User data processing method, device, equipment and storage medium combining RPA and AI

Abstract

The present application provides a user data processing method, device, equipment and storage medium combining RPA and AI. Applied to Natural Language Processing (NLP) of artificial intelligence AI, the method includes: obtaining the user's access information; determining the user's target business intention based on the access information, wherein the target business intention corresponds to multiple word slots, and each word slot corresponds to a group of query statements; using the query statements of the word slots corresponding to the target business intention to interact with the user in a conversation to obtain the conversation statements input by the user; based on the conversation statements, generating and outputting the user data corresponding to the target business intention to understand the intention. The present application can automatically interact with the user and obtain user data through the word slots and query statements corresponding to the preset business intentions, thereby improving the efficiency of obtaining user data.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a user data processing method, device, equipment and storage medium combining RPA (Robotic Process Automation) and AI (Artificial Intelligence). Background Art

[0002] Robotic Process Automation (RPA) simulates human operations on computers through specific "robot software" and automatically executes process tasks according to rules. Artificial Intelligence (AI) is a technical science that studies and develops theories, methods, technologies and application systems for simulating, extending and expanding human intelligence.

[0003] With the development of Internet technology, more and more companies are promoting their products through the Internet. Users can learn about the business products they are interested in by browsing the product promotion pages provided by the company. The company interacts with the users browsing the page through the dialog box on the page, and records the user's business intentions, contact information and other data, so as to carry out appropriate product promotion to the user in the future.

[0004] In the related technology, in the application scenario of natural language processing (NLP) in the field of computer technology, manual customer service interacts with users browsing the page, and the manual customer service fills in the user's data in the user data table.

[0005] However, since a human customer service representative can only interact with a few users at the same time, and the speed of manually filling in user data is slow, the efficiency of obtaining user data is low during the process of intent understanding. Summary of the invention

[0006] The embodiments of the present application provide a user data processing method, apparatus, device, and storage medium that combine RPA and AI to solve the problem of low efficiency in manually obtaining user data.

[0007] In a first aspect, an embodiment of the present application provides a user data processing method combining RPA and AI, which is applied to natural language processing (Natural Language Processing, NLP) of artificial intelligence AI, including:

[0008] Obtain user access information;

[0009] Determine the user's target business intention according to the access information, wherein the target business intention corresponds to a plurality of word slots, and each word slot corresponds to a set of query statements;

[0010] Performing conversational interaction with the user using the query sentence of the word slot corresponding to the target business intent to obtain the conversation sentence input by the user;

[0011] According to the conversation sentence, user data corresponding to the target business intent is generated and output to understand the intent.

[0012] In a possible implementation manner, the number of unfilled word slots is at least one;

[0013] Using the query sentence corresponding to the unfilled word slot to perform conversation interaction with the user includes:

[0014] For each unfilled word slot, push the query sentence corresponding to the unfilled word slot to the user, and receive the conversation sentence input by the user;

[0015] After receiving the conversation statement input by the user, the method further includes:

[0016] Determine whether there is a new business intent according to the conversation sentence, and when it is determined that there is the new business intent, perform conversation interaction with the user using a query sentence of a word slot corresponding to the new business intent to fill the word slot corresponding to the new business intent;

[0017] After the word slot corresponding to the new business intent is filled, jump to execute the following steps: use the query sentence of the word slot corresponding to the target business intent to interact with the user in a conversation.

[0018] In a possible implementation, before using the query statement of the word slot corresponding to the new business intent to perform conversational interaction with the user, the method further includes:

[0019] For the word slots in the filled word slots that are preset and shared with the new business intent, data is retained.

[0020] In a possible implementation, generating and outputting user data corresponding to the target service intent according to the conversation statement includes:

[0021] Extracting words that match the word slots corresponding to the target business intent from the conversation sentences, and filling the words into the corresponding word slots;

[0022] According to the word slot corresponding to the target business intent and the vocabulary filled therein, user data corresponding to the target business intent is generated and output.

[0023] In a possible implementation, the word slot includes a contact information word slot;

[0024] Generating and outputting user data corresponding to the target service intent according to the word slot corresponding to the target service intent and the vocabulary filled therein, including:

[0025] Determining whether the contact information word slot is filled;

[0026] When the contact information word slot is filled, user data corresponding to the target business intent is generated and output according to the word slot corresponding to the target business intent and the words filled therein.

[0027] In a possible implementation, when the contact information word slot is filled, the method further includes:

[0028] It is determined that there is a jump instruction, and before the jump, data is retained for a preset word slot shared with the target business intent.

[0029] In a possible implementation, the conversational statements include question statements;

[0030] The method further comprises:

[0031] The words matching the word slots corresponding to the target business intent are extracted from the question sentence and filled into the corresponding word slots.

[0032] In a possible implementation, the method further includes:

[0033] Reply to the question statement, and after replying to the question statement, jump to the following steps: perform conversational interaction with the user using the query statement of the word slot corresponding to the target business intent;

[0034] Wherein, responding to the question statement includes:

[0035] After receiving the question statement, searching for a question matching the question statement from a preset question-and-answer database, and pushing the answer text corresponding to the question matching the question statement to the user, wherein the question-and-answer database includes multiple sets of corresponding questions and answer texts;

[0036] Alternatively, based on the table lookup logic corresponding to the question statement, data matching the question statement is searched from a preset knowledge data table, and an answer text is generated according to the data matching the question statement, and the answer text is pushed to the user.

[0037] In a possible implementation manner, the access information includes:

[0038] Keywords that users search for.

[0039] In a possible implementation manner, the access information further includes:

[0040] User attribute information, wherein the attribute information includes at least one of the following: page information of the page currently visited by the user, question statements sent by the user on the currently visited page, user tags, and user network address information;

[0041] A welcome message is determined according to the attribute information, and the welcome message is pushed to the user.

[0042] In a second aspect, an embodiment of the present application provides a user data processing device combining RPA and AI, which is applied to natural language processing (NLP) of artificial intelligence (AI), including:

[0043] The acquisition module is used to obtain the user's access information;

[0044] A processing module, configured to determine a target business intention of the user according to the access information, wherein the target business intention corresponds to a plurality of word slots, and each word slot corresponds to a set of query statements;

[0045] An interaction module, used to perform conversation interaction with the user by using the query sentence of the word slot corresponding to the target business intent, and obtain the conversation sentence input by the user;

[0046] The output module is used to generate and output user data corresponding to the target business intent based on the conversation sentence to understand the intent.

[0047] In a possible implementation manner, the interaction module is specifically used to:

[0048] Determine unfilled word slots in the word slots corresponding to the target business intent;

[0049] The query sentence corresponding to the unfilled word slot is used to perform conversation interaction with the user.

[0050] In a possible implementation manner, the number of unfilled word slots is at least one;

[0051] The interaction module is specifically used for:

[0052] For each unfilled word slot, push the query sentence corresponding to the unfilled word slot to the user, and receive the conversation sentence input by the user;

[0053] The interaction module is further used for:

[0054] Determine whether there is a new business intent according to the conversation sentence, and when it is determined that there is the new business intent, perform conversation interaction with the user using a query sentence of a word slot corresponding to the new business intent to fill the word slot corresponding to the new business intent;

[0055] After the word slot corresponding to the new business intent is filled, jump to execute the following steps: use the query sentence of the word slot corresponding to the target business intent to interact with the user in a conversation.

[0056] In a possible implementation manner, the interaction module is further used to:

[0057] Before using the query statement of the word slot corresponding to the new business intent to interact with the user in a conversation, data is retained for the word slots preset in the filled word slots and shared with the new business intent.

[0058] In a possible implementation manner, the output module is specifically used to:

[0059] Extracting words that match the word slots corresponding to the target business intent from the conversation sentences, and filling the words into the corresponding word slots;

[0060] According to the word slot corresponding to the target business intent and the vocabulary filled therein, user data corresponding to the target business intent is generated and output.

[0061] In a possible implementation, the word slot includes a contact information word slot;

[0062] The output module is specifically used for:

[0063] Determining whether the contact information word slot is filled;

[0064] When the contact information word slot is filled, user data corresponding to the target business intent is generated and output according to the word slot corresponding to the target business intent and the words filled therein.

[0065] In a possible implementation manner, the output module is further used to:

[0066] When the contact information word slot is filled, it is determined that there is a jump instruction, and before the jump, data is retained for the preset word slot shared with the target business intent.

[0067] In a possible implementation, the conversational statements include question statements;

[0068] The output module is further used for:

[0069] The words matching the word slots corresponding to the target business intent are extracted from the question sentence and filled into the corresponding word slots.

[0070] In a possible implementation manner, the interaction module is further used to:

[0071] After receiving the question statement, reply to the question statement, and after replying to the question statement, jump to execute the following steps: use the inquiry statement of the word slot corresponding to the target business intent to interact with the user in a conversation;

[0072] Wherein, responding to the question statement includes:

[0073] Searching for a question matching the question statement from a preset question-and-answer database, and pushing the answer text corresponding to the question matching the question statement to the user, wherein the question-and-answer database includes a plurality of sets of corresponding questions and answer texts;

[0074] Alternatively, based on the table lookup logic corresponding to the question statement, data matching the question statement is searched from a preset knowledge data table, and an answer text is generated according to the data matching the question statement, and the answer text is pushed to the user.

[0075] In a possible implementation manner, the access information includes:

[0076] Keywords that users search for.

[0077] In a possible implementation manner, the access information further includes:

[0078] User attribute information, wherein the attribute information includes at least one of the following: page information of the page currently visited by the user, question statements sent by the user on the currently visited page, user tags, and user network address information;

[0079] The interaction module is further used for:

[0080] A welcome message is determined according to the attribute information, and the welcome message is pushed to the user.

[0081] In a third aspect, an embodiment of the present application provides a user data processing device, including: at least one processor and a memory;

[0082] The memory stores computer-executable instructions;

[0083] The at least one processor executes the computer-executable instructions stored in the memory, so that the at least one processor executes the user data processing method combining RPA and AI as described in the first aspect and various possible implementations of the first aspect above.

[0084] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, in which computer-executable instructions are stored. When a processor executes the computer-executable instructions, the user data processing method combining RPA and AI as described in the first aspect and various possible implementations of the first aspect is implemented.

[0085] The user data processing method, apparatus, device and storage medium combining RPA and AI provided in the embodiments of the present application obtain the user's access information, determine the user's target business intent based on the access information, the target business intent corresponds to multiple word slots, each word slot corresponds to a group of query statements, and then use the query statements of the word slots corresponding to the target business intent to interact with the user in a conversation, obtain the conversation statements input by the user, generate and output user data corresponding to the target business intent based on the conversation statements, and can automatically interact with the user through the word slots and query statements corresponding to the preset business intent to obtain user data, thereby improving the efficiency of obtaining user data. BRIEF DESCRIPTION OF THE DRAWINGS

[0086] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0087] Figure 1 A schematic diagram of the architecture of a user data processing system combining RPA and AI provided in one embodiment of the present application;

[0088] Figure 2 A flowchart of a user data processing method combining RPA and AI provided in one embodiment of the present application;

[0089] Figure 3 A schematic diagram of a session window provided in an embodiment of the present application;

[0090] Figure 4 A flowchart of a user data processing method combining RPA and AI provided in another embodiment of the present application;

[0091] Figure 5 A flowchart of a user data processing method combining RPA and AI provided in another embodiment of the present application;

[0092] Figure 6 A schematic diagram of a program unit for implementing a user data processing method combining RPA and AI provided in an embodiment of the present application;

[0093] Figure 7 A schematic diagram of the structure of a user data processing device combining RPA and AI provided in one embodiment of the present application;

[0094] Figure 8 A schematic diagram of the hardware structure of a user data processing device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0095] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution 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 are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0096] Figure 1 This is a schematic diagram of the architecture of a user data processing system combining RPA and AI provided in one embodiment of the present application. Figure 1 As shown, the user data processing system combining RPA and AI may include a terminal 11 and a server 12. The terminal may be a mobile phone, a tablet, a laptop computer, a desktop computer, etc., which is not limited here.

[0097] In an application scenario, the server 12 can be used as the execution subject of the user data processing method combining RPA and AI provided in this embodiment. The user can access the company's web page through the terminal 11, and the terminal 11 sends the user's access information to the server 12. The server 12 determines the user's target business intention based on the user's access information, and then sends the query statement of the word slot corresponding to the target business intention to the terminal 11. The terminal 11 displays the query statement to the user, and sends the conversation statement entered by the user to the server 12 to realize the conversation interaction with the user. The server 12 generates user data based on the conversation statement in the process of conversation interaction with the user, and sends the user data to the terminal of the enterprise staff so that the enterprise staff can organize the user data, or promote products based on the user data.

[0098] In another application scenario, the terminal 11 can be used as the execution subject of the user data processing method combining RPA and AI provided in this embodiment. The user can access the company's webpage through the terminal 11. The terminal 11 can determine the user's target business intention based on the user's access information, and then display the query sentence of the word slot corresponding to the target business intention to the user, and receive the conversation sentence input by the user to realize the conversation interaction with the user. The terminal 11 generates user data based on the conversation sentence in the process of conversation interaction with the user. The terminal 11 can send the user data to the terminal of the enterprise staff through the server 12, so that the enterprise staff can organize the user data, or promote products based on the user data.

[0099] Compared with the prior art in which manual customer service interacts with users who visit the page and records user data, which is inefficient, this embodiment can automatically interact with users and obtain user data through word slots and inquiry statements corresponding to preset business intents, thereby improving the efficiency of obtaining user data.

[0100] Figure 2 A flowchart of a user data processing method combining RPA and AI provided in one embodiment of the present application. The execution subject of the user data processing method combining RPA and AI provided in this embodiment is a user data processing device, which can be the above-mentioned terminal or server, and is not limited here.

[0101] It should be noted that the "processing" in this application refers to the data processing process that combines robotic process automation (RPA) and artificial intelligence (AI). That is to say, the data processing process is a fully automated data processing process, and the data processing process is also combined with artificial intelligence (AI) to achieve fully automated analysis and processing of user data.

[0102] The present application can be specifically applied to Natural Language Processing (NLP) of Artificial Intelligence AI, Natural Language Processing (NLP), that is, the field of computer science, artificial intelligence, and linguistics that focuses on the interaction between computers and human (natural) language.

[0103] For example, the present application realizes the full-process automated data processing process to obtain the user's access information in a fully automated manner, and executes the step of determining the user's target business intent based on the access information, wherein the target business intent corresponds to multiple word slots, and each word slot corresponds to a set of query statements. Then, combined with the natural language processing NLP of artificial intelligence AI, the query statements of the word slots corresponding to the target business intent are used to interact with the user in a conversation to obtain the conversation statements input by the user; based on the conversation statements, the user data corresponding to the target business intent is generated and output to understand the intent.

[0104] like Figure 2 As shown, the method includes:

[0105] S201. Obtain user access information.

[0106] S202: Determine the user's target business intent according to the access information, wherein the target business intent corresponds to a plurality of word slots, and each word slot corresponds to a set of query statements.

[0107] It should be noted that the “plurality” mentioned in the present application refers to two or more than two, that is, a plurality of word slots refers to two or more than two word slots.

[0108] Among them, business intent can be used to characterize the user's business intention, that is, business intent can be used to describe the business services that the user needs to obtain, and in the embodiment of the present application, the business intent corresponding to the obtained access information can be called the target business intent.

[0109] In this embodiment, when a user accesses a page, the user data processing device may obtain the user's access information. Optionally, the access information may include: keywords searched by the user. The access information may also include: page information of the page currently accessed by the user, question statements sent by the user on the currently accessed page, etc., which are not limited here. The user data processing device may determine the user's target business intention based on one or more access information.

[0110] For example, the keyword may be a keyword entered by the user on the company's search page. The user data processing device may determine the user's target business intent based on the keyword searched by the user. Taking the academic consulting scenario as an example, if the user searches for "study tour" on the company's search page, the page will jump from the company's search page to the company's business display page. The businesses displayed on the business display page include "study tour" and "study abroad". The user data processing device may determine that the user's target business intent is "study tour" instead of "study abroad" based on the keyword searched by the user as "study tour", and subsequently use the query statement corresponding to the word slot of "study tour" to interact with the user in a conversation.

[0111] Optionally, the access information may also include:

[0112] User attribute information, wherein the attribute information includes but is not limited to at least one of the following: page information of the page currently visited by the user, question statements sent by the user on the currently visited page, user tags, and user network address information.

[0113] For example, if the page currently visited by the user is a page for the "study and travel" service, the user data processing device can determine that the user's target service intention is "study and travel". If the question statement sent by the user on the page currently visited is "Which country is good for studying abroad", the user data processing device can determine that the user's target service intention is "study and travel". The user's label can be a label generated for the user by the user data processing device based on the user's historical browsing records and sessions, or it can be a label set by the user himself, which is not limited here. For example, if the label set by the user is "international student", the user data processing device can determine that the user's target service intention is "study abroad". The user data processing device can determine the user's location based on the user's network address information. For example, if the user data processing device determines that the user is located in the United States, English will be used to communicate with the user.

[0114] The above method may further include:

[0115] Determine a welcome message based on the attribute information and push the welcome message to the user.

[0116] In this embodiment, when a user visits the current page, the user data processing device can determine a welcome message based on the attribute information. For example, welcome messages in different languages, different user tags, and different business intentions can be pre-set, and the user data processing device determines the corresponding welcome message based on the attribute information, and pushes the welcome message to the user to implement personalized welcome message push and attract the user to engage in conversation. For example, according to the different country pages visited by the user, welcome messages introducing different countries can be pushed.

[0117] A user data processing device is pre-set with multiple business intents, each business intent corresponds to multiple word slots, and each word slot corresponds to a group of query statements. A group of query statements may include one query statement or multiple query statements, which is not limited here. A word slot corresponding to a business intent is the corresponding information in the user data that needs to be obtained by the business intent. The query statement corresponding to a word slot is used to ask questions to the user so as to extract the word slot from the conversation statement replied by the user.

[0118] Taking the academic scenario as an example, three business intents are preset in the user data processing device, namely "study abroad", "study and travel" and "examination and training". Among them, the word slots corresponding to "study abroad" may include: study abroad country, admission degree, personal situation, grade in school, preparation stage, telephone, social network account, title, city, customer source, etc. The word slots corresponding to "study and travel" may include: study abroad country, personal situation, telephone, social network account, title, city, customer source, etc. The word slots corresponding to "examination and training" may include: language courses, language situation, telephone, social network account, title, city, customer source, etc.

[0119] Among them, the query sentence corresponding to the word slot study abroad country may include "Have you considered studying abroad in any country? For example: the United States, the United Kingdom, Canada, etc."; the query sentence corresponding to the word slot admission degree may include "What level of study abroad course do you want to study? For example: undergraduate, graduate, etc."; the query sentence corresponding to the word slot grade in school may include "What grade are you currently in, for example: sophomore, junior, etc.". The query sentence corresponding to the word slot preparation stage may include "Please tell me about your current preparation stage, for example: your usual grades, language skills, etc."

[0120] S203: Use the query sentence of the word slot corresponding to the target business intent to perform conversation interaction with the user to obtain the conversation sentence input by the user.

[0121] S204: Generate and output user data corresponding to the target business intent based on the conversation sentence to understand the intent.

[0122] In this embodiment, after determining the target business intent of the user, the user data processing device obtains the query sentence of the word slot corresponding to the target business intent, pushes the query sentence to the user, and then receives the conversation sentence replied by the user to realize the conversation interaction with the user. The user data processing device can extract the vocabulary used to fill the word slot from the conversation sentence, and then generate the user data corresponding to the target business intent according to the word slot and the corresponding vocabulary, and can output the user data to the staff of the enterprise in the form of table, text, etc., for example, send the user data to the terminal of the staff so that the user can view or download it from the terminal.

[0123] In a possible application scenario, the user data processing device is a server. The server obtains the query statement of the word slot corresponding to the target business intent and sends the query statement to the terminal. The terminal displays the query statement in the conversation window, receives the conversation statement input by the user in the conversation window, and sends the conversation statement to the server to realize the conversation interaction with the user. Figure 3As shown, the conversation window is superimposed on the current page for display, and the terminal displays the query statement sent by the server in the conversation window. The user can select the conversation window and enter the conversation statement in the conversation window in text form. The content of the conversation interaction in this example takes the business intention "study abroad" as an example, and the current page can be a page displaying the study abroad business.

[0124] In a possible application scenario, the user data processing device is a terminal. The terminal obtains the query sentence of the word slot corresponding to the target business intent, displays the query sentence in the conversation window, and receives the conversation sentence input by the user in the conversation window to realize the conversation interaction with the user.

[0125] The embodiment of the present application obtains the user's access information, determines the user's target business intent based on the access information, the target business intent corresponds to multiple word slots, each word slot corresponds to a group of query statements, and then uses the query statements of the word slots corresponding to the target business intent to interact with the user in a conversation, obtains the conversation statements input by the user, generates and outputs user data corresponding to the target business intent based on the conversation statements, and can automatically interact with the user and obtain user data through the word slots and query statements corresponding to the preset business intent, thereby improving the efficiency of obtaining user data.

[0126] Optionally, the conversation sentence includes a question sentence. The method may further include:

[0127] Extract words that match the word slots corresponding to the target business intent from the question sentence and fill them into the corresponding word slots.

[0128] In this embodiment, during the conversation interaction process, the conversation sentence input by the user may be a reply to the pushed inquiry sentence, or may be a question sentence actively asked by the user. For the question sentence input by the user, the user data processing device can identify whether there is a word in the question sentence that matches the word slot corresponding to the target business intent, and if so, the word can be extracted and filled into the corresponding word slot.

[0129] For example, during a conversation interaction, the user inputs a question sentence such as "What is the ranking of schools in the United States?" The user data processing device can reply to the question sentence and extract the word "United States" from the question sentence and fill it into the word slot "study abroad country".

[0130] During the conversation interaction with the user, this embodiment not only fills the word slots based on the user's replies to the questions asked, but also identifies vocabulary matching the word slots from the question sentences actively asked by the user to fill the word slots. By filling the word slots when the user actively asks questions, the efficiency of word slot filling can be improved.

[0131] Optionally, the above method may further include:

[0132] After receiving the question statement, reply to the question statement, and after replying to the question statement, jump to execute the following steps: use the inquiry statement of the word slot corresponding to the target business intent to interact with the user in a conversation.

[0133] In this embodiment, when an inquiry statement in the word slot corresponding to the target business intent is used to interact with the user in a conversation, if a question statement input by the user is received, the original conversation process of the target business intent is terminated, and the question statement input by the user is first replied. After replying to the question statement, the conversation process is jumped back to the original conversation process of the target business intent to continue the conversation with the user.

[0134] Among them, responding to the question statement includes:

[0135] Questions matching the question statement are searched from a preset question-and-answer database, and answer texts corresponding to the questions matching the question statement are pushed to the user, wherein the question-and-answer database includes multiple sets of corresponding questions and answer texts.

[0136] Alternatively, based on the table lookup logic corresponding to the question statement, data matching the question statement is searched from a preset knowledge data table, and an answer text is generated based on the data matching the question statement, and the answer text is pushed to the user.

[0137] In this embodiment, the user data processing device may reply to the question statement in any one of the following two implementation modes.

[0138] In a first possible implementation, the user data processing device may determine the answer text of the question statement based on a preset question-and-answer database, and push the answer text to the user. The question-and-answer database includes multiple sets of corresponding questions and answer texts. The user data processing device may search the question-and-answer database for a question that matches the question statement, and push the answer text of the question to the user.

[0139] In a second possible implementation, the user data processing device can search for data matching the question statement from a preset knowledge data table, generate an answer text from the data, and push it to the user. For example, the knowledge data table can record the tuition fees for each grade of each school in the United States. When the question statement is "How much is the tuition fee for the first grade of university at school X in the United States", the user data processing device can search for the tuition fee value corresponding to school X and the first grade of university from the knowledge data table as Y, generate an answer text from the tuition fee value such as "The tuition fee for the first grade of university at school X is Y", and push the answer text to the user.

[0140] This embodiment can accurately determine the answer text of the question statement through the preset question and answer database or knowledge data table, thereby improving the accuracy of answering questions.

[0141] Figure 4 A flowchart of a user data processing method combining RPA and AI provided in another embodiment of the present application. This embodiment describes in detail the specific implementation process of using the query statement corresponding to the target business intent to interact with the user in a conversation. Figure 4 As shown, the method includes:

[0142] S401: Obtain user access information.

[0143] In this embodiment, S401 and Figure 2 S201 in the embodiment is similar and will not be described again here.

[0144] S402: Determine the user's target business intention according to the access information, wherein the target business intention corresponds to a plurality of word slots, and each word slot corresponds to a group of query statements.

[0145] In this embodiment, S402 and Figure 2 S202 in the embodiment is similar and will not be described again here.

[0146] S403: Determine unfilled word slots in the word slots corresponding to the target business intent.

[0147] S404: Use the query sentence corresponding to the unfilled word slot to conduct conversation interaction with the user.

[0148] In this embodiment, after the user data processing device determines the target business intent, it can obtain all word slots corresponding to the target business intent. Since some of the word slots corresponding to the target business intent may have been filled, and some may not be filled, the user data processing device can determine the unfilled word slots. For example, after pushing a welcome message to the user, a question statement input by the user is received, and vocabulary has been extracted based on the user's question statement to fill some word slots, then the query statements of these word slots do not need to be pushed to the user again. The user data processing device uses the query statements corresponding to the unfilled word slots to interact with the user in a conversation.

[0149] This embodiment only pushes query sentences of unfilled word slots to the user, avoiding the problem of repeated pushes, which can improve the user experience and increase the efficiency of obtaining user data.

[0150] Optionally, there is at least one unfilled word slot. S404 may include:

[0151] For each unfilled word slot, the query sentence corresponding to the unfilled word slot is pushed to the user, and the conversation sentence input by the user is received.

[0152] In this embodiment, when there are multiple unfilled word slots in the target business intent, the query statements corresponding to the unfilled word slots can be pushed to the user in a preset order multiple times, and the conversation statements input by the user can be received.

[0153] Optionally, the above method may further include:

[0154] Determine whether there is a new business intent based on the conversation sentence, and when it is determined that there is a new business intent, use the query sentence of the word slot corresponding to the new business intent to interact with the user in the conversation to fill the word slot corresponding to the new business intent.

[0155] After the word slot corresponding to the new business intent is filled, jump to the following steps: use the query sentence of the word slot corresponding to the target business intent to interact with the user in a conversation.

[0156] In this embodiment, during the process of using the query statement of the target business intent to interact with the user, if the conversation statement input by the user is related to another business intent, it is determined that there is a new business intent, and the query statement of the new business intent is used to interact with the user in a conversation, so as to obtain the user data corresponding to the new business intent. After the word slot corresponding to the new business intent is filled, it jumps back and continues to use the query statement of the target business intent to interact with the user in a conversation.

[0157] Taking the academic scenario as an example, if the query sentence in the word slot corresponding to "study abroad" is used to interact with the user, and the user enters the conversation sentence "What conditions are required for study tour", the conversation process about "study abroad" is terminated, and the query sentence in the word slot corresponding to "study tour" is used to interact with the user. After the word slot corresponding to "study tour" is filled, the conversation process about "study abroad" is switched to continue the conversation interaction with the user.

[0158] In this embodiment, when the user generates a new service intention, timely conversation interaction with the user is performed according to the new service intention, which can improve the user experience, enhance the user's willingness to continue the conversation, and improve the success rate of user data acquisition.

[0159] In one implementation, before using a query statement of a word slot corresponding to a new business intent to conduct a conversation interaction with a user, data is retained for the word slots preset in the filled word slots and shared with the new business intent.

[0160] In this embodiment, the same word slots may exist in the word slots of different business intents, and the same word slots of two or more business intents can be set as shared word slots in advance. Still taking the above-mentioned academic scenario as an example, the word slots corresponding to the business intent "study abroad" may include: study abroad country, admission degree, personal situation, grade in school, preparation stage, phone, social network account, title, city, customer source. The word slots corresponding to the business intent "study tour" may include: study tour country, personal situation, phone, social network account, phone, title, city, customer source. The word slots corresponding to the business intent "examination and training" may include: language courses, language situation, phone, social network account, title, city, customer source. Among them, phone, social network account, title, city, customer source, these word slots can be preset as shared word slots of the above three business intents; personal situation, this word slot can be preset as a shared word slot of "study abroad" and "study tour".

[0161] When the user's business intent switches from the target business intent to a new business intent, before the conversation interaction, the word slots currently filled with the new business intent can be shared with the new business intent. The data can be retained, such as automatically filling the vocabulary of these shared word slots into the corresponding word slots of the new business intent, or storing the vocabulary of these shared word slots for use when generating user data for the new business intent. Alternatively, if the word slots themselves are shared by two business intents, the new business intent will receive feedback that the word slots have been filled when the corresponding word slots need to be filled, so that when the conversation interaction is carried out with the user according to the query statements of the word slots corresponding to the new business intent, it is not necessary to repeat the questions of these word slots to ask the user. By sharing word slots, it is possible to avoid repeatedly asking the same user the same question, improve the intelligence of the robot, and enhance the user experience.

[0162] S405: Generate and output user data corresponding to the target business intent according to the conversation statement.

[0163] In this embodiment, S405 and Figure 2 S204 in the embodiment is similar and will not be described again here.

[0164] Figure 5 This is a flowchart of a user data processing method combining RPA and AI provided in another embodiment of the present application. This embodiment describes in detail the specific implementation process of generating user data corresponding to the target business intent according to the conversation sentence. Figure 5 As shown, the method includes:

[0165] S501. Obtain user access information.

[0166] In this embodiment, S501 and Figure 2 S201 in the embodiment is similar and will not be described again here.

[0167] S502: Determine the user's target business intent according to the access information, wherein the target business intent corresponds to a plurality of word slots, and each word slot corresponds to a group of query statements.

[0168] In this embodiment, S502 and Figure 2 S202 in the embodiment is similar and will not be described again here.

[0169] S503: Use the query sentence of the word slot corresponding to the target business intent to perform conversation interaction with the user to obtain the conversation sentence input by the user.

[0170] In this embodiment, S503 and Figure 2 S203 in the embodiment is similar and will not be described again here.

[0171] S504: Extract words that match the word slots corresponding to the target business intent from the conversation sentences, and fill the words into the corresponding word slots.

[0172] S505 : Generate and output user data corresponding to the target business intent according to the word slots corresponding to the target business intent and the words filled therein.

[0173] In this embodiment, during the conversation interaction, the user data processing device can identify the vocabulary in the conversation sentence that matches the word slot corresponding to the target business intent, extract the vocabulary, and fill the vocabulary into the corresponding word slot. After the conversation interaction ends, based on the word slot corresponding to the target business intent and the vocabulary filled therein, user data corresponding to the target business intent is generated and output.

[0174] Optionally, the word slot includes a contact information word slot; S505 may include:

[0175] Determine whether the contact word slot is filled;

[0176] When the contact information word slot is filled, user data corresponding to the target business intent is generated and output according to the word slot corresponding to the target business intent and the words filled therein.

[0177] In this embodiment, the contact information can be a telephone number, a social account, an email account, etc., which is not limited here. Among all the word slots corresponding to each business intent, there is at least one contact information word slot, which is used to fill in the user's contact information so that subsequent staff can contact the user through the contact information to promote products or provide services. During the conversation with the user, the user data processing device determines whether the contact information word slot of the target business intent is filled. If so, it generates and outputs the user data corresponding to the target business intent based on the word slot corresponding to the target business intent and the vocabulary filled therein, otherwise the user data is not generated.

[0178] Since the user's contact information is the most critical in the user data, only by obtaining the user's contact information can the subsequent staff contact the user. If the user's contact information cannot be obtained, even if the user data corresponding to the target business intent is generated, the user data is useless. This embodiment determines whether the contact information word slot is filled, and only generates and outputs user data when the contact information word slot is filled, which can avoid generating useless user data and improve the effect of obtaining user data. Optionally, the user data processing device can guide the user to leave contact information according to the preset information retention process and words. For example, it is preferred to guide the user to leave a phone number. If the user refuses to leave a phone number, other contact information is obtained.

[0179] In one implementation, when it is determined that the contact information word slot is filled and it is determined that there is a jump instruction, data is retained for the preset word slot shared with the target business intent before the jump.

[0180] In this embodiment, when it is determined that the contact information word slot is filled, it can be determined whether there is a jump instruction, which is used to instruct to jump from the conversation flow of the target business intent to the conversation flow of another business intent. When it is determined that there is a jump instruction, a corresponding jump is performed, and before the jump, data is retained for the preset word slot shared with the target business intent. By sharing the word slot, repeated questions about the same question to the same user are avoided, the intelligence of the robot is improved, and the user experience is improved.

[0181] Figure 6 A schematic diagram of a program unit for implementing a user data processing method combining RPA and AI provided in an embodiment of the present application. The user data processing method combining RPA and AI provided in this embodiment can be implemented by the following program units: a preprocessing unit 601, an event triggering unit 602, a general business central control unit 603, a branch business central control unit 604, and a branch business execution unit 605. These program units run in a user data processing device.

[0182] The pre-processing unit 601 is used for initialization settings, such as loading and initializing other units, question and answer databases, knowledge data tables, etc.

[0183] The event trigger unit 602 is used, on the one hand, to actively initiate a corresponding conversation or welcome message to the user when the page is initially loaded or switched, and on the other hand, to read the user's access information, such as: the keywords searched by the user, the user ID, the customer source, the device information, the page information of the page currently visited by the user, the question statement sent by the user on the currently visited page, the user's tags, the user's network address information, etc.

[0184] The general business central control unit 603 is used to identify the user's current target business intention according to the user's access information, and control the program flow to be distributed to the corresponding branch business central control unit 604, wherein each branch business central control unit 604 is used to control a conversation flow corresponding to a business intention, and the conversation flow is specifically executed by the branch business execution unit 605 to implement conversation interaction with the user according to the query statement of the word slot corresponding to the business intention, fill the word slot, and generate output user data. After each branch business central control unit 604 controls and executes the conversation flow, the program flow will jump back to the general business central control unit 603.

[0185] The branch service central control unit 604 is used to control the process of executing the session flow by the corresponding branch service execution unit 605. Each branch service central control unit 604 controls one branch service execution unit 605.

[0186] The branch business execution unit 605 is used to execute the conversation process of the corresponding business intent according to the control of the branch business central control unit 604, conduct conversation interaction with the user according to the query statement of the word slot corresponding to the business intent, and obtain the user data corresponding to the word slot through multiple rounds of conversations. Each branch business execution unit 605 corresponds to a business intent. In this embodiment, taking the academic scenario as an example, there are three branch business execution units 605, corresponding to the three business intents "study abroad", "study tour" and "exam training".

[0187] When a branch service central control unit 604 controls the corresponding branch service execution unit 605 to execute the conversation process of the service intention, if it is determined that the user has a new service intention according to the conversation statement input by the user, the branch service central control unit 604 controls the branch service execution unit 605 to terminate the conversation process and jump the program flow back to the main service central control unit 603, and the main service central control unit 603 allocates the program flow to the branch service central control unit 604 corresponding to the new service intention to realize the switching of the service intention. After the execution of the conversation process of the new service intention is completed, the main service central control unit 603 jumps the program flow to the original branch service execution unit 604 to continue to execute the previously terminated conversation process.

[0188] In the prior art, manual customer service interacts with users browsing the page, and the manual customer service fills the user's data into the user data table. Since one manual customer service can only interact with a few users at the same time, and the speed of manually filling in user data is slow, the efficiency of obtaining user data is low. In addition, there are the following problems: the number of manual customer service is small during peak customer flow and at night, making it difficult to interact with users in a timely manner; the interaction level of manual customer service is uneven, making it difficult to ensure the quality of user data obtained.

[0189] The user data processing method provided by this embodiment combines RPA and AI, automatically interacts with users through the word slots and query statements corresponding to the preset business intent, obtains user data, and thus improves the efficiency of obtaining user data, can guide user intentions, explore user needs, and efficiently complete the collection of user data. This embodiment can have conversations with many users at the same time, handle repetitive problems, and can subsequently transfer to manual or continue independent reception; it can have conversations with users at night, reduce labor costs, and improve experience and results; it can imitate the best sales customer service and maintain stable business capabilities.

[0190] Figure 7 This is a schematic diagram of the structure of a user data processing device combining RPA and AI provided in one embodiment of the present application. Figure 7 As shown, the user data processing device 70 combining RPA and AI includes: an acquisition module 701, a processing module 702, an interaction module 703 and an output module 704.

[0191] Natural Language Processing (NLP) applied to artificial intelligence (AI).

[0192] The acquisition module 701 is used to acquire the user's access information.

[0193] The processing module 702 is used to determine the user's target business intention according to the access information, wherein the target business intention corresponds to a plurality of word slots, and each word slot corresponds to a group of query statements.

[0194] The interaction module 703 is used to perform conversation interaction with the user by using the query sentence of the word slot corresponding to the target business intent to obtain the conversation sentence input by the user.

[0195] The output module 704 is used to generate and output user data corresponding to the target business intent according to the conversation sentence, so as to understand the intent.

[0196] Optionally, the interaction module 703 is specifically configured to:

[0197] Determine unfilled word slots in the word slots corresponding to the target business intent;

[0198] Use the query sentences corresponding to the unfilled word slots to interact with the user in a conversation.

[0199] Optionally, the number of unfilled word slots is at least one;

[0200] The interaction module 703 is specifically used for:

[0201] For each unfilled word slot, push the query sentence corresponding to the unfilled word slot to the user, and receive the conversation sentence input by the user;

[0202] The interaction module 703 is further used for:

[0203] Determine whether there is a new business intent based on the conversation sentence, and when it is determined that there is a new business intent, use the query sentence of the word slot corresponding to the new business intent to interact with the user in the conversation, and fill the word slot corresponding to the new business intent;

[0204] After the word slot corresponding to the new business intent is filled, jump to the following steps: use the query sentence of the word slot corresponding to the target business intent to interact with the user in a conversation.

[0205] Optionally, the interaction module 703 is further configured to:

[0206] Before using the query statement of the word slot corresponding to the new business intent to interact with the user in a conversation, data is retained for the word slots that are preset and shared with the new business intent among the filled word slots.

[0207] Optionally, the output module 704 is specifically configured to:

[0208] Extract words that match the word slots corresponding to the target business intent from the conversation sentences, and fill the words into the corresponding word slots;

[0209] According to the word slots corresponding to the target business intent and the words filled therein, user data corresponding to the target business intent is generated and output.

[0210] Optionally, the word slot includes a contact word slot;

[0211] The output module 704 is specifically used for:

[0212] Determine whether the contact word slot is filled;

[0213] When the contact information word slot is filled, user data corresponding to the target business intent is generated and output according to the word slot corresponding to the target business intent and the words filled therein.

[0214] Optionally, the output module 704 is further configured to:

[0215] When the contact information word slot is filled, it is determined that there is a jump instruction, and before the jump, data is retained for the preset word slot shared with the target business intent.

[0216] Optionally, the conversational statements include question statements;

[0217] The output module 704 is further used for:

[0218] Extract words that match the word slots corresponding to the target business intent from the question sentence and fill them into the corresponding word slots.

[0219] Optionally, the interaction module 703 is further configured to:

[0220] After receiving the question statement, reply to the question statement, and after replying to the question statement, jump to execute the following steps: use the inquiry statement of the word slot corresponding to the target business intent to interact with the user in a conversation;

[0221] Among them, responding to the question statement includes:

[0222] Searching for questions matching the question statement from a preset question-and-answer database, and pushing the answer text corresponding to the question matching the question statement to the user, wherein the question-and-answer database includes multiple sets of corresponding questions and answer texts;

[0223] Alternatively, based on the table lookup logic corresponding to the question statement, data matching the question statement is searched from a preset knowledge data table, and an answer text is generated based on the data matching the question statement, and the answer text is pushed to the user.

[0224] Optionally, the access information includes:

[0225] Keywords that users search for.

[0226] Optionally, the access information also includes:

[0227] User attribute information, wherein the attribute information includes at least one of the following: page information of the page currently visited by the user, question statements sent by the user on the currently visited page, user tags, and user network address information;

[0228] The interaction module 703 is further used for:

[0229] Determine a welcome message based on the attribute information and push the welcome message to the user.

[0230] The user data processing device combining RPA and AI provided in the embodiment of the present application can be used to execute the above-mentioned method embodiment. Its implementation principle and technical effects are similar, and this embodiment will not be repeated here.

[0231] Figure 8 This is a schematic diagram of the hardware structure of a user data processing device provided in one embodiment of the present application. Figure 8 As shown, the user data processing device 80 provided in this embodiment includes: at least one processor 801 and a memory 802. The user data processing device 80 also includes a communication component 803. The processor 801, the memory 802 and the communication component 803 are connected via a bus 804.

[0232] In the specific implementation process, at least one processor 801 executes the computer execution instructions stored in the memory 802, so that at least one processor 801 executes the user data processing method combining RPA and AI as described above.

[0233] The specific implementation process of the processor 801 can be found in the above method embodiment, and its implementation principle and technical effect are similar, so this embodiment will not be repeated here.

[0234] In the above Figure 8 In the illustrated embodiment, it should be understood that the processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), etc. A general-purpose processor may be a microprocessor or any conventional processor. The steps of the method disclosed in the application may be directly implemented as being executed by a hardware processor, or may be executed by a combination of hardware and software modules in the processor.

[0235] The memory may include a high-speed RAM memory, and may also include a non-volatile storage NVM, such as at least one disk storage.

[0236] The bus may be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, the bus in the drawings of the present application is not limited to only one bus or one type of bus.

[0237] The present application also provides a computer-readable storage medium, which stores computer-executable instructions. When a processor executes the computer-executable instructions, the user data processing method combining RPA and AI as described above is implemented.

[0238] The above-mentioned readable storage medium 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 memory, flash memory, magnetic disk or optical disk. The readable storage medium can be any available medium that can be accessed by a general or special-purpose computer.

[0239] An exemplary readable storage medium is coupled to a processor so that the processor can read information from the readable storage medium and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can be located in an application specific integrated circuit (ASIC). Of course, the processor and the readable storage medium can also exist in the device as discrete components.

[0240] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present application.

Claims

1. A user data processing method combining RPA and AI, characterized in that: Natural Language Processing (NLP) applied to artificial intelligence (AI), including: Obtain user access information; Determine the user's target business intent according to the access information, wherein the target business intent corresponds to a plurality of word slots, each word slot corresponds to a group of query statements; the target business intent is the business intent corresponding to the access information obtained; Performing conversational interaction with the user using the query sentence of the word slot corresponding to the target business intent to obtain the conversation sentence input by the user; According to the conversation sentence, generate and output user data corresponding to the target business intent to understand the intent; Wherein, the conversation sentence includes a question sentence; and the method further includes: Extracting words that match the word slots corresponding to the target business intent from the question sentence and filling them into the corresponding word slots; After receiving the question statement, reply to the question statement, and after replying to the question statement, jump to execute the following steps: use the inquiry statement of the word slot corresponding to the target business intent to interact with the user in a conversation; Wherein, responding to the question statement includes: Searching for a question matching the question statement from a preset question-and-answer database, and pushing the answer text corresponding to the question matching the question statement to the user, wherein the question-and-answer database includes multiple sets of corresponding questions and answer texts; or, Based on the table lookup logic corresponding to the question statement, search for data matching the question statement from a preset knowledge data table, generate an answer text according to the data matching the question statement, and push the answer text to the user; Generating and outputting user data corresponding to the target business intent according to the conversation statement, including: Extracting words that match the word slots corresponding to the target business intent from the conversation sentences, and filling the words into the corresponding word slots; Generate and output user data corresponding to the target business intent according to the word slot corresponding to the target business intent and the vocabulary filled therein; The word slots include contact information word slots; Generating and outputting user data corresponding to the target service intent according to the word slot corresponding to the target service intent and the vocabulary filled therein, including: Determining whether the contact information word slot is filled; When the contact information word slot is filled, generating and outputting user data corresponding to the target business intent according to the word slot corresponding to the target business intent and the words filled therein; When the contact information slot is not filled, no user data is generated.

2. The method according to claim 1, characterized in that Using the query sentence of the word slot corresponding to the target business intent to interact with the user in a conversation includes: Determine unfilled word slots in the word slots corresponding to the target business intent; The query sentence corresponding to the unfilled word slot is used to perform conversation interaction with the user.

3. The method according to claim 2, characterized in that The number of the unfilled word slots is at least one; and using the query statement corresponding to the unfilled word slots to perform conversational interaction with the user includes: For each unfilled word slot, push the query sentence corresponding to the unfilled word slot to the user; Receive conversation statements input by the user; After receiving the conversation statement input by the user, the method further includes: Determine whether there is a new business intent based on the conversation statement, and when it is determined that the new business intent exists, use the query statement of the word slot corresponding to the new business intent to interact with the user in a conversation, fill the word slot corresponding to the new business intent, and after the word slot corresponding to the new business intent is filled, jump to execute the following steps: use the query statement of the word slot corresponding to the target business intent to interact with the user in a conversation.

4. The method according to claim 3, characterized in that Before using the query statement of the word slot corresponding to the new business intent to perform conversation interaction with the user, the method further includes: For the word slots in the filled word slots that are preset and shared with the new business intent, data is retained.

5. The method according to claim 1, characterized in that When the contact information word slot is filled, the method further includes: It is determined that there is a jump instruction, and before the jump, data is retained for a preset word slot shared with the target business intent.

6. The method according to any one of claims 1 to 5, characterized in that: The access information includes: Keywords that users search for.

7. The method according to any one of claims 1 to 5, characterized in that: The access information also includes: User attribute information, wherein the attribute information includes at least one of the following: page information of the page currently visited by the user, question statements sent by the user on the currently visited page, user tags, and user network address information; The method further comprises: A welcome message is determined according to the attribute information, and the welcome message is pushed to the user.

8. A user data processing device combining RPA and AI, characterized in that: Natural language processing (NLP) applied to artificial intelligence (AI) includes: The acquisition module is used to obtain the user's access information; A processing module, configured to determine a target business intent of the user according to the access information, wherein the target business intent corresponds to a plurality of word slots, each word slot corresponds to a group of query statements; the target business intent is the business intent corresponding to the access information obtained; An interaction module, used to perform conversation interaction with the user by using the query sentence of the word slot corresponding to the target business intent, and obtain the conversation sentence input by the user; An output module, used to generate and output user data corresponding to the target business intent according to the conversation sentence, so as to understand the intent; Wherein, the conversational sentences include question sentences; and the output module is further used for: Extract words that match the word slots corresponding to the target business intent from the question sentence and fill them into the corresponding word slots The interaction module is further used for: After receiving the question statement, reply to the question statement, and after replying to the question statement, jump to execute the following steps: use the inquiry statement of the word slot corresponding to the target business intent to interact with the user in a conversation; Wherein, responding to the question statement includes: Searching for a question matching the question statement from a preset question-and-answer database, and pushing the answer text corresponding to the question matching the question statement to the user, wherein the question-and-answer database includes a plurality of sets of corresponding questions and answer texts; Alternatively, based on the table lookup logic corresponding to the question statement, data matching the question statement is searched from a preset knowledge data table, and an answer text is generated according to the data matching the question statement, and the answer text is pushed to the user; The output module is specifically used to extract words that match the word slots corresponding to the target business intent from the conversation sentence, and fill the words into the corresponding word slots; Generate and output user data corresponding to the target business intent according to the word slot corresponding to the target business intent and the vocabulary filled therein; The word slots include contact information word slots; The output module is further used to determine whether the contact information word slot is filled; When the contact information word slot is filled, generating and outputting user data corresponding to the target business intent according to the word slot corresponding to the target business intent and the words filled therein; When the contact information slot is not filled, no user data is generated.

9. A user data processing device, characterized in that: include: at least one processor and memory; The memory stores computer-executable instructions; The at least one processor executes the computer-executable instructions stored in the memory, so that the at least one processor executes the user data processing method combining RPA and AI as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions. When the processor executes the computer-executable instructions, the user data processing method combining RPA and AI as described in any one of claims 1 to 7 is implemented.

Images