Conversation mediation system and related method

Abstract

The invention provides a dialogue intermediary system. The dialogue intermediary system comprises an intermediary input generation module, a context management module and a response summarization module. The intermediary input generation module is configured to generate an intermediary input based on a user input and auxiliary information related to the user input, and transmit the intermediary input to the at least one dialog application accordingly. The context management module is configured to extract the auxiliary information from user data and update the user data based on a response to the intermediary input generated by the at least one dialog application. The response summarizing module is configured to summarize and summarize responses generated by the at least one dialog application to the intermediary input to generate a comprehensive response.

Description

Technical Field

[0001] The present invention relates to a dialogue application program. More specifically, it relates to a dialogue mediation system and related methods for interacting with a dialogue application program to enhance the user experience and context response ability. Background Art

[0002] Dialogue application programs based on Artificial Intelligence (AI) (such as chatbots, dialogue agents, and dialogue systems, especially those driven by large language models such as ChatGPT) have become increasingly popular as tools for providing information and solutions to users. Despite their increasing adoption rate, these AI-based dialogue application programs have some limitations. First, AI-based dialogue application programs have a limited knowledge base and can only provide information based on the data they were pre-trained on, which may not be comprehensive or up-to-date. Second, AI-based dialogue application programs typically have limited interaction memory. This means that AI-based dialogue application programs lack the ability to remember long or complex conversation histories, which may be necessary for providing context-sensitive responses. Third, AI-based dialogue application programs typically do not have access to personal user preferences, local information, or the context of user actions. This means that AI-based dialogue application programs lack user data. Given this, users may find that the information and solutions provided by AI-based dialogue application programs are general rather than tailored to their specific needs or context. Therefore, it is necessary to provide a complete solution to improve the above problems. Summary of the Invention

[0003] An object of the present invention is to provide a dialogue mediation system and related methods for aggregating and summarizing conversations between a user and multiple dialogue application programs. The dialogue mediation system and related methods serve as an intermediate layer between the user and multiple dialogue application programs. Specifically, the dialogue mediation system and methods collect user input and context information, query multiple sources accordingly to obtain relevant data, and then aggregate and summarize the responses from multiple sources into a comprehensive response for the user. In addition, the comprehensive response is also stored as user-specific context for future interactions, thereby enhancing personalization and efficiency. By doing so, the dialogue application program can provide a more comprehensive and context-aware response to the user. This method not only enhances the accuracy and completeness of the information provided by the dialogue application program, but also enhances the user experience by providing personalized and context-aware responses to the user.

[0004] According to one embodiment, a dialogue mediation system is provided. The dialogue mediation system includes: a mediation input generation module, a context management module, and a response summarization module. The mediation input generation module is configured to generate a mediation input based on a user input and auxiliary information related to the user input, and accordingly send the mediation input to at least one dialogue application. The context management module is configured to extract the auxiliary information from user data and update the user data based on a response to the mediation input generated by the at least one dialogue application. The response summarization module is configured to summarize and aggregate the responses to the mediation input generated by the at least one dialogue application to generate a comprehensive response.

[0005] In some embodiments, the user data includes personal information, behavioral information, preference information, and interaction history information.

[0006] In some embodiments, the mediation input generation module is further configured to send the mediation input to at least one information retrieval system, and the response summarization module is configured to summarize and aggregate the responses to the mediation input generated by the at least one dialogue application and the at least one information retrieval system to generate the comprehensive response.

[0007] In some embodiments, the at least one information retrieval system is a search engine.

[0008] In some embodiments, the mediation input generation module is further configured to send the user input to at least one information retrieval system, and accordingly generate the mediation input based on the response of the at least one information retrieval system, the user input, and the auxiliary information.

[0009] In some embodiments, the at least one dialogue application operates based on a language model.

[0010] In some embodiments, the context management module is further configured to synchronously update the user data among multiple user devices.

[0011] According to one embodiment, a dialogue mediation method is provided. The method includes: extracting auxiliary information related to the user input from user data; generating a mediation input based on the user input and the auxiliary information; sending the mediation input to at least one dialogue application; summarizing and aggregating the responses to the mediation input generated by the at least one dialogue application to generate a comprehensive response; and updating the user data based on the response to the mediation input generated by the at least one dialogue application.

[0012] In some embodiments, the user data includes personal information, behavioral information, preference information, and interaction history information.

[0013] In some embodiments, the dialogue mediation method further includes: sending the mediation input to at least one information retrieval system; and summarizing and aggregating responses to the mediation input from the at least one dialogue application and the at least one information retrieval system to generate the comprehensive response.

[0014] In some embodiments, the at least one information retrieval system is a search engine.

[0015] In some embodiments, the dialogue mediation method further includes: sending the user input to at least one information retrieval system; and generating the mediation input based on the response of the at least one information retrieval system, the user input, and the auxiliary information.

[0016] In some embodiments, the at least one dialogue application operates based on a language model.

[0017] In some embodiments, the dialogue mediation method further includes: synchronously updating the user data among multiple user devices.

[0018] Those skilled in the art can undoubtedly understand these and other objectives of the present invention after reading the following detailed description of the preferred embodiments shown in the drawings. The detailed description will be given in the following embodiments with reference to the drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The present invention can be more comprehensively understood by reading the subsequent detailed description and referring to the examples given in the drawings.

[0020] Figure 1 The schematic diagram of the architecture of a dialogue mediation system according to an embodiment of the present invention is shown.

[0021] Figure 2 The schematic diagram of the interaction process executed by a dialogue mediation system according to the first embodiment of the present invention is shown.

[0022] Figure 3 The schematic diagram of the interaction process executed by a dialogue mediation system according to the second embodiment of the present invention is shown.

[0023] Figure 4 The schematic diagram of the interaction process executed by a dialogue mediation system according to the third embodiment of the present invention is shown.

[0024] Figure 5 The schematic diagram of a dialogue mediation method according to an embodiment of the present invention is shown.

[0025] In the following detailed description, for purposes of illustration, numerous specific details are set forth in order for those skilled in the art to better understand embodiments of the present invention. However, it is apparent that one or more embodiments may be practiced without these specific details, and different embodiments may be combined according to requirements and should not be limited to the embodiments listed in the figures. Detailed Embodiments

[0026] The following description is a preferred embodiment for implementing the present invention, which is only used to illustrate the technical features of the present invention and not to limit the scope of the present invention. Throughout the specification and claims, certain terms are used to refer to specific elements. Those skilled in the art should understand that manufacturers may use different names to refer to the same element. Therefore, the specification and claims do not use the difference in names as a way to distinguish elements, but use the difference in the functions of elements as the basis for distinction. The terms "element", "system" and "device" used in the present invention may be entities related to a computer, where the computer may be hardware, software, or a combination of hardware and software. The terms "comprising" and "including" mentioned in the following description and claims are open-ended terms and should be interpreted as meaning "including, but not limited to...". In addition, the term "coupled" means an indirect or direct electrical connection. Therefore, if a device is described as being coupled to another device in the text, it means that the device can be directly electrically connected to the other device, or indirectly electrically connected to the other device through other devices or connection means.

[0027] In this specification, the reference to "an embodiment" or "an embodiment (an example)" means that the specific features, structures, or characteristics described in relation to the embodiment or example are included in at least one embodiment of the embodiments of the present invention. Therefore, the phrase "in an embodiment" or "in an embodiment" that appears throughout this specification does not necessarily refer to the same embodiment. In addition, the specific features, structures, or characteristics may be combined in any suitable combination and / or sub-combination in one or more embodiments.

[0028] Figure 1 A schematic diagram showing the architecture of a dialogue mediation system according to an embodiment of the present invention is shown. As shown in the figure, the dialogue mediation system 100 may receive user input from one of the user devices 50_1 to 50_O belonging to the same user 50. In particular, the dialogue mediation system 100 may be deployed on at least one of the user devices 50_1 to 50_O, or implemented as a cloud service that can be accessed by the user devices 50_1 to 50_O. In addition, user input may be obtained from any one of the user devices 50_1 to 50_O through various human-computer interaction modes, including but not limited to the user's keyboard input and voice commands.

[0029] The user input can be a user query, a user question, a user prompt, and / or a user request. The dialogue mediation system 100 can be implemented by a rule-based method or an artificial intelligence (AI)-based method. As an intermediary, the dialogue mediation system 100 converts the user input into a mediator input to interact with multiple conversational applications 200_1 to 200_N and / or multiple information retrieval systems 300_1 to 300_M, thereby improving the efficiency of the interaction.

[0030] The conversational applications 200_1 to 200_N include any software or system capable of having a conversation or chat with the user 50. The conversational applications 200_1 to 200_N can include, but are not limited to, applications developed using methods based on rules, templates, retrieval, decision trees, language models, or any combination thereof. The conversational applications 200_1 to 200_N utilize datasets and algorithms to understand, generate, and process natural language text in a conversational manner. This approach allows the conversational applications 200_1 to 200_N to participate in dynamic and coherent conversations, understand user input, and generate contextually relevant responses. In addition, the information retrieval systems 300_1 to 300_M include any software or system capable of performing information retrieval tasks. This includes, but is not limited to, search engines, database management systems, information retrieval platforms, etc. The information retrieval systems 300_1 to 300_M are capable of parsing and analyzing user queries to extract relevant information from databases and using algorithms to ensure the accuracy and relevance of the retrieved information.

[0031] The dialogue mediation system 100 can parse and process natural language user input using predefined rules. In some embodiments, the dialogue mediation system 100 can integrate artificial intelligence technologies to more precisely understand and process user input and generate more accurate and relevant mediator input. In some embodiments, the mediator input generated by the dialogue mediation system is not limited to natural language expressions and can also include non-natural language symbols, codes, or other forms of expressions to meet the interaction rules or preferences of the conversational applications 200_1 to 200_N.

[0032] In some embodiments, the dialogue mediation system 100 can customize the mediation input by identifying / distinguishing the inherent characteristics and capabilities of the dialogue applications 200_1 to 200_N to elicit the best response. This means that the dialogue mediation system 100 can generate multiple mediation inputs for the same user input, so that each mediation input is closely combined with the inherent characteristics and functions of the dialogue applications 200_1 to 200_N.

[0033] The dialogue mediation system 100 includes a mediator input generating module 110, a context management module 120, and a response aggregation module 130. Whenever the dialogue mediation system 100 receives a user input, the context management module 120 searches for user data stored on one or more user devices 50_1 to 50_O, and extracts relevant auxiliary information to ensure that the dialogue applications 200_1 to 200_N and / or the information retrieval systems 300_1 to 300_M have a more comprehensive understanding of the user's intention.

[0034] Generally, the user data includes various information to describe user attributes and preferences. For example, it includes personal information such as age, gender, occupation, educational background, and / or place of residence. In one embodiment, the user data also includes real-world behavior and / or online behavior information, which includes movement trajectories, locations, browsing histories, search histories, purchase patterns, and / or activities on social media platforms such as likes, shares, and comments to depict the user's interests. In one embodiment, the user data can also include preference information, which includes user-specific preference settings such as language preferences. For the dialogue interaction or communication with the dialogue applications 200_1 to 200_N, the user data includes interaction history information, which includes historical conversations (e.g., questions and responses) with the dialogue applications 200_1 to 200_N. That is to say, the user data enables the dialogue applications 200_1 to 200_N to generate more personalized, accurate, and context-aware responses based on a deep understanding of the user's needs and preferences.

[0035] For example, if the user input is an inquiry about surrounding restaurants or a suggestion about popular dating locations, the context management module 120 can search the user data to retrieve auxiliary information such as the user's age, gender, place of residence, favorite food, etc. For example, if the user input is an inquiry about dinner suggestions, the context management module 120 can search the user data to retrieve auxiliary information about the user's past activity records. This may include what the user had for breakfast and lunch, the user's current location, and the user's favorite food. Such auxiliary information allows the dialogue applications 200_1 to 200_N and the multiple information retrieval systems 300_1 to 300_M to customize their responses in a more precise manner.

[0036] Based on the original user input and the auxiliary information extracted from the user data, the intermediary input generation module 110 generates one or more intermediary inputs. The dialogue mediation system 100 sends the one or more intermediary inputs to the dialogue applications 200_1 to 200_N and the information retrieval systems 300_1 to 300_M respectively. According to the one or more intermediary inputs, the dialogue applications 200_1 to 200_N and the information retrieval systems 300_1 to 300_M output corresponding responses. The one or more intermediary inputs ensure that the responses generated by the dialogue applications 200_1 to 200_N and the information retrieval systems 300_1 to 300_M are fine-tuned to adapt to the user input and user data, thus achieving a highly personalized and effective user experience. In addition, in some embodiments, the dialogue mediation system 100 first sends the user input to the information retrieval systems 300_1 to 300_M, and then generates one or more intermediary inputs based on the information provided by the information retrieval systems 300_1 to 300_M, the user input, and the auxiliary information. This is because the information provided by the information retrieval systems 300_1 to 300_M may be more up-to-date.

[0037] The responses from the dialogue applications 200_1 to 200_N and the information retrieval systems 300_1 to 300_M will be summarized and aggregated by the response summarization module 130. The response summarization module 130 accordingly generates a consolidated response, which will be easier for the user to understand. In some embodiments, the dialogue mediation system 100 can integrate AI technology, which can significantly improve the proficiency in analyzing and summarizing the responses from the dialogue applications 200_1 to 200_N and the information retrieval systems 300_1 to 300_M, enabling the dialogue mediation system 100 to provide more detailed and tailored interactions, thereby enhancing the user's overall experience and the value obtained from the dialogue applications 200_1 to 200_N and the information retrieval systems 300_1 to 300_M.

[0038] On the other hand, the context management module 120 can update user data by using the comprehensive response generated by the response summarization module 130 or the original responses directly from the conversation applications 200_1 to 200_N and the information retrieval systems 300_1 to 300_M. For example, the context management module 120 can use the comprehensive response or the original response to update the interaction history information in the user data. In one embodiment, the context management module 120 can synchronize the updated user data among the user devices 50_1 to 50_O.

[0039] Figure 2 FIG. illustrates a schematic diagram of an interaction process executed by the dialogue mediation system 100 according to the first embodiment of the present invention. In step S111, the dialogue mediation system 100 receives a user input D1. In step S112, the dialogue mediation system 100 searches the user data to extract auxiliary information D2 related to the user input D1. In step S113, the dialogue mediation system 100 generates a mediation input D3 based on the user input D1 and the auxiliary information D2. In steps S114, S115, and S116, the dialogue mediation system 100 sends the mediation input D3 to the conversation applications A and B, and the information retrieval system C. In step S117, the dialogue mediation system 100 receives responses D4 to the mediation input D3 generated by the conversation applications A and B and the information retrieval system C. In step S118, the dialogue mediation system 100 summarizes and aggregates the responses D4, and accordingly generates a comprehensive response D5 for the user. In addition, the information contained in the comprehensive response D5 will be saved as part of the user data (understandably, updating the user data).

[0040] Figure 3 FIG. illustrates a schematic diagram of an interaction process executed by the dialogue mediation system 100 according to the second embodiment of the present invention. In step S211, the dialogue mediation system 100 receives a user input D1. In step S212, the dialogue mediation system 100 searches the user data to extract auxiliary information D2 related to the user input D1. In step S213, the dialogue mediation system 100 generates a mediation input D3 based on the user input D1 and the auxiliary information D2. In step S214, the dialogue mediation system 100 sends the mediation input D3 to the conversation application. In step S215, the dialogue mediation system 100 receives a response D4 to the mediation input D3 generated by the conversation application. In step S216, the dialogue mediation system 100 summarizes and aggregates the response D4, and accordingly generates a comprehensive response D5 for the user. Correspondingly, the information contained in the comprehensive response D5 will be saved as part of the user data (understandably, updating the user data).

[0041] Figure 4Schematic diagram of an interaction process executed by the dialogue mediation system 100 according to the third embodiment of the present invention. At step S311, the dialogue mediation system 100 receives the user input D1. At step S312, the dialogue mediation system 100 searches the user data to extract auxiliary information D2 related to the user input D1. At step S313, the dialogue mediation system 100 generates a mediation input D3 based on the user input D1 and the auxiliary information D2. At step S314, the dialogue mediation system 100 sends the mediation input D3 to the dialogue application. At step S315, the dialogue mediation system 100 receives a response D4 to the mediation input D3 generated by the dialogue application. At step S316, the dialogue mediation system 100 summarizes and aggregates the response D4 and accordingly generates a comprehensive response D5 for the user. Correspondingly, the information contained in the comprehensive response D5 will be saved as part of the user data (understandably, updating the user data). At step S317, the update of the user data will be synchronized among all user devices 50_1 to 50_O belonging to the user 50.

[0042] Figure 5 Schematic diagram of a dialogue mediation method according to an embodiment of the present invention. As Figure 5 shown, the method of the present invention includes the following simplified process:

[0043] Step S410: Receive user input;

[0044] Step S420: Extract auxiliary information related to the user input from the user data;

[0045] Step S430: Generate a mediation input based on the user input and the auxiliary information;

[0046] Step S440: Send the mediation input to at least one dialogue application;

[0047] Step S450: Summarize and aggregate the responses to the mediation input generated by at least one dialogue application to generate a comprehensive response; and

[0048] Step S460: Update the user data based on the responses to the mediation input generated by at least one dialogue application.

[0049] Since the principles and specific details of the above steps have been described in detail through the above embodiments, they will not be repeated here. It should be noted that the above process can also be improved by adding other additional steps or making appropriate modifications and adjustments to better improve flexibility and further improve the efficiency of using the dialogue application.

[0050] In summary, the dialogue mediation system and related methods of the present invention act as mediators to convert user input into mediator input to facilitate intelligent interaction with various dialogue applications. Thanks to the auxiliary information extracted from user data, the mediator input generated by the dialogue mediation system is not just a simple response to the words expressed by the user, but a fusion of user intent, context awareness (situational awareness), and refined understanding. Therefore, the dialogue mediation system and method of the present invention significantly ensure a more coherent, relevant, and user-friendly interaction experience, improving the quality and efficiency of the interaction between the user and the dialogue application.

[0051] According to current embodiments, the embodiments can be implemented as a device, a method, or a computer program product. Thus, the current embodiments can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects, which can generally be referred to herein as a "module" or a "system". Additionally, the current embodiments can take the form of a computer program product embodied in any tangible expression medium having computer-usable program code. In terms of hardware, the present invention can be implemented by applying the following technologies or related combinations: individual operational logics of logic gates capable of performing logical functions according to data signals, and appropriate combinational logics of application specific integrated circuits (ASICs), programmable gate arrays (PGAs), or field programmable gate arrays (FPGAs).

[0052] Flowcharts and block diagrams illustrate the possible architectures, functions, and operations of systems, methods, and computer program products according to various embodiments of the present embodiment. In this regard, each block in the flowchart or block diagram can represent a module, a segment, or a part of code that includes one or more executable instructions for implementing the specified logical function. It should also be noted that each module in the block diagram and / or flowchart, as well as combinations of each module in the block diagram and / or flowchart, can be implemented by a dedicated hardware system that performs the specified function or action, or by a combination of dedicated hardware and computer instructions. These computer program instructions can be stored in a computer-readable medium that instructs a computer or other programmable data processing device to operate in a specific manner, such that the instructions stored in the computer-readable medium produce a manufactured article that includes an instruction apparatus for implementing the functions / actions specified in the flowchart and / or block diagram blocks.

[0053] Although the present invention has been described by way of examples and in terms of preferred embodiments, it should be understood that the invention is not limited to the disclosed embodiments. On the contrary, it is intended to cover various modifications and similar constructions (as would be apparent to those skilled in the art), e.g., combinations or substitutions of different features in different embodiments. Accordingly, the scope of the appended claims should be given the broadest interpretation to cover all such modifications and similar constructions.

Claims

1. A conversation mediation system, comprising: an intermediary input generation module, configured to generate an intermediary input based on a user input and auxiliary information related to the user input, and send the intermediary input to at least one dialog application accordingly; a context management module configured to extract the auxiliary information from the user data and update the user data based on a response to the mediator input generated by the at least one dialog application; A response aggregation module is configured to summarize and aggregate responses to the mediator input generated by the at least one dialog application to generate a composite response.

2. The conversation mediation system according to claim 1, wherein: The user data includes personal information, behavior information, preference information and interaction history information.

3. The conversation mediation system according to claim 1, wherein: The mediated input generation module is further configured to send the mediated input to at least one information retrieval system, and the response aggregation module is configured to summarize and aggregate responses to the mediated input generated by the at least one dialog application and the at least one information retrieval system to generate the composite response.

4. The conversation mediation system according to claim 3, wherein: The at least one information retrieval system is a search engine.

5. The conversation mediation system according to claim 1, wherein: The intermediate input generation module is further configured to send the user input to at least one information retrieval system, and accordingly generate the intermediate input based on a response of the at least one information retrieval system, the user input and the auxiliary information.

6. The conversation mediation system according to claim 1, wherein: The at least one conversational application operates based on a language model.

7. The conversation mediation system according to claim 1, wherein: The context management module is further configured to synchronously update the user data among multiple user devices.

8. A conversation mediation method, comprising: Receive user input; Extracting auxiliary information related to the user input from the user data; generating an intermediate input based on the user input and the auxiliary information; sending the mediated input to at least one conversation application; summarizing and aggregating responses to the mediator input generated by the at least one dialog application to generate a composite response; The user data is updated based on a response to the broker input generated by the at least one dialog application.

9. The conversation mediation method according to claim 8, wherein: The user data includes personal information, behavior information, preference information and interaction history information.

10. The conversation mediation method according to claim 8, further comprising: sending the mediated input to at least one information retrieval system; and Responses to the mediator input from the at least one dialog application and the at least one information retrieval system are summarized and aggregated to generate the composite response.

11. The conversation mediation method according to claim 10, wherein: The at least one information retrieval system is a search engine.

12. The conversation mediation method according to claim 8, further comprising: sending the user input to at least one information retrieval system; and The intermediary input is generated based on a response of the at least one information retrieval system, the user input, and the auxiliary information.

13. The conversation mediation method according to claim 8, wherein: The at least one conversational application operates based on a language model.

14. The conversation mediation method according to claim 8, further comprising: The user data is updated synchronously among multiple user devices.

Images