plurality of tag detection tasks and the target terminal device is any one of the plurality of terminal devices; Send the task information of the target call record insertion task to the target terminal device to trigger the detection client in the target terminal device to construct the corresponding call record for each of the multiple first phone numbers to be detected by the target tag detection task, and call the call record insertion interface to write the call records corresponding to the multiple first phone numbers into the local call record database of the target terminal device. In response to the target terminal device completing the target call record insertion task, a multi-agent system is invoked to interact with the target terminal device to execute the target tag detection task; During the execution of the target tag detection task, the multi-agent system acquires screenshot information and / or view hierarchy information of the current call log list interface displayed by the target terminal device; based on the screenshot information and / or view hierarchy information of the current call log list interface, it acquires the specific text content corresponding to each of the multiple first phone numbers; based on the specific text content corresponding to each of the multiple first phone numbers, it determines the tag corresponding to each of the multiple first phone numbers sequentially through knowledge base matching and Internet search.

2. The method according to claim 1, characterized in that, The multi-agent system includes a master agent and sub-agents. The system interacts with the target terminal device to perform the target tag detection task, including: The main intelligent agent interacts with the target terminal device to collect specific text content corresponding to each of the multiple first phone numbers; The sub-agent determines the tag for each of the multiple first phone numbers based on the specific text content corresponding to each of the multiple first phone numbers.

3. The method according to claim 2, characterized in that, The main intelligent agent interacts with the target terminal device to collect specific text content corresponding to each of the multiple first phone numbers, including: The main intelligent agent triggers the target terminal device to display the current call log list interface, and takes a screenshot of the current call log list interface and / or obtains the view hierarchy structure information of the current call log list interface; The main AI agent extracts specific text content from at least one second phone number in the current call log list interface based on the screenshot information and / or view hierarchy information of the current call log list interface, so as to obtain specific text content of at least one third phone number. The second phone number is any one of the plurality of first phone numbers, and the third phone number is any one of the at least one second phone number. The main intelligent agent determines whether the collection of specific text content from the multiple first phone numbers has been completed. If the judgment result is negative, the main agent triggers the target terminal device to slide the current call record list interface, and repeats the steps of triggering the target terminal device to take a screenshot of the current call record list interface and / or obtain the view hierarchy information of the current call record list interface, and the subsequent steps, until the specific text content of the multiple first phone numbers is collected.

4. The method according to claim 3, characterized in that, Before the main agent determines whether the collection of specific text content from the multiple first phone numbers is complete, the process also includes: For any fourth phone number for which extraction of specific text content fails, and the fourth phone number is any one of at least two second phone numbers, the main intelligent agent triggers the target terminal device to display the call details interface of the fourth phone number, and takes a screenshot of the call details interface of the fourth phone number and / or obtains the view hierarchy structure information of the call details interface of the fourth phone number; The main AI agent extracts specific text content from the fourth phone number based on screenshots of the call details interface and / or view hierarchy information.

5. The method according to claim 3, characterized in that, Before taking a screenshot of the current call log list interface, the following steps are also included: The main AI agent determines whether a pop-up window exists on the current call log list interface; If a pop-up window exists, the main intelligent trigger will close the pop-up window on the current call log list interface of the target terminal device.

6. The method according to claim 3, characterized in that, The sub-agent determines the respective tagging labels of the plurality of first phone numbers based on the specific text content corresponding to each of the plurality of first phone numbers, including: The sub-agent performs the following operations: For any given first phone number, match it with multiple tags stored in the knowledge base based on the specific text content of the first phone number; If there is a tag that successfully matches a specific text content of the first phone number, then the successfully matched tag will be used as the tag of the first phone number; If no tag successfully matches the specific text content of the first phone number, then the specific text content of the first phone number is entered into the Internet platform for searching, and the search results are obtained; The tag for the first phone number is determined based on the search results; If the tag for the first phone number cannot be determined based on the search results, a prompt will appear indicating that the tag for the first phone number should be determined manually.

7. The method according to any one of claims 1 to 6, characterized in that, Multiple tag detection tasks include: Retrieve multiple tag detection tasks from the task queue, prioritizing them according to their execution priority. The task execution priority of the tag detection task in the task queue is determined according to the source channel of the tag detection task.

8. The method according to any one of claims 1 to 6, characterized in that, Also includes: Detect the task execution status of the target tag detection task; If an abnormality is detected in the execution of the target tag detection task, the process returns to the step of selecting the target terminal device corresponding to the target tag detection task from the multiple terminal devices based on the task information of the multiple tag detection tasks and the device information of the multiple terminal devices, until the maximum number of retries is reached.

9. The method according to claim 8, characterized in that, Also includes: The system receives a failure context reported by the detection client, wherein the detection client performs an integrity check on the write result of the target call record insertion task; if the write result of the target call record insertion task fails the integrity check, the system reports the failure context to the detection server; the failure context includes: device information of the terminal device that failed to write, the number of call records that failed to write, and the phone number that failed to write. The task execution status and task information of the target tag detection task are updated according to the failure context so that the phone numbers that failed to be written can be rewritten into the call record database to continue tag detection.

10. The method according to any one of claims 1 to 6, characterized in that, Based on the task information of the multiple tag detection tasks and the device information of the multiple terminal devices, the target terminal device corresponding to the target tag detection task is selected from the multiple terminal devices, including: Based on the task information of the multiple tag detection tasks, the device information of multiple terminal devices, and the pre-configured mapping relationship between the source channels of the tag detection tasks and the models of the terminal devices, the target terminal device corresponding to the target tag detection task is selected from the multiple terminal devices.

11. The method according to any one of claims 1 to 6, characterized in that, Also includes: In response to the addition of a new terminal device to the terminal device cluster, an environmental integrity test is performed on the new terminal device; If the new terminal device does not meet the environmental integrity requirements, then the new terminal device will be repaired with the goal of making the new terminal device meet the environmental integrity requirements. Bind a purpose tag to the new terminal device.

12. The method according to any one of claims 1 to 6, characterized in that, Also includes: The number of call records already constructed in the local call record database of the target terminal device is periodically detected; If the number of constructed call records exceeds a set threshold, then delete the number of constructed call records in the local call record database of the target terminal device.

13. An electronic device, characterized in that, include: Memory and processor; The memory is used to store computer programs; The processor is coupled to the memory for executing the computer program to perform the steps of the method according to any one of claims 1-12.

14. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it causes the processor to perform the steps of the method according to any one of claims 1-12.

15. A computer program product, characterized in that, Includes a computer program / instruction that, when executed by a processor, causes the processor to perform the steps of the method according to any one of claims 1-12.

Images