TWI900034B - Realize human-computer interaction and evaluate recommendation through artificial intelligence system and method thereof - Google Patents

Abstract

A realize human-computer interaction and evaluate recommendation through artificial intelligence system and a method thereof are provided. Basic information of virtual customer and personality information of virtual customer are generated randomly by virtual customer artificial intelligence platform when salesperson terminal device interface with virtual customer artificial intelligence platform. Customer language model which is correspondence with personality information of virtual customer is selected by virtual customer artificial intelligence platform. Salesperson terminal device and virtual customer artificial intelligence platform are human-computer interaction using corresponding customer language model. Business service evaluation and business service improvement suggestion are generated for human-computer interaction by observer artificial intelligence platform. Therefore, the efficiency of realizing human-computer interaction and evaluate recommendation through artificial intelligence may be achieved.

Description

System and method for realizing human-computer interaction and evaluation suggestion through artificial intelligence

An evaluation and recommendation system and method thereof, particularly a system and method thereof, in which a virtual client artificial intelligence platform provides human-computer interaction, and an observer artificial intelligence platform evaluates and provides recommendations on the human-computer interaction through artificial intelligence.

Artificial intelligence (AI) is a field of computer science that refers to the ability of computers, programs, and other systems and machines to independently draw inferences through data analysis and simulate human-like thinking logic and execution strategies.

Currently, artificial intelligence (AI) can enable one-on-one human-machine interaction. For example, AlphaGo, an AI Go program developed by Google DeepMind in London, UK, has demonstrated AI's potential in logical judgment and automated tasks, creating endless possibilities for future applications.

Existing AI logical judgments often require extensive human-computer interaction to evaluate their effectiveness. This is time-consuming and labor-intensive, requiring human review of the results. Developing a system that can rapidly evaluate human-computer interaction results and provide recommendations is a challenge that needs to be addressed.

In summary, it can be seen that the previous technology has long had the problem that using artificial intelligence to evaluate the results of human-computer interaction is time-consuming and labor-intensive, and does not provide any suggestions for the human-computer interaction results. Therefore, it is necessary to propose improved technical means to solve this problem.

In view of the problems that the prior art has in using artificial intelligence to evaluate the results of human-computer interaction, which is time-consuming and labor-intensive and does not provide recommendations for the human-computer interaction results, the present invention discloses a system and method for implementing human-computer interaction and evaluation recommendations through artificial intelligence, wherein:

The system disclosed herein uses artificial intelligence to implement human-machine interaction and provide evaluation and recommendations. It includes a salesperson terminal device, a virtual customer artificial intelligence platform, and an observer artificial intelligence platform. The virtual customer artificial intelligence platform further includes a first interface module, a virtual customer data generation module, a selection module, and a human-machine interaction module. The observer artificial intelligence platform further includes a second interface module and a business service evaluation and recommendation module.

The salesperson's terminal device interfaces with the virtual customer artificial intelligence platform and the observer artificial intelligence platform, providing human-machine interaction messages to the virtual customer artificial intelligence platform, obtaining human-machine interaction response messages from the virtual customer artificial intelligence platform, providing corresponding human-machine interaction messages and human-machine interaction response messages to the observer artificial intelligence platform, and obtaining corresponding business service evaluations and business service improvement suggestions from the observer artificial intelligence platform.

The first interface module of the virtual customer artificial intelligence platform is to provide an interface with the salesperson terminal device, obtain human-machine interaction information from the salesperson terminal device, and provide human-machine interaction response information to the salesperson terminal device; the virtual customer data generation module of the virtual customer artificial intelligence platform is to randomly generate virtual customer basic information and virtual customer personal information when interfacing with the salesperson terminal device. The virtual customer artificial intelligence platform's selection module selects a corresponding customer language model based on the virtual customer's personal data, and the customer language model is trained and established using training data corresponding to the virtual customer's personal data; and the virtual customer artificial intelligence platform's human-computer interaction module uses the selected customer language model to generate human-computer interaction response messages from human-computer interaction inquiry messages.

The second interface module of the observer artificial intelligence platform is to provide an interface with the salesperson terminal device, obtain the corresponding human-machine interaction information and human-machine interaction answer information from the salesperson terminal device, and provide business service evaluation and business service improvement suggestions to the salesperson terminal device; and the business service evaluation and suggestion module of the observer artificial intelligence platform uses the dialogue evaluation language model to evaluate the human-machine interaction inquiry information and human-machine interaction answer information. The system evaluates the smoothness of human-computer interaction, customer sentiment in human-computer interaction, emergency response in human-computer interaction, data collection of human-computer interaction inquiry messages, semantic accuracy of human-computer interaction inquiry messages, and professionalism of product and service introductions in human-computer interaction inquiry messages. It generates business service evaluations based on corresponding evaluation items and generates business service improvement suggestions based on the evaluation process and evaluation items.

The method disclosed in the present invention for implementing human-computer interaction and evaluation suggestions through artificial intelligence includes the following steps:

First, the salesperson terminal device is interfaced with the virtual customer artificial intelligence platform and the observer artificial intelligence platform respectively; then, when the virtual customer artificial intelligence platform is interfaced with the salesperson terminal device, the virtual customer artificial intelligence platform randomly generates virtual customer basic information and virtual customer personality data; then, the virtual customer artificial intelligence platform selects the corresponding customer language model based on the virtual customer personality data, and the customer language model is trained using the corresponding virtual customer personality data. Training data is established; then, the salesperson terminal device provides human-machine interaction information to the virtual customer artificial intelligence platform; then, the virtual customer artificial intelligence platform uses speech-to-text technology to convert the human-machine interaction information into human-machine interaction inquiry information; then, the virtual customer artificial intelligence platform uses the selected customer language model to generate human-machine interaction answer information from the human-machine interaction inquiry information; then, the virtual customer artificial intelligence platform provides the human-machine interaction answer information to the salesperson terminal device; then, the observer artificial intelligence platform obtains the corresponding human-machine interaction message and human-machine interaction answer message from the salesperson's terminal device; then, the observer artificial intelligence platform uses speech-to-text technology to convert the human-machine interaction voice message into a human-machine interaction inquiry message; then, the observer artificial intelligence platform uses the dialogue evaluation language model to evaluate the human-machine interaction inquiry message and the human-machine interaction answer message for the smoothness of human-machine interaction, the customer emotion of human-machine interaction, and the human-machine interaction. The system evaluates the emergency response, data collection of human-machine interaction inquiry messages, the semantic accuracy of human-machine interaction inquiry messages, and the professionalism of product and service introductions in human-machine interaction inquiry messages. It generates business service evaluations based on corresponding evaluation items and generates business service improvement suggestions based on the evaluation process and evaluation items. Finally, the Observer AI platform provides business service evaluations and business service improvement suggestions to the salesperson's terminal device.

The system and method disclosed herein are as described above. When the virtual customer artificial intelligence platform interfaces with a salesperson's terminal device, the virtual customer artificial intelligence platform randomly generates basic virtual customer data and virtual customer personality data. Based on the virtual customer personality data, a corresponding customer language model is selected. The salesperson's terminal device and the virtual customer artificial intelligence platform then interact using the corresponding customer language model. The observer artificial intelligence platform then generates a business service evaluation and business service improvement suggestions based on this human-computer interaction.

Through the above-mentioned technical means, the present invention can achieve the technical effect of realizing human-computer interaction and evaluation suggestions through artificial intelligence.

The following will be used in conjunction with drawings and embodiments to explain in detail the implementation of the present invention, so that the implementation process of how the present invention applies technical means to solve technical problems and achieve technical effects can be fully understood and implemented accordingly.

The following first describes the system disclosed in the present invention for implementing human-computer interaction and evaluation suggestions through artificial intelligence. Please refer to "Figure 1", which is a system block diagram of the system disclosed in the present invention for implementing human-computer interaction and evaluation suggestions through artificial intelligence.

The system disclosed herein, which implements human-machine interaction and evaluation and recommendation through artificial intelligence, includes a salesperson terminal device 10, a virtual customer artificial intelligence platform 20, and an observer artificial intelligence platform 30. The virtual customer artificial intelligence platform 20 further includes a first interface module 21, a virtual customer data generation module 22, a selection module 23, and a human-machine interaction module 24. The observer artificial intelligence platform 30 further includes a second interface module 31 and a business service evaluation and recommendation module 32.

The salesperson terminal device 10 may be, for example, a general computer, a laptop, a tablet, a smartphone, etc. These examples are given here for illustration only and do not limit the scope of application of the present invention. The first interface module 21 of the virtual customer artificial intelligence platform 20 and the second interface module 31 of the observer artificial intelligence platform 30 provide the salesperson terminal device 10 with an interface via a wired transmission method or a wireless transmission method. The aforementioned wired transmission method may be, for example, a cable network, an optical fiber network, etc., and the aforementioned wireless transmission method may be, for example, Wi-Fi, a mobile communication network (e.g., 4G, 5G, 6G, etc.), etc. These examples are given here for illustration only and do not limit the scope of application of the present invention.

When the virtual customer artificial intelligence platform 20 is interfaced with the salesperson terminal device 10, the virtual customer data generation module 22 of the virtual customer artificial intelligence platform 20 randomly generates virtual customer basic data and virtual customer personal data. The virtual customer basic data may include, for example, name, registered address, correspondence address, contact number, mobile phone number, etc. The virtual customer personal data may include, for example, a combination of defensive personality, confrontational personality, positive personality, negative personality, rational thinking personality, impulsive personality, etc. These are merely examples and are not intended to limit the scope of application of the present invention.

Then, the selection module 23 of the virtual customer artificial intelligence platform 20 can select the corresponding customer language model 41 according to the virtual customer personality data randomly generated by the virtual customer data generation module 22 of the virtual customer artificial intelligence platform 20. It is worth noting that the customer language model 41 is trained and established by the external model building device 40 using the training data that conforms to the virtual customer personality data, and then imported into the virtual customer artificial intelligence platform 20 to conform to the human-computer interaction of the virtual customer personality data. Specifically, if the virtual customer personality data is "negative personality", the training data used by the customer language model 41 is negative. For example, data such as: I don't need a recommendation, I'm busy now, I don't need this, etc., is used to establish customer language model 41 to match the human-computer interaction of the virtual customer personality data as "negative personality". If the virtual customer personality data is "positive personality", the training data used by customer language model 41 may be, for example,: I want to ask about the details of the claim conditions, Can you provide a spreadsheet, About the payment method of the premium, etc., to establish customer language model 41 to match the human-computer interaction of the virtual customer personality data as "positive personality". This is only an example for illustration and does not limit the scope of application of the present invention.

When a salesperson operates the salesperson terminal device 10, he or she inputs human-machine interaction information through the touch unit, microphone, near-field communication, wireless transmission, etc. provided by the salesperson terminal device 10. Then, the salesperson terminal device 10 can provide the human-machine interaction information to the virtual customer artificial intelligence platform 20. The virtual customer artificial intelligence platform 20 obtains the human-machine interaction information from the salesperson terminal device 10 through the first interface module 21. It is worth noting that the above-mentioned human-machine interaction information modes include text mode, voice mode, image mode, and video mode. These are only examples for illustration and do not limit the scope of application of the present invention.

Next, the human-computer interaction module 24 of the virtual customer artificial intelligence platform 20 can use the customer language model 41 selected by the selection module 23 to generate a human-computer interaction answer message from the human-computer interaction inquiry message, and then the first interface module 21 of the virtual customer artificial intelligence platform 20 provides the human-computer interaction answer message to the salesperson terminal device 10.

It is worth noting that the virtual customer artificial intelligence platform 20 further includes a first modal conversion module 25, which provides for converting the modalities of human-machine interaction inquiry messages and human-machine interaction response messages. It is worth noting that the first modal conversion module 25 uses a combination of speech-to-text, text-to-speech, optical character recognition, AI image generation, and AI video generation technologies to convert the modalities of human-machine interaction inquiry messages/human-machine interaction response messages.

Specifically, if the mode of the human-machine interactive inquiry message is text mode, and the first interface module 21 obtains from the salesperson terminal device 10 that the mode to be converted is image mode, the first mode conversion module 25 can use AI image generation technology to convert the human-machine interactive inquiry message in text mode into a human-machine interactive inquiry message in image mode; if the mode of the human-machine interactive answer message is voice mode, and the first interface module 21 obtains from the salesperson terminal device 10 that the mode to be converted is image mode, the first mode conversion module 25 can use AI image generation technology to convert the human-machine interactive inquiry message in text mode into a human-machine interactive inquiry message in image mode; The service terminal device 10 determines that the mode to be converted is an image mode. The first mode conversion module 25 first uses speech-to-text technology to convert the human-machine interaction response message in the voice mode into a human-machine interaction response message in the text mode. Then, it uses AI image generation technology to convert the human-machine interaction response message in the text mode into a human-machine interaction response message in the image mode. This is merely an example and does not limit the scope of application of the present invention.

After the salesperson terminal device 10 obtains the human-machine interaction response message from the virtual customer artificial intelligence platform 20, the salesperson terminal device 10 then provides the corresponding human-machine interaction message and human-machine interaction response message to the observer artificial intelligence platform 30. The observer artificial intelligence platform 30 can then obtain the corresponding human-machine interaction message and human-machine interaction response message from the salesperson terminal device 10 through the second interface module 31.

The business service evaluation and recommendation module 32 of the observer artificial intelligence platform 30 uses a dialogue evaluation language model 42 to evaluate the fluency of human-machine interaction, customer emotions in human-machine interaction, emergency handling in human-machine interaction, data collection of human-machine interaction inquiry messages, semantic accuracy of human-machine interaction inquiry messages, and professionalism of product and service introductions in human-machine interaction inquiry messages in human-machine interaction inquiry messages. It generates business service evaluations based on corresponding evaluation items and generates business service improvement suggestions based on the evaluation process and the evaluation of the evaluation items.

Specifically, the business service evaluation and suggestion module 32 uses the dialogue evaluation language model 42 to identify the number of particles in the human-computer interaction inquiry message to evaluate the fluency of the human-computer interaction. Assuming that the number of particles used in the human-computer interaction inquiry message is 4, the fluency of the human-computer interaction can be evaluated as "relatively smooth" (which can be obtained from a pre-established particle number evaluation comparison table). Then, based on the evaluation process and the evaluation items, business service improvement suggestions are generated. For example, it is recommended to familiarize yourself with the claims content of XX insurance, to practice more oral conversation, etc. These are just examples for illustration and do not limit the scope of application of the present invention.

The business service evaluation and suggestion module 32 uses the dialogue evaluation language model 42 to identify emotional words/emotional sentences in the human-computer interaction response message (for example: happy, bored, please don't explain anymore, I don't need insurance... etc., which are only given as examples here and do not limit the scope of application of the present invention) to evaluate the customer's emotions in the human-computer interaction, and then generates business service improvement suggestions based on the evaluation process and the evaluation of the evaluation items, such as: attention should be paid to the problems that customers are dissatisfied with, to build high customer satisfaction, please familiarize yourself with the insurance-related explanation methods... etc., which are only given as examples here and do not limit the scope of application of the present invention.

The business service evaluation and suggestion module 32 uses the dialogue evaluation language model 42 to identify the basic customer information in the human-machine interactive answer message to realize the data collection and evaluation of the human-machine interactive inquiry message. Assuming that the human-machine interactive inquiry message contains the name, mobile phone number and email address, but lacks the contact number and communication address, the data collection and evaluation of the human-machine interactive inquiry message is 60% (or 6/1 0, 0.6, etc., are given here as examples only and are not intended to limit the scope of application of the present invention). Business service improvement suggestions are then generated based on the evaluation process and the evaluation of the evaluation items. For example, suggestions such as: "The collection of basic customer information is insufficient; please inquire for more relevant information" and "Inquiring for too much basic information may cause customer disgust" are given here as examples only and are not intended to limit the scope of application of the present invention.

The business service evaluation and suggestion module 32 uses the dialogue evaluation language model 42 to compare the product and service introduction description of the human-computer interaction inquiry message with the product and service introduction template to achieve a professional evaluation of the product and service introduction in the human-computer interaction inquiry message. Assuming that the comparison between the product and service introduction description and the product and service introduction template is 85%, the professional evaluation of the product and service introduction is 85% (or 8. 5/10, 0.85, etc., are merely examples here and are not intended to limit the scope of application of the present invention). Business service improvement suggestions are then generated based on the evaluation process and the evaluation of the evaluation items, such as: whether the professional requirements of product and service introductions have been met, whether the professional requirements of product and service introductions still need to be further strengthened, etc. These are merely examples here and are not intended to limit the scope of application of the present invention.

It is worth noting that the dialogue evaluation language model 42 is also created by the external model building device 40 using the corresponding training data, and then imported into the observer artificial intelligence platform 30 to provide human-computer interaction evaluation, so as to generate business service evaluations based on corresponding evaluation items, and generate business service improvement suggestions based on the evaluation process and the evaluation of the evaluation items.

After the business service evaluation and suggestion module 32 generates the business service evaluation and business service improvement suggestions, the second interface module 31 of the observer artificial intelligence platform 30 provides the business service evaluation and business service improvement suggestions to the salesperson terminal device 10.

It is worth noting that the observer artificial intelligence platform 30 further includes a second modal conversion module 33, which provides a mode conversion between human-computer interaction inquiry messages and human-computer interaction answer messages, as well as a mode conversion between business service evaluations and business service improvement suggestions. It is worth noting that the relevant description of the second modal conversion module 33 can refer to the description of the above-mentioned first modal conversion module 25 and will not be repeated here.

Please refer to "Figure 2" and "Figure 3". "Figure 2" is a schematic diagram of an embodiment of the present invention that uses artificial intelligence to implement human-computer interaction and evaluation suggestions for human-computer interaction and human-computer interaction response messages, and "Figure 3" is a schematic diagram of an embodiment of the present invention that uses artificial intelligence to implement business service evaluation and business service improvement suggestions for human-computer interaction and evaluation suggestions.

In "Figure 2" and "Figure 3", a complete embodiment is presented, that is, a complete human-machine interaction implementation in which the salesperson terminal device 10 and the virtual customer artificial intelligence platform 20 perform human-machine interaction messages 51 and human-machine interaction response messages 52, and the observer artificial intelligence platform 30 performs human-machine interaction between the salesperson terminal device 10 and the virtual customer artificial intelligence platform 20. The business service evaluation 53 and business service improvement suggestions 54 are merely examples and are not intended to limit the scope of application of the present invention. For the contents of the human-computer interaction message 51 and the human-computer interaction answer message 52, please refer to "Figure 2" and will not be further described here. For the contents of the business service evaluation 53 and business service improvement suggestions 54, please refer to "Figure 3" and will not be further described here.

Next, the operation method of the present invention will be described below, and please refer to "Figure 4A" and "Figure 4B" at the same time. "Figure 4A" and "Figure 4B" are flowcharts of the method of the present invention for realizing human-computer interaction and evaluation suggestions through artificial intelligence.

The method disclosed in the present invention for implementing human-computer interaction and evaluation suggestions through artificial intelligence includes the following steps:

First, the salesperson terminal device interfaces with the virtual customer artificial intelligence platform and the observer artificial intelligence platform (step 601). Then, when the virtual customer artificial intelligence platform interfaces with the salesperson terminal device, the virtual customer artificial intelligence platform randomly generates virtual customer basic information and virtual customer personality data (step 602). Then, the virtual customer artificial intelligence platform selects the corresponding customer language model based on the virtual customer personality data, and the customer language model is trained using the training data of the corresponding virtual customer personality data. Establish (step 603); then, the salesperson terminal device provides the human-machine interaction message to the virtual customer artificial intelligence platform (step 604); then, the virtual customer artificial intelligence platform uses speech-to-text technology to convert the human-machine interaction message into a human-machine interaction inquiry message (step 605); then, the virtual customer artificial intelligence platform uses the selected customer language model to generate a human-machine interaction answer message from the human-machine interaction inquiry message (step 606); then, the virtual customer artificial intelligence platform provides the human-machine interaction answer message to the salesperson The operator's terminal device (step 607); then, the observer artificial intelligence platform obtains the corresponding human-machine interaction message and human-machine interaction answer message from the operator's terminal device (step 608); then, the observer artificial intelligence platform uses speech-to-text technology to convert the human-machine interaction voice message into a human-machine interaction inquiry message (step 609); then, the observer artificial intelligence platform uses the dialogue evaluation language model to evaluate the human-machine interaction inquiry message and the human-machine interaction answer message for the smoothness of the human-machine interaction and the customer sentiment of the human-machine interaction. The business service evaluation is generated based on the evaluation process and the evaluation of the evaluation items, and business service improvement suggestions are generated based on the evaluation process and the evaluation of the evaluation items (step 610). Finally, the observer artificial intelligence platform provides the business service evaluation and business service improvement suggestions to the salesperson's terminal device (step 611).

In summary, when the virtual customer AI platform interfaces with the salesperson's terminal device, it randomly generates basic and personalized data for the virtual customer. Based on this data, it selects a corresponding customer language model, allowing the salesperson's terminal device and the virtual customer AI platform to interact using the corresponding customer language model. The observer AI platform then generates a business service evaluation and suggestions for business service improvements based on this interaction.

This technology can solve the problems of previous technologies that use artificial intelligence to evaluate human-computer interaction results, which are time-consuming and labor-intensive, and lack the ability to provide recommendations based on the results of human-computer interaction. It can then achieve the technical effectiveness of using artificial intelligence to realize human-computer interaction and evaluation recommendations.

While the embodiments disclosed above are intended to limit the scope of patent protection for this invention, these are not intended to directly limit the scope of patent protection for this invention. Anyone skilled in the art may make minor changes in the form and details of the implementation without departing from the spirit and scope of this invention. The scope of patent protection for this invention shall remain subject to the scope of the attached patent application.

10: Salesperson terminal device

20: Virtual Customer Artificial Intelligence Platform

21: First interface module

22: Virtual customer data generation module

23: Select module

24: Human-computer interaction module

25: First mode conversion module

30:Observer Artificial Intelligence Platform

31: Second interface module

32: Business Service Evaluation and Suggestion Module

33: Second mode conversion module

40: Model building device

41:Customer Language Model

42: Dialogue Evaluation Language Model

51: Human-computer interaction information

52: Human-computer interaction answering message

53:Business Service Evaluation

54: Business service improvement suggestions

Step 601: The salesperson terminal device interfaces with the virtual customer artificial intelligence platform and the observer artificial intelligence platform respectively.

Step 602: When the virtual customer artificial intelligence platform is connected to the salesperson terminal device, the virtual customer artificial intelligence platform randomly generates virtual customer basic information and virtual customer personal information.

Step 603: The virtual customer artificial intelligence platform selects a corresponding customer language model based on the virtual customer personality data. The customer language model is trained using the training data corresponding to the virtual customer personality data.

Step 604: The salesperson terminal device provides human-machine interaction information to the virtual customer artificial intelligence platform

Step 605: The virtual customer artificial intelligence platform uses speech-to-text technology to convert human-computer interaction messages into human-computer interaction inquiry messages.

Step 606: The virtual customer artificial intelligence platform uses the selected customer language model to generate a human-computer interaction answer message from the human-computer interaction inquiry message.

Step 607: The virtual customer artificial intelligence platform provides human-machine interactive answer messages to the salesperson’s terminal device

Step 608: The observer artificial intelligence platform obtains the corresponding human-machine interaction information and human-machine interaction response information from the salesperson terminal device.

Step 609: The observer artificial intelligence platform uses speech-to-text technology to convert the human-computer interaction voice message into a human-computer interaction inquiry message.

Step 610: The observer artificial intelligence platform uses the dialogue evaluation language model to evaluate the human-machine interaction inquiry message and the human-machine interaction answer message on the fluency of human-machine interaction, the customer sentiment of human-machine interaction, the emergency handling of human-machine interaction, the data collection of human-machine interaction inquiry message, the semantic accuracy of human-machine interaction inquiry message, and the professionalism of the product and service introduction in the human-machine interaction inquiry message. The platform generates a business service evaluation based on corresponding evaluation items, and generates business service improvement suggestions based on the evaluation process and the evaluation items.

Step 611: The Observer AI platform provides business service evaluation and business service improvement suggestions to the salesperson’s terminal device

Figure 1 is a system block diagram of a system for implementing human-machine interaction and evaluation suggestions through artificial intelligence according to the present invention. Figure 2 is a schematic diagram of a human-machine interaction message and a human-machine interaction response message according to an embodiment of the present invention for implementing human-machine interaction and evaluation suggestions through artificial intelligence. Figure 3 is a schematic diagram of a business service evaluation and business service improvement suggestion embodiment of the present invention for implementing human-machine interaction and evaluation suggestions through artificial intelligence. Figures 4A and 4B are flow charts of a method for implementing human-machine interaction and evaluation suggestions through artificial intelligence according to the present invention.

10: Salesperson terminal device

20: Virtual Customer Artificial Intelligence Platform

21: First Interface Module

22: Virtual Customer Data Generation Module

23: Select module

24: Human-computer interaction module

25: First modal conversion module

30: Observer Artificial Intelligence Platform

31: Second interface module

32: Business Service Evaluation and Suggestion Module

33: Second modal conversion module

41: Customer Language Model

42: Dialogue Evaluation Language Model

Claims (10)

A system for implementing human-machine interaction and evaluation suggestions through artificial intelligence includes: a salesperson terminal device, which interfaces with a virtual customer artificial intelligence platform and an observer artificial intelligence platform, provides human-machine interaction messages to the virtual customer artificial intelligence platform, obtains human-machine interaction response messages from the virtual customer artificial intelligence platform, provides the corresponding human-machine interaction messages and human-machine interaction response messages to the observer artificial intelligence platform, and obtains a corresponding business service evaluation and business service improvement suggestion from the observer artificial intelligence platform; a virtual customer artificial intelligence platform, the virtual customer artificial intelligence platform further including: a first interface module, providing an interface with the salesperson terminal device, obtaining the human-machine interaction message from the salesperson terminal device, and providing the human-machine interaction answer message to the salesperson terminal device; a virtual customer data generation module, randomly generating a virtual customer basic data and a virtual customer personality data when interfacing with the salesperson terminal device; a selection module, selecting a corresponding customer language model based on the virtual customer personality data, the customer language model being trained and established using training data corresponding to the virtual customer personality data; and a human-machine interaction module, using the selected customer language model to generate the human-machine interaction answer message from the human-machine interaction inquiry message; and An observer artificial intelligence platform, the observer artificial intelligence platform further comprising: a second interface module, providing an interface with the salesperson terminal device, obtaining the corresponding human-machine interaction message and the human-machine interaction response message from the salesperson terminal device, and providing the business service evaluation and the business service improvement suggestions to the salesperson terminal device; and A business service evaluation and suggestion module uses a dialogue evaluation language model to evaluate the human-machine interaction inquiry message and the human-machine interaction answer message on the fluency of human-machine interaction, customer sentiment in human-machine interaction, emergency handling in human-machine interaction, data collection of the human-machine interaction inquiry message, semantic accuracy of the human-machine interaction inquiry message, and professionalism of product and service introductions in the human-machine interaction inquiry message. The module generates the business service evaluation by providing corresponding evaluations based on different evaluation items, and generates business service improvement suggestions based on the evaluation process and the evaluation of the evaluation items. A system for implementing human-computer interaction and evaluation and recommendation through artificial intelligence as described in claim 1, wherein the business service evaluation and recommendation module uses the conversation evaluation language model to identify the number of interjections in the human-computer interaction inquiry message to evaluate the fluency of the human-computer interaction, and the business service evaluation and recommendation module uses the conversation evaluation language model to identify emotional words/emotional sentences in the human-computer interaction answer message to evaluate the customer's emotion in the human-computer interaction. A system for implementing human-computer interaction and evaluation and suggestion through artificial intelligence as described in claim 1, wherein the business service evaluation and suggestion module uses the dialogue evaluation language model to identify the basic customer information in the human-computer interaction answer message to implement data collection and evaluation of the human-computer interaction inquiry message, and the business service evaluation and suggestion module uses the dialogue evaluation language model to compare the product and service introduction description in the human-computer interaction inquiry message with the product and service introduction template to implement professional evaluation of the product and service introduction in the human-computer interaction inquiry message. As described in claim 1, the system for realizing human-computer interaction and evaluation suggestions through artificial intelligence, wherein the observer artificial intelligence platform further includes providing the business service evaluation and the business service improvement suggestions to a supervisory terminal device when the number of evaluation items in the business service evaluation that are lower than the evaluation threshold is greater than or equal to the quantity threshold. A system for implementing human-computer interaction and evaluation suggestions through artificial intelligence as described in claim 1, wherein the virtual customer artificial intelligence platform further includes a first modal conversion module for converting the modality of the human-computer interaction inquiry message and the modality of the human-computer interaction answer message; and the observer artificial intelligence platform further includes a second modal conversion module for converting the modality of the human-computer interaction inquiry message and the modality of the business service evaluation and the business service improvement suggestion. A method for implementing human-computer interaction and evaluation recommendations through artificial intelligence comprises the following steps: a salesperson terminal device interfaces with a virtual customer artificial intelligence platform and an observer artificial intelligence platform; when the virtual customer artificial intelligence platform interfaces with the salesperson terminal device, the virtual customer artificial intelligence platform randomly generates virtual customer basic information and virtual customer personality information; the virtual customer artificial intelligence platform selects a corresponding customer language model based on the virtual customer personality information, and the customer language model is trained using training data corresponding to the virtual customer personality information; the salesperson terminal device provides human-computer interaction messages to the virtual customer artificial intelligence platform; The virtual customer artificial intelligence platform uses speech-to-text technology to convert the human-machine interaction message into a human-machine interaction inquiry message; the virtual customer artificial intelligence platform uses the selected customer language model to generate a human-machine interaction answer message from the human-machine interaction inquiry message; the virtual customer artificial intelligence platform provides the human-machine interaction answer message to the salesperson terminal device; the observer artificial intelligence platform obtains the corresponding human-machine interaction message and the human-machine interaction answer message from the salesperson terminal device; the observer artificial intelligence platform uses speech-to-text technology to convert the human-machine interaction voice message into the human-machine interaction inquiry message; The observer artificial intelligence platform uses a dialogue evaluation language model to evaluate the human-machine interaction inquiry message and the human-machine interaction answer message in terms of human-machine interaction fluency, customer sentiment in human-machine interaction, emergency response in human-machine interaction, data collection of the human-machine interaction inquiry message, semantic accuracy of the human-machine interaction inquiry message, and professionalism of product and service introductions in the human-machine interaction inquiry message, thereby generating a business service evaluation based on corresponding evaluation items and generating a business service improvement suggestion based on the evaluation process and the evaluation items. The observer artificial intelligence platform provides the business service evaluation and the business service improvement suggestion to the salesperson terminal device. A method for implementing human-computer interaction and evaluation suggestions through artificial intelligence as described in claim 6, wherein the observer artificial intelligence platform uses the dialogue evaluation language model to identify the number of interjections in the human-computer interaction inquiry message to evaluate the fluency of the human-computer interaction, and the observer artificial intelligence platform uses the dialogue evaluation language model to identify emotional words/emotional sentences in the human-computer interaction answer message to evaluate the customer's emotions in the human-computer interaction. A method for implementing human-computer interaction and evaluation suggestions through artificial intelligence as described in claim 6, wherein the observer artificial intelligence platform uses the dialogue evaluation language model to identify basic customer information in the human-computer interaction inquiry message to implement data collection and evaluation of the human-computer interaction inquiry message, and the observer artificial intelligence platform uses the dialogue evaluation language model to compare the product and service introduction description in the human-computer interaction inquiry message with a product and service introduction template to implement a professional evaluation of the product and service introduction in the human-computer interaction inquiry message. As described in claim 6, the method for realizing human-computer interaction and evaluation suggestions through artificial intelligence, wherein the observer artificial intelligence platform further includes the step of providing the business service evaluation and the business service improvement suggestions to a competent terminal device when the number of evaluation items in the business service evaluation that are lower than the evaluation threshold is greater than or equal to the quantity threshold. The method for implementing human-computer interaction and evaluation suggestions through artificial intelligence as described in claim 6, wherein the method for implementing human-computer interaction and evaluation suggestions through artificial intelligence further includes the following steps: the virtual customer artificial intelligence platform converts the mode of the human-computer interaction inquiry message; the virtual customer artificial intelligence platform converts the mode of the human-computer interaction answer message; the observer artificial intelligence platform converts the mode of the human-computer interaction inquiry message; and the observer artificial intelligence platform converts the mode of the business service evaluation and the mode of the business service improvement suggestion.