CN117390336A - A web page process automation method, device, equipment and storage medium - Google Patents

Abstract

The application provides a webpage process automation method, device, equipment and storage medium, and relates to the technical field of artificial intelligence, wherein the method comprises the following steps: and processing the natural language instruction acquired by the webpage to obtain a leaf node and instruction key value pair which need the intelligent agent to operate, so that the intelligent agent realizes webpage process automation according to the leaf node and instruction key value pair. Because the key value pairs and the leaf nodes are simple elements in the operation of the webpage process, for the condition of replacing the webpage interface or the application environment, the trained intelligent agent can still analyze the natural language instruction through the operation to obtain the instruction key value pairs and the operable leaf nodes so as to realize the automation of the webpage process, reduce the condition of redeploying RPA under different application scenes and reduce the development cost.

Description

A web page process automation method, device, equipment and storage medium

Technical field

This application relates to the field of artificial intelligence technology, and in particular to a web page process automation method, device, equipment and storage medium.

Background technique

With the increasing improvement of enterprise information construction, a large number of Robotic Process Automation (RPA) products have flooded into the market. By simulating human operations between electronic systems, RPA products can not only efficiently complete a large number of repetitive tasks, but also It can realize a large amount of data integration, migration and processing with high efficiency and high quality, greatly improving office efficiency.

Currently, RPA has the disadvantage of poor generalization ability. Each RPA process only corresponds to a fixed keyboard and mouse operation sequence and UI interaction position. Therefore, each web page job has a specific RPA process and cannot be applied to other scenarios, leading to generalization. Poor ability; in addition, if the style of buttons or input boxes on the same web page is arbitrarily changed, the RPA process may not be able to locate the UI that needs to be operated, leading to misoperations.

Therefore, when changing the web interface or using RPA in a different application environment, RPA needs to be redeployed, and redeploying RPA takes a lot of time and requires high costs.

Contents of the invention

In view of this, this application provides a web page process automation method and device, aiming to make RPA applicable to a variety of application scenarios, reduce the redeployment of RPA in different application scenarios, and reduce development costs.

In the first aspect, this application provides a web page process automation method, which method includes:

Obtain the natural language instructions received by the web page plug-in, and the natural language instructions include multiple types of instructions;

Process the natural language instructions to obtain multiple instruction key-value pairs;

Parse the document object model of the web page according to the natural language instructions to obtain operable leaf nodes in the web page;

The plurality of instruction key-value pairs and the operable leaf nodes are input into a pre-trained agent to automatically execute the natural language instructions.

Optionally, training the agent includes:

The recurrent neural network is used as the core network of the agent, and the agent is trained with instruction key-value pairs as input, operable leaf nodes as state space, clicking text boxes and inputting natural language text as action space.

Optionally, the natural language instructions received by the web page acquisition plug-in include:

Obtain the natural language instructions received within the preset time period, and store the natural language instructions received within the preset time period in the order of reception time.

Optionally, when the natural language instruction is a download instruction, the download instruction includes the storage address after the download is completed, and the method further includes:

After completing the download instruction, the download content included in the download instruction is stored according to the storage address.

Optionally, the operable leaf nodes include:

One or more of text boxes, buttons, drop-down boxes, and check boxes in web pages.

In the second aspect, this application provides a web page process automation device, which includes:

An acquisition unit, used to acquire natural language instructions received by the web page plug-in, where the natural language instructions include multiple types of instructions;

A first processing unit, configured to process the natural language instructions to obtain multiple instruction key-value pairs;

A second processing unit configured to parse the document object model of the web page according to the natural language instruction to obtain operable leaf nodes in the web page;

An input unit is used to input the plurality of instruction key-value pairs and the operable leaf nodes into a pre-trained agent to automatically execute the natural language instruction.

Optionally, the device also includes a training unit for using the recurrent neural network as the core network of the agent, using instruction key-value pairs as input, operable leaf nodes as the state space, clicking on the text box and inputting natural language. The text trains the agent in action space.

Optionally, the acquisition unit is specifically used for:

Obtain the natural language instructions received within the preset time period, and store the natural language instructions received within the preset time period in the order of reception time.

Optionally, the device further includes a storage unit. When the natural language instruction is a download instruction, the download instruction includes a storage address after the download is completed. After the download instruction is completed, the storage The unit is configured to store the download content included in the download instruction according to the storage address.

Optionally, the operable leaf nodes include:

One or more of text boxes, buttons, drop-down boxes, and check boxes in web pages.

In a third aspect, the application provides a device. The device includes a memory and a processor. The memory is used to store instructions or codes. The processor is used to execute the instructions or codes, so that the device executes the foregoing. The web page process automation method described in any one of the first aspects.

In a fourth aspect, the present application provides a computer-readable storage medium. Code is stored in the computer-readable storage medium. When the code is run, the device running the code implements any of the foregoing first aspects. The described web page process automation method.

This application provides a web page process automation method. When executing the method, first obtain the natural language instructions received by the web page plug-in. The natural language instructions include multiple types of instructions, and then process the natural language instructions to obtain multiple instruction key-value pairs. According to the natural language instructions, The language instruction parses the document object model of the web page to obtain the operable leaf nodes in the web page, and inputs the multiple instruction key-value pairs and the operable leaf nodes into the pre-trained agent to automatically execute the Describe natural language instructions. In this way, by processing the natural language instructions obtained from the web page, the leaf nodes and instruction key-value pairs that need to be operated by the agent are obtained, so that the agent can realize the automation of the web page process based on the leaf nodes and instruction key-value pairs. Since key-value pairs and leaf nodes are simple elements in web page process operations, when the web interface is changed or the application environment is changed, the trained agent can still parse the natural language instructions through operations to obtain the instruction key-value pairs and available Operate leaf nodes to automate web page processes, reduce redeployment of RPA in different application scenarios, and reduce development costs.

Description of the drawings

In order to explain this embodiment or the technical solution in the prior art more clearly, the drawings needed to be used in the description of the embodiment or the prior art will be briefly introduced below. Obviously, the drawings in the following description are only For some embodiments of the present application, those of ordinary skill in the art can also obtain other drawings based on these drawings without exerting creative efforts.

Figure 1 is a flow chart of a web page process automation method provided by an embodiment of the present application;

Figure 2 is a schematic structural diagram of a web page process automation device provided by an embodiment of the present application;

FIG. 3 is a schematic structural diagram of a computer device provided by an embodiment of the present application.

Detailed ways

As mentioned in the background technology of this application, in the web process automation method in the prior art, RPA needs to be trained on the entire operation process. When the web page or application scenario changes, the training needs to be re-trained, which takes a lot of time and makes the cost relatively high. high.

In order to solve the above technical problems, embodiments of the present application provide a web page process automation method, which method includes:

First obtain the natural language instructions received by the web page plug-in, which include multiple types of instructions, then process the natural language instructions to obtain multiple instruction key-value pairs, and parse the web page according to the natural language instructions. The document object model obtains the operable leaf nodes in the web page, and inputs the multiple instruction key-value pairs and the operable leaf nodes into the pre-trained agent to automatically execute the natural language instructions. In this way, by processing the natural language instructions obtained from the web page, the leaf nodes and instruction key-value pairs that need to be operated by the agent are obtained, so that the agent can realize the automation of the web page process based on the leaf nodes and instruction key-value pairs. Since key-value pairs and leaf nodes are simple elements in web page process operations, when the web interface is changed or the application environment is changed, the trained agent can still parse the natural language instructions through operations to obtain the instruction key-value pairs and available Operate leaf nodes to automate web page processes, reduce redeployment of RPA in different application scenarios, and reduce development costs.

It should be noted that the user information (including but not limited to user equipment information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all It is information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data need to comply with the relevant laws, regulations and standards of relevant countries and regions.

In order to make the purpose, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments These are only some of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the scope of protection of this application.

Figure 1 is a flow chart of a web page process automation method provided by an embodiment of the present application. As shown in Figure 1, the web page process automation method provided by the embodiment of the present application may include:

S101. Obtain the natural language instructions received by the web page plug-in. The natural language instructions include multiple types of instructions.

Web plug-ins, also known as browser plug-ins, are used to extend browser functions, enrich browsing experience, and meet different functional requirements. After the web page plug-in in the embodiment of this application is started, a text box will be displayed on the front-end web page, and a natural language instruction can be input through the text box. For example: the natural language instruction input by the user is: "Help me download the last three Monthly bills."

Natural language refers to languages used daily such as Chinese, English, and French. It is a language that evolved from the development of human society, rather than a man-made language. It is an important tool for human beings to learn and live. In summary, natural language refers to artificial language that is conventional in human society and is different from language such as programming.

The natural language instructions include multiple types, specifically, they may include download instructions, fill-in instructions, and search instructions. The downloading instruction specifically refers to downloading a certain type of file; the filling instruction refers to filling in a certain form; and the searching instruction refers to searching for a certain content, such as data or files.

S102. Process the natural language instructions to obtain multiple instruction key-value pairs.

After obtaining the natural language instructions through the front-end web page plug-in, the natural language instructions are transmitted to the back-end for processing. The processing process specifically includes identifying the natural language instructions through a natural language model.

Natural language processing model refers to a model that can perform natural language processing. Natural Language Processing (NLP) is a method of using computers to process, understand and use human language. It is a branch of artificial intelligence. The natural language processing model can process the form, pronunciation, meaning and other information of natural language, that is, the input, output, recognition, analysis, understanding, generation, etc. of characters, words, sentences, and chapters, to achieve human-computer interaction. exchange of information.

The instruction key-value pair refers to the key-value pair related to the instruction obtained after processing the natural language instruction. The key-value pair (key: value) is a simple correspondence relationship, and the key corresponds to the corresponding value, for example : {Time: three months}, where "time" is the key and "three months" is the corresponding value, which is combined into a key-value pair. Specifically, multiple key-value pairs can be obtained by processing natural language instructions. For example, the natural language instruction input by the user is: "Help me download the bills of the last three months." After the natural language processing model, "{time" can be generated. : three months, document type: running bill, command: download}" such key-value pairs.

S103. Parse the document object model of the web page according to the natural language instruction to obtain operable leaf nodes in the web page.

Document Object Model (DOM for short) is a programming interface for network documents. DOM represents documents as nodes and objects to enable interaction with the page through programming languages. Generally speaking, all the content presented on the page is regarded as a node in the DOM document. For example, element labels are element nodes, annotated content is annotation nodes, text content is text nodes, document is a document node, etc. In this application, operable controls in the DOM are used as leaf nodes, and the operable leaf nodes include one or more of text boxes, buttons, drop-down boxes, and check boxes in web pages.

The operable leaf nodes are different in different application scenarios. For example, when the natural language instruction is a download type instruction, the operable leaf nodes include text boxes, buttons, and drop-down boxes in the web page; when the natural language instruction is a fill-in type instruction When commanding, the operable leaf nodes include text boxes, buttons, drop-down boxes and check boxes in the web page.

S104. Input the plurality of instruction key-value pairs and the operable leaf nodes into a pre-trained agent to automatically execute the natural language instruction.

After obtaining the instruction key-value pair and operable leaf nodes, the pre-trained agent automatically performs the corresponding actions to complete the web page job.

The agent is an agent based on reinforcement learning. After training, the agent has learned various atomic tasks such as clicking buttons, clicking drop-down boxes, selecting check boxes, entering text, etc., and any web page operation process, such as Downloading files, filling out forms, etc. can all be broken down into a combination of these atomic tasks. As long as the agent is trained to handle atomic tasks, it can be generalized to other complex workflows through reinforcement learning. In addition, because the reinforcement learning agent operates directly on the DOM elements of the web page, it is not sensitive to attribute descriptions such as the position and shape of the UI, and thus has extremely high style generalization capabilities.

Training the agent includes: using Rerrent Neural Network (RNN) as the core network of the agent, using instruction key-value pairs as input, operable leaf nodes as the state space, clicking on the text box and input Natural language text is used to train the agent in the action space. In order to enable the reinforcement learning agent to identify the UI component categories corresponding to various DOM elements (i.e. buttons, text boxes, drop-down boxes, etc.) and learn to align command key-value pairs with operable leaf nodes, it is necessary to collect information about different web pages during the training phase. Data sets described by the DOM of UI components such as buttons, drop-down boxes, and check boxes, and these data sets are annotated. For example, a search button, the DOM can be represented as "<buttontype="search">Search</button>" or "<a href="#"class="button">Search</a>" , they are all marked as buttons. The richer the data set, the stronger the generalization of the trained agent. Then the simulation environment is generated reversely based on these DOM data sets. The agent is trained in the simulation environment until it can determine the UI component category based only on the DOM description.

Since the agent that has undergone reinforcement learning takes action by judging the state of the current environment, in the web page operation process, the state is the DOM information of the web page, and the action is click or input. , drag and drop and other operations. Therefore, when an abnormal situation occurs on the web page, the agent determines the output behavior by judging the changes in the DOM elements of the web page. If the web page exception does not affect the key operable DOM leaf nodes, the job process will not be interrupted, thus achieving the effect of being able to cope with abnormal network conditions.

In addition, the agent trained based on reinforcement learning described in the embodiments of this application can quickly adapt to new web pages and complete corresponding instructions. Since the reinforcement learning agent uses web page DOM elements as input during training, it will not be affected by factors such as web page layout, structure, color, style, etc., but is only related to operable leaf nodes, so it can quickly adapt to different visual styles. The web interface avoids the situation where the intelligent agent cannot recognize the button or input box style or position due to changes in the web page.

In an implementation manner of the embodiment of the present application, obtaining the natural language instructions received by the web page plug-in includes: obtaining the natural language instructions received within a preset time period, and processing the preset time period in the order of the time of receipt. The natural language instructions received within are stored.

When there are multiple tasks that need to be realized through process automation, multiple task instructions can be obtained at one time, such as downloading user a's bills for one year; downloading user b's bills for three months, and then follow the chronological order of obtaining instructions. Store the obtained instructions and complete the instructions in sequence. In order to ensure that the instructions can be completed in the order they appear, instructions can be stored in chronological order.

In an implementation manner of the embodiment of the present application, when the natural language instruction is a download instruction, the download instruction includes the storage address after the download is completed, and the method further includes:

After completing the download instruction, the download content included in the download instruction is stored according to the storage address.

The downloading instruction is, for example: "Download the bills within three months and store them on the desktop." Then, through the web page process automation method, the instruction can be completed, and when the storage address changes, it can also be stored to the target address. Since the agent in this application has learned to operate the leaf nodes in the web page, that is, small components, such as clicking buttons, clicking drop-down boxes, selecting check boxes, entering text, etc., whether it is changing the application scenario or changing Web page layout and agents can be realized through click components. Compared with the need to redeploy RPA once in different scenarios, the method described in the embodiment of this application can reduce development costs.

The above examples of this application provide a web page process automation method. The method includes: first obtaining the natural language instructions received by the web page plug-in. The natural language instructions include multiple types of instructions, and then processing the natural language instructions to obtain multiple types of instructions. instruction key-value pairs, parse the document object model of the web page according to the natural language instruction, obtain operable leaf nodes in the web page, and input the plurality of instruction key-value pairs and the operable leaf nodes in advance The intelligent agent is trained to automatically execute the natural language instructions. In this way, by processing the natural language instructions obtained from the web page, the leaf nodes and instruction key-value pairs that need to be operated by the agent are obtained, so that the agent can realize the automation of the web page process based on the leaf nodes and instruction key-value pairs. Since key-value pairs and leaf nodes are simple elements in web page process operations, when the web interface is changed or the application environment is changed, the trained agent can still parse the natural language instructions through operations to obtain the instruction key-value pairs and available Operate leaf nodes to automate web page processes, reduce redeployment of RPA in different application scenarios, and reduce development costs.

The above are some specific implementations of a web page process automation method provided by embodiments of the present application. Based on this, the present application also provides corresponding devices. The following will introduce the device provided by the embodiment of the present application from the perspective of functional modularization.

FIG. 2 is a schematic structural diagram of a web page process automation device provided by an embodiment of the present application. As shown in FIG. 2 , the web page process automation device 200 provided by the embodiment of the present application includes:

The acquisition unit 210 is used to acquire the natural language instructions received by the web page plug-in, where the natural language instructions include multiple types of instructions;

The first processing unit 220 is used to process the natural language instructions to obtain multiple instruction key-value pairs;

The second processing unit 230 is configured to parse the document object model of the web page according to the natural language instruction to obtain operable leaf nodes in the web page;

The input unit 240 is used to input the plurality of instruction key-value pairs and the operable leaf nodes into a pre-trained agent to automatically execute the natural language instructions.

In an implementation manner of the embodiment of the present application, the device further includes a training unit for using a recurrent neural network as the core network of the agent, using instruction key-value pairs as input and operable leaf nodes as the state space. , click the text box and input natural language text to train the intelligent agent in the action space.

In an implementation manner of the embodiment of the present application, the acquisition unit is specifically used to:

Obtain the natural language instructions received within the preset time period, and store the natural language instructions received within the preset time period in the order of reception time.

In an implementation manner of the embodiment of the present application, the device further includes a storage unit. When the natural language instruction is a download instruction, the download instruction includes the storage address after the download is completed. After downloading the instruction, the storage unit is used to store the download content included in the download instruction according to the storage address.

In an implementation manner of the embodiment of this application, the operable leaf nodes include:

One or more of text boxes, buttons, drop-down boxes, and check boxes in web pages.

The embodiments of this application also provide corresponding equipment and computer storage media for implementing the solution provided by the embodiments of this application.

As shown in Figure 3, computer device 01 is embodied in the form of a general computing device. The components of the computer device 01 may include, but are not limited to: one or more processors or processing units 03, a system memory 08, and a bus 04 connecting different system components (including the system memory 08 and the processing unit 03).

Bus 04 represents one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, a graphics acceleration port, a processor, or a local bus using any of a variety of bus structures. For example, these architectures include, but are not limited to, the Industry Standard Architecture (ISA) bus, the Micro Channel Architecture (MAC) bus, the Enhanced ISA bus, the Video Electronics Standards Association (VESA) local bus, and the Peripheral Component Interconnect ( PCI) bus.

Computer device 01 typically includes a variety of computer system readable media. These media can be any available media that can be accessed by computer device 01, including volatile and non-volatile media, removable and non-removable media.

System memory 08 may include computer system readable media in the form of volatile memory, such as random access memory (RAM) 09 and/or cache memory 10 . Computer device 01 may further include other removable/non-removable, volatile/non-volatile computer system storage media. By way of example only, storage system 11 may be used to read and write to non-removable, non-volatile magnetic media (not shown in Figure 3, commonly referred to as a "hard drive"). Although not shown in FIG. 3, a disk drive may be provided for reading and writing to removable non-volatile disks (e.g., "floppy disks"), and for removable non-volatile optical disks (e.g., CD-ROM, DVD-ROM or other optical media) that can read and write optical disc drives. In these cases, each drive can be connected to bus 04 via one or more data media interfaces. The memory 08 may include at least one program product having a set of (eg, at least one) program modules configured to perform the functions of various embodiments of the present application.

A program/utility 12 having a set of (at least one) program modules 13 including, but not limited to, an operating system, one or more application programs, other program modules, and program data, may be stored, for example, in memory 08 , each of these examples or some combination may include the implementation of a network environment. The program module 13 generally performs the functions and/or methods in the embodiments described in this application.

Computer device 01 may also communicate with one or more external devices 02 (e.g., keyboard, pointing device, display 07, etc.), with one or more devices that enable a user to interact with computer device 01, and/or with Any device (such as a network card, modem, etc.) that enables the computer device 01 to communicate with one or more other computing devices. This communication can occur through the input/output (I/O) interface 06. Furthermore, the computer device 01 may also communicate with one or more networks (eg, a local area network (LAN), a wide area network (WAN), and/or a public network, such as the Internet) through the network adapter 05. As shown in Figure 3, network adapter 05 communicates with other modules of computer device 01 via bus 04. It should be understood that, although not shown in Figure 3, other hardware and/or software modules may be used in conjunction with computer device 01, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tapes drives and data backup storage systems, etc.

The processor unit 03 executes various functional applications and data processing by running programs stored in the system memory 08 , for example, implementing the methods provided by the embodiments of the present application.

From the description of the above embodiments, those skilled in the art can clearly understand that all or part of the steps in the methods of the above embodiments can be implemented by means of software plus a general hardware platform. Based on this understanding, the technical solution of the present application can be embodied in the form of a software product. The computer software product can be stored in a storage medium, such as read-only memory (English: read-only memory, ROM)/RAM, disk, Optical disc, etc., including a number of instructions to cause a computer device (which can be a personal computer, a server, or a network communication device such as a router) to execute the methods described in various embodiments or certain parts of the embodiments of this application.

It should be noted that in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply that these entities or operations are mutually exclusive. any such actual relationship or sequence exists between them. Furthermore, the terms "comprises," "comprises," or any other variations thereof are intended to cover a non-exclusive inclusion such that a process, method, article, or apparatus that includes a list of elements includes not only those elements, but also those not expressly listed other elements, or elements inherent to the process, method, article or equipment. Without further limitation, an element defined by the statement "comprises a..." does not exclude the presence of additional identical elements in a process, method, article, or apparatus that includes the stated element.

It should also be noted that each embodiment in this specification is described in a progressive manner. The same and similar parts between various embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments. place. In particular, the device and system embodiments are described simply because they are basically similar to the method embodiments. For relevant details, please refer to the partial description of the method embodiments. The device and system embodiments described above are only illustrative. The units described as separate components may or may not be physically separated. The components indicated as units may or may not be physical units, that is, they may be located in One location, or it can be distributed across multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. Persons of ordinary skill in the art can understand and implement the method without any creative effort.

The above is only a specific implementation mode of the present application, but the protection scope of the present application is not limited thereto. Any person familiar with the technical field can easily think of changes or modifications within the technical scope disclosed in the present application. Replacements shall be covered by the protection scope of this application. Therefore, the protection scope of this application should be subject to the protection scope of the claims.

Claims (10)

1. A web page process automation method, characterized in that the method includes: Obtain the natural language instructions received by the web page plug-in, and the natural language instructions include multiple types of instructions; Process the natural language instructions to obtain multiple instruction key-value pairs; Parse the document object model of the web page according to the natural language instructions to obtain operable leaf nodes in the web page; The plurality of instruction key-value pairs and the operable leaf nodes are input into a pre-trained agent to automatically execute the natural language instructions. 2. The method according to claim 1, characterized in that training the intelligent agent includes: The recurrent neural network is used as the core network of the agent, and the agent is trained with instruction key-value pairs as input, operable leaf nodes as state space, clicking text boxes and inputting natural language text as action space. 3. The method according to claim 1, characterized in that said obtaining the natural language instructions received by the web page plug-in includes: Obtain the natural language instructions received within the preset time period, and store the natural language instructions received within the preset time period in the order of reception time. 4. The method according to claim 1, characterized in that when the natural language instruction is a download instruction, the download instruction includes a storage address after the download is completed, and the method further includes: After completing the download instruction, the download content included in the download instruction is stored according to the storage address. 5. The method according to any one of claims 1 or 2, characterized in that the operable leaf nodes include: One or more of text boxes, buttons, drop-down boxes, and check boxes in web pages. 6. A web page process automation device, characterized in that the device includes: An acquisition unit, used to acquire natural language instructions received by the web page plug-in, where the natural language instructions include multiple types of instructions; A first processing unit, configured to process the natural language instructions to obtain multiple instruction key-value pairs; A second processing unit configured to parse the document object model of the web page according to the natural language instruction to obtain operable leaf nodes in the web page; An input unit is used to input the plurality of instruction key-value pairs and the operable leaf nodes into a pre-trained agent to automatically execute the natural language instruction. 7. The device according to claim 6, characterized in that the device further includes a training unit for using a recurrent neural network as the core network of the agent, using instruction key-value pairs as input and operable leaf nodes. The intelligent agent is trained as a state space, clicking a text box and inputting natural language text as an action space. 8. The device according to claim 6, characterized in that the acquisition unit is specifically used to: Obtain the natural language instructions received within the preset time period, and store the natural language instructions received within the preset time period in the order of reception time. 9. A computing device, characterized in that the computing device includes: a memory and a processor; The memory is used to store computer programs; The processor is configured to implement the web page process automation method according to any one of claims 1 to 5 when executing the computer program. 10. A computer-readable storage medium, characterized in that a computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, the web page according to any one of claims 1 to 5 is implemented. Process automation methods.