Human machine interaction method based on conversation history, device, apparatus and storage medium

Abstract

To provide a human machine interaction method based on a conversation history for enhancing response accuracy and efficiency of a human machine interaction, a device, an apparatus and a storage medium.SOLUTION: A human machine interaction method acquires user's current conversation information and a conversation history database. The conversation history database includes a plurality of history storage objects, the history storage objects represent user's conversation history information and analysis results of the conversation history information, and the analysis results include time stamps and word meaning information of the conversation history information. Also, the method determines a history storage object associated with the current conversation information as a target storage object from the conversation history database, and determines response information of the current conversation information according to the target storage object.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present disclosure relates to the field of human-machine interaction in the field of artificial intelligence, and in particular to a method, apparatus, device and storage medium for human-machine interaction based on conversation history. [Background technology]

[0002] With the development of artificial intelligence technology, users' demands for interaction experiences during human-machine interaction are increasing. Users can input information as needed and automatically receive responses to that information.

[0003] Users expect accurate response information during human-machine interaction, so how to provide efficient and accurate responses to users has become an urgent issue. Summary of the Invention [Problem to be solved by the invention]

[0004] The present disclosure provides a method, apparatus, device, and storage medium for human-machine interaction based on conversation history. [Means for solving the problem]

[0005] According to a first aspect of the present disclosure, there is provided a human-machine interaction method based on a conversation history, the human-machine interaction method comprising: acquiring a user's current conversation information and a conversation history database, the conversation history database including a plurality of history storage objects, the history storage objects representing the user's conversation history information and an analysis result of the conversation history information, the analysis result including a timestamp and semantic information of the conversation history information; determining a history storage object associated with the current conversation information from the conversation history database as a target storage object; determining response information for the current conversation information in response to the target storage object.

[0006] According to a second aspect of the present disclosure, there is provided a human-machine interaction device based on a conversation history, the human-machine interaction device comprising: an acquisition unit used to acquire a user's current conversation information and a conversation history database, the conversation history database including a plurality of history storage objects, the history storage objects representing the user's conversation history information and an analysis result of the conversation history information, the analysis result including timestamps and semantic information of the conversation history information; a target determination unit, used for determining a history storage object associated with the current conversation information from the conversation history database as a target storage object; a response determination unit, used for determining response information of the current conversation information according to the target storage object;

[0007] According to a third aspect of the present disclosure, there is provided an electronic device, the electronic device comprising: at least one processor; a memory communicatively coupled to the at least one processor; The memory stores instructions executable by the at least one processor, and when executed by the at least one processor, the instructions enable the at least one processor to perform the method according to the first aspect of the present disclosure.

[0008] According to a fourth aspect of the present disclosure, there is provided a non-transitory computer-readable storage medium having stored thereon computer instructions, said computer instructions being used to cause a computer to perform a method according to the first aspect of the present disclosure.

[0009] According to a fifth aspect of the present disclosure, there is provided a computer program which, when executed by a processor, implements the steps of the method according to the first aspect of the present disclosure. [Effects of the Invention]

[0010] The techniques disclosed herein improve the efficiency and accuracy of human-machine interaction.

[0011] It should be noted that the contents described in this section are not intended to identify key or important features of the embodiments of the present disclosure, nor are they intended to limit the scope of the present disclosure. Other features of the present disclosure will be readily apparent from the following detailed description. [Brief explanation of the drawings]

[0012] The drawings are used for better understanding of the present technical solution and are not intended to limit the present disclosure. [Figure 1] 1 is a flowchart of a human-machine interaction method based on conversation history provided according to an embodiment of the present disclosure. [Figure 2] 1 is a flowchart of a human-machine interaction method based on conversation history provided according to an embodiment of the present disclosure. [Figure 3] 1 is a flowchart of a human-machine interaction method based on conversation history provided according to an embodiment of the present disclosure. [Figure 4] FIG. 1 is a structural block diagram of a human-machine interaction device based on conversation history provided according to an embodiment of the present disclosure; [Figure 5] FIG. 1 is a structural block diagram of a human-machine interaction device based on conversation history provided according to an embodiment of the present disclosure; [Figure 6] FIG. 1 is a block diagram of an electronic device for implementing a human-machine interaction method based on a conversation history according to an embodiment of the present disclosure. [Figure 7] FIG. 1 is a block diagram of an electronic device for implementing a human-machine interaction method based on a conversation history according to an embodiment of the present disclosure. DETAILED DESCRIPTION OF THE INVENTION

[0013]

[0023] Hereinafter, exemplary embodiments of the present disclosure will be described with reference to the drawings. For ease of understanding, various details of the embodiments of the present disclosure are included therein and should be considered merely exemplary. Therefore, those skilled in the art should recognize that various changes and modifications can be made to the embodiments described in the detailed description without departing from the scope and spirit of the present disclosure. Similarly, for the sake of clarity and conciseness, the following description will omit descriptions of well-known functions and structures.

[0014] Human-machine interaction refers to processing conversational information input by a user and sending corresponding response information to the user. For example, the conversational information input by the user can be processed using a large language model (LLM). To return more accurate information to the user, the response information is usually combined with the user's conversation history information to make it more responsive to the user's actual needs.

[0015] However, user conversation history information is typically long, and models like LLM are unable to effectively analyze conversation history information due to input length limitations. Currently, methods such as truncation and segmentation are considered when dealing with long conversations. Text truncation refers to directly cutting out a portion of the conversation history information as input, while segmented input refers to splitting the conversation history information of a long conversation into multiple short sequences and inputting them into the model separately. However, text truncation and segmented input result in information loss, which affects the efficiency and accuracy of response information determination, reducing the efficiency and accuracy of human-machine interaction and affecting the user's interaction experience.

[0016] The present disclosure provides a human-machine interaction method, device, equipment and storage medium based on conversation history applied to the field of human-machine interaction in the field of artificial intelligence, so as to increase the efficiency and accuracy of human-machine interaction and improve user experience.

[0017] Note that the model in this example is not a model for a specific user and cannot reflect the personal information of a specific user. Note that the data in this example is from an open dataset.

[0018] In the technical solution disclosed herein, the collection, storage, use, processing, transmission, provision and disclosure of personal information of users involved shall all comply with the provisions of relevant laws and regulations and shall not violate public order and morals.

[0019] To help readers better understand the implementation principles of the present disclosure, the following examples are described in more detail in conjunction with FIGS. 1-7.

[0020] 1 is a flowchart of a method for human-machine interaction based on a conversation history provided according to an embodiment of the present disclosure, which can be performed by a human-machine interaction device based on a conversation history. As shown in FIG. 1, the method includes steps S101 to S103.

[0021] In S101, the user's current conversation information and conversation history database are obtained, and the conversation history database includes multiple history storage objects, which represent the user's conversation history information and the analysis results of the conversation history information, and the analysis results include timestamps and semantic information of the conversation history information.

[0022] For example, when a user engages in human-machine interaction, the user can input conversation information in real time, and the user can input the conversation information by text, voice, etc. For example, the user can input their own question and receive an answer to the question through human-machine interaction. The conversation information currently input by the user can be obtained in real time, i.e., current conversation information can be obtained. Each time a user engages in human-machine interaction, the conversation information can be stored as new conversation history information, and the conversation history information can be stored in a conversation history database. The conversation information when a user engages in human-machine interaction can include conversation information input by the user and conversation information replied by the robot. New conversation history information can be obtained in real time or at a scheduled time to update the conversation history database.

[0023] A conversation history database can be acquired simultaneously with the acquisition of a user's current conversation information. The conversation history database can include multiple history storage objects, each representing a sentence or a section of conversation history information and an analysis result of the conversation history information. The analysis result can include information such as a timestamp and semantic information of the conversation history information. For example, conversation information uttered per sentence is treated as one conversation history information, and each conversation history information corresponds to one history storage object, and the history storage object can represent the text content of the conversation history information. After one conversation history information is acquired, the transmission time of the conversation history information can be identified and stored as a timestamp in the history storage object of the conversation history information. Furthermore, a semantic analysis of the conversation history information can be performed, and semantic information such as emotions, themes, and keywords expressed in the conversation history information obtained thereby can be stored in the history storage object of the conversation history information. The history storage object may also be referred to as an IMO (Interactive Memory Object). In this embodiment, the method of handling the semantic analysis is not particularly limited.

[0024] The conversation history database may include a data table, with each row representing an IMO and columns representing fields stored in the IMO. For example, the fields may include the text content of the conversation history information, a timestamp, a heading, keywords, a theme category, a sentiment tag, and the like. The heading may summarize the content of the conversation history information, the keywords may be important phrases in the text content of the conversation history information, the theme category may be a theme or field related to the conversation history information, such as a transportation category or a food category, and the sentiment tag may represent the sentiment expressed by the conversation history information, such as a positive sentiment or a negative sentiment. Each time an IMO is obtained, the IMO is stored in the conversation history database.

[0025] This embodiment further includes the steps of identifying the transmission time of the current conversation information and determining the transmission time as the timestamp of the current conversation information, performing a semantic analysis process on the current conversation information and obtaining semantic information of the current conversation information, determining the timestamp of the current conversation information and the semantic information of the current conversation information as the analysis result of the current conversation information, and determining the current conversation information and the analysis result of the current conversation information as a new history storage object and storing it in the conversation history database.

[0026] Specifically, after the current conversation information is acquired, the current conversation information can be analyzed to obtain an analysis result of the current conversation information, and the text content of the current conversation information and the analysis result are stored in the conversation history database as a new history storage object. The analysis result can include information such as a timestamp and semantic information of the conversation information. Identifying the transmission time of the current conversation information may be, for example, identifying the time when the conversation information is transmitted from the user or the time when the conversation information is transmitted from the robot. Setting the transmission time of the current conversation information as the timestamp of the current conversation information means attaching the transmission time of the conversation information to each piece of conversation information.

[0027] Furthermore, a semantic analysis algorithm can be preset to perform semantic analysis processing on the current conversation information to obtain semantic information of the current conversation information, where the semantic information can include headings, keywords, topic categories, sentiment tags, etc. The timestamp of the current conversation information and the semantic information of the current conversation information are determined as the analysis result of the current conversation information. The current conversation information and the analysis result of the current conversation information are determined as a new IMO and stored in the conversation history database. The IMO can be compressed or encrypted to save storage space and ensure data security.

[0028] The beneficial effect of this setup is that when a user engages in human-machine interaction, a new history storage object can be generated according to the content of the conversation, and the conversation history database can be automatically updated, thereby improving the accuracy and efficiency of determining response information and improving the user experience.

[0029] In this embodiment, the method further includes a step of identifying the conversation information immediately before the current conversation information in the conversation history database, and a step of assigning a predetermined logical identifier to the current conversation information and the conversation information immediately before the current conversation information, indicating that there is a logical structure and chronological order between the two conversation information.

[0030] Specifically, a relation can be established between different IMOs so that the logical structure and chronological order of the conversation are reflected. For example, a previous sentence can be related to a next sentence. That is, for each current conversation, the conversation immediately preceding the current conversation can be identified. A single predefined logical identifier that can indicate that there is a logical structure and chronological order between the two conversations is assigned to the two conversations.

[0031] In the conversation history database, a preset logical identifier is added for these two conversation information, and when searching for one of the conversation information, the other conversation information that is logically related to the conversation information can be found according to the logical identifier.

[0032] The beneficial effect of such an arrangement is that by adding logical identifiers, association relationships between different IMOs can be identified, facilitating subsequent searches and queries, thereby improving the efficiency and accuracy of human-machine interaction.

[0033] In this embodiment, the method further includes the steps of acquiring a timestamp of a history storage object in the conversation history database according to a predetermined information update period, and deleting the history storage object from the conversation history database when it is determined that the length of storage time of the history storage object in the conversation history database has exceeded a predetermined threshold value for the length of time according to the timestamp of the history storage object.

[0034] Specifically, the conversation history database can be updated periodically. An information update cycle can be preset, for example, 24 hours. According to the preset information update cycle, timestamps of all history storage objects in the conversation history database are obtained, and the timestamps can represent the transmission times of the conversation history information of the history storage objects. For example, the transmission times of each conversation history information are obtained every 24 hours.

[0035] The current time is identified, and it is determined whether the length of time the history storage object has been stored in the conversation history database exceeds a preset threshold value according to the current time and the timestamp of the history storage object. For example, the preset threshold value is one year, and it is determined whether the time difference between the transmission time of the conversation history information and the current time exceeds one year. If it is determined that the length of time the history storage object has been stored in the conversation history database exceeds the preset threshold value, the history storage object can be deleted from the conversation history database, but if the preset threshold value is not exceeded, the history storage object is retained in the conversation history database.

[0036] The beneficial effect of such an installation is that it can improve system performance by cleaning up old IMOs and optimizing the database configuration.

[0037] Users can update the conversation history database themselves. For example, users can send database viewing commands via a graphical interface or voice commands, and then view the IMOs in the conversation history database in a visualization interface. They can also perform operations such as editing, deleting, and deduplication on the IMOs in the conversation history database. Human intervention improves the quality of the IMOs, which in turn improves the accuracy and efficiency of subsequent human-machine interactions.

[0038] In S102, a history storage object associated with the current conversation information is determined as a target storage object from the conversation history database.

[0039] For example, when a robot needs to respond to a user's conversation information, it can search for an IMO associated with the current conversation information as a target storage object from a conversation history database. The target storage object can be searched according to the context of the current conversation information and the user's actual needs. For example, the robot may perform a semantic analysis of the current conversation information and search for an IMO that is most similar to the semantic meaning of the current conversation information as a related target storage object, or may search for an IMO that has the most overlapping characters with the current conversation information as a related target storage object.

[0040] The identification rules for the target memory object can be set in advance, and the target memory object is identified according to, for example, the degree of matching of the theme category, the number of overlapping characters, the similarity of the meaning, etc. In this embodiment, there are no particular limitations on the identification rules for the target memory object.

[0041] In S103, response information for the current conversation information is determined according to the target storage object.

[0042] For example, after the target memory object is obtained, the target memory object and the current conversation information can be combined to determine response information for the current conversation information, and the response information can be sent to the user. For example, the response information for the current conversation information can be extracted from the target memory object. An information extraction algorithm can be preset, for example, to extract important information from the target memory object and summarize it as a complete phrase. In this embodiment, the information extraction algorithm is not particularly limited.

[0043] There may be one or more target storage objects, and if multiple target storage objects are found, response information can be determined in combination with the multiple target storage objects. For example, response information for the current conversation information can be extracted from each of the multiple target storage objects, and the extracted multiple response information can be compiled as final response information.

[0044] In an embodiment of the present disclosure, a user can obtain current conversation information and a conversation history database in real time when engaging in human-machine interaction. The conversation history database can include multiple history storage objects, each of which can represent analysis results such as the user's conversation history information, timestamps of the conversation history information, and semantic information. This allows for the generation and storage of individual history storage objects for each piece of conversation history information. According to the analysis results of the history storage objects, a history storage object associated with the current conversation information is determined as a target storage object from the conversation history database. A response is then sent to the user based on the target storage object. By generating a history storage object, the cutting of long conversation histories is reduced, memory capacity is improved, and the history information associated with the current conversation can be more accurately searched, improving the accuracy and efficiency of human-machine interaction and improving the user's interaction experience.

[0045] FIG. 2 is a flowchart of a human-machine interaction method based on conversation history provided by an embodiment of the present disclosure.

[0046] In this embodiment, the step of determining a history storage object associated with the current conversation information from the conversation history database as a target storage object includes the steps of determining an association value between each history storage object and the current conversation information in the conversation history database, which represents the degree of association between the history storage object and the current conversation information, and identifying a target storage object from the conversation history database according to the association value corresponding to each history storage object.

[0047] As shown in FIG. 2, the method includes the following steps S201 to S204.

[0048] In S201, the user's current conversation information and conversation history database are obtained, and the conversation history database includes multiple history storage objects, which represent the user's conversation history information and the analysis results of the conversation history information, and the analysis results include timestamps and semantic information of the conversation history information.

[0049] Illustratively, this step can refer to the above step S101, and therefore will not be described again.

[0050] In S202, a relevance value between each history storage object and the current conversation information is determined in the conversation history database, and the relevance value represents the degree of relevance between the history storage object and the current conversation information.

[0051] For example, for each IMO in the conversation history database, a relevance value between the IMO and the current conversation information is determined, and the relevance value may represent the degree of relevance between the IMO and the current conversation information. A higher relevance value indicates a higher degree of match between the conversation history information of the IMO and the current conversation information. For example, if the topic category of the IMO is the same as the topic category of the current conversation information, the relevance value is high.

[0052] In this embodiment, the step of determining the association value between each history storage object and the current conversation information in the conversation history database includes the steps of: determining a distance value between the history storage object and the current conversation information according to the timestamp of the history storage object, which represents the temporal proximity between the history storage object and the current conversation information; determining a correlation value between the history storage object and the current conversation information according to the semantic information of the history storage object, which represents the degree of semantic association between the conversation history information represented by the history storage object and the current conversation information; and determining the association value between the history storage object and the current conversation information according to the distance value and correlation value corresponding to the history storage object.

[0053] Specifically, for each IMO, the timestamp of the IMO is identified, i.e., the transmission time of the IMO is identified. The transmission time of the current conversation information is identified, and a distance value between the IMO and the current conversation information is determined according to the timestamp of the IMO and the transmission time of the current conversation information. The distance value can represent the temporal proximity between the IMO and the current conversation information. For example, the time difference between the timestamp of the IMO and the transmission time of the current conversation information can be determined, and the distance value can be determined according to the time difference. The smaller the time difference, the smaller the distance value, i.e., the closer the IMO is to the current conversation information in time.

[0054] A semantic analysis of the current conversation information is performed to obtain the semantic information of the current conversation information. The semantic information of the history storage object is determined, and a correlation value between the history storage object and the current conversation information is determined according to the semantic information of the history storage object and the semantic information of the current conversation information. The correlation value can represent the degree of semantic association between the conversation history information represented by the history storage object and the current conversation information, and the greater the similarity between the semantics of the current conversation information and the semantics of the history storage object, the greater the correlation value. For example, the greater the similarity between the theme category of the current conversation information and the theme category of the history storage object, the greater the correlation value.

[0055] The semantic information can include information such as headings, keywords, topic categories, and sentiment tags, and these pieces of information can be combined to calculate a correlation value. For example, the headings of the current conversation information are matched with the headings of the history storage object to obtain a numerical value as a matching result, and the keywords of the current conversation information are matched with the keywords of the history storage object to obtain another numerical value as a matching result. The two numerical values ​​are added to obtain a correlation value.

[0056] The distance value and the correlation value corresponding to the history storage object are combined to determine the relevance value between the history storage object and the current conversation information. For example, the distance value and the correlation value can be added to obtain the relevance value, or the average value of the distance value and the correlation value can be calculated as the relevance value.

[0057] The beneficial effect of such a setup is that for each IMO, the proximity and correlation with the current conversation information are calculated, and then they are combined to obtain a relevance value, which can more accurately search for the target memory object associated with the current conversation information, improve response accuracy, and improve user experience.

[0058] In this embodiment, the semantic information includes a sentiment tag that represents the sentiment expressed by the conversation history information represented by the history storage object, and the step of determining the correlation value between the history storage object and the current conversation information according to the semantic information of the history storage object includes the steps of: performing sentiment analysis on the current conversation information to obtain the sentiment information of the current conversation information; determining the similarity between the sentiment information of the current conversation information and the sentiment tag of the history storage object; and determining the correlation value between the history storage object and the current conversation information according to the similarity.

[0059] Specifically, the semantic information may include a sentiment tag that can represent the sentiment expressed by the conversation history information represented by the history storage object. For each IMO, the sentiment tag of the IMO can be obtained, that is, the sentiment expressed in the conversation history information of the IMO can be identified.

[0060] A sentiment analysis algorithm is preset, and sentiment analysis is performed on the current conversation information to obtain the sentiment information of the current conversation information, that is, to identify the sentiment expressed in the current conversation information. In this embodiment, the sentiment analysis algorithm is not particularly limited.

[0061] A similarity determination algorithm is preset to calculate the similarity between the sentiment information of the current conversation information and the sentiment tag of the history storage object. The similarity can represent the degree of similarity between the sentiment expressed by the current conversation information and the conversation history information of the history storage object. For example, if the sentiment information of the current conversation information represents a positive sentiment and the sentiment tag of the history storage object also represents a positive sentiment, the similarity can be high. In this embodiment, the preset similarity determination algorithm is not particularly limited.

[0062] A correlation value between the history storage object and the current conversation information is determined according to the similarity between the sentiment information of the current conversation information and the sentiment tag of the history storage object. The higher the similarity, the higher the correlation value. For example, the similarity can be determined as the correlation value.

[0063] The beneficial effect of such a configuration is that the correlation value can be determined according to information associated with the sentiment of the semantic information, making the target memory object more matched with the current conversation information and making the answer information more suited to the user's current state, thereby increasing the accuracy of identifying the answer information and improving the user experience.

[0064] In this embodiment, the step of determining the association value between the history storage object and the current conversation information according to the distance value and correlation value corresponding to the history storage object includes the steps of: determining the phrase structure of the conversation history information represented by the history storage object, which represents the grammar and word order of the conversation history information; determining the importance value of the history storage object based on the phrase structure, which represents the grammatical consistency and contextual importance of the conversation history information represented by the history storage object; and determining the association value between the history storage object and the current conversation information according to the distance value, correlation value, and importance value corresponding to the history storage object.

[0065] Specifically, for each IMO, the phrase structure of the conversation history information represented by the IMO can be determined. The phrase structure can represent the grammar and word order of the conversation history information, i.e., the grammar and sentence word order used in the conversation history information can be determined. For example, the subject, predicate, object, etc. of the conversation history information can be determined, and whether the conversation history information is completely, clearly, and vividly expressed can be determined. The grammatical structure and word order can always reveal the importance of information. For example, in many languages, important elements such as subjects, predicates, and objects are usually placed in the center of a sentence, while other modifying elements such as modifiers and adverbs may be placed at the edges of the sentence. The phrase structure can also represent the sentence pattern of the conversation history information. For example, special sentence patterns such as emphasis constructions and inversion constructions are often used to emphasize the importance of information.

[0066] An importance value of a history storage object can be determined based on the phrase structure, and the importance value can represent the grammatical consistency and contextual importance of the conversation history information represented by the history storage object. Whether a phrase is consistent can be determined based on the grammatical structure, and the more consistent and readable the phrase, the higher the importance value is considered to be. For example, if the positions of the subject and predicate in the conversation history information are reversed, the importance value can be determined to be low. Even the same phrase may have different importance depending on the context. For example, "It looks like it's going to rain tomorrow" may be important when planning outdoor activities, but may not be as important when discussing a historical event. In other words, the importance value of the conversation history information can be determined by combining the phrase structure and the context of the conversation history information.

[0067] For each IMO, a distance value, a correlation value, and an importance value can be calculated, and these three indices can be calculated together to obtain the relevance value of the IMO, e.g., these three indices can be added together to obtain the final relevance value.

[0068] The beneficial effect of such a setting is that it determines the importance of the IMO, and the higher the importance, the higher the possibility of identifying it as the target memory object, thereby increasing the accuracy of identifying answer information and improving the user's interaction experience.

[0069] This embodiment further includes the steps of determining the number of times the conversation history information represented by the history storage object appears in the conversation history database, and adjusting the importance value of the history storage object according to the number of times it appears.

[0070] Specifically, there may be overlapping IMOs in the conversation history database, and for the overlapping IMOs, the number of times the IMOs appear in the conversation history database can be determined. Depending on the number of times they appear, the importance value of the IMO can be determined or adjusted. For example, the more times an IMO appears, the higher the importance value.

[0071] The IMO importance may be determined based on the lexical structure, and then adjusted based on the frequency of occurrence. The IMO importance may be obtained by performing a weighted sum based on the grammatical structure and frequency of occurrence.

[0072] The beneficial effect of such a configuration is that the importance value of an IMO can be increased according to the number of occurrences, so that when selecting a target storage object, an IMO with a high number of occurrences can be selected preferentially, thereby contributing to meeting the actual needs of users and improving the user experience.

[0073] In this embodiment, the step of determining the relevance value between the history storage object and the current conversation information according to the distance value, correlation value, and importance value corresponding to the history storage object includes the step of performing weighted processing of the distance value, correlation value, and importance value corresponding to the history storage object, and determining the weighted processing result as the relevance value between the history storage object and the current conversation information.

[0074] Specifically, the weights of the three indicators, distance value, correlation value, and importance value, can be set according to actual needs or preset rules. After the distance value, correlation value, and importance value of the IMO are obtained, weighted processing can be performed based on the preset weights of these three indicators. For example, weighted addition can be performed. The weighted calculation result is determined, and the calculation result is determined as the correlation value between the IMO and the current conversation information.

[0075] The beneficial effect of such an arrangement is that, when these three indicators are considered together, it increases the accuracy of identifying relevant values ​​and therefore the accuracy of identifying answer information.

[0076] In S203, a target storage object is identified from the conversation history database according to the association value corresponding to each history storage object.

[0077] For example, after the relevance value of each IMO is obtained, a target storage object can be identified from the conversation history database according to the magnitude of each relevance value, for example, the IMO with the largest relevance value can be identified as the target storage object.

[0078] In this embodiment, for each IMO, a relevance value between the IMO and the current conversation information can be calculated, thereby finding the most relevant conversation history information to assist in responding to the current conversation information, improving the accuracy of human-machine interaction, and improving the user experience.

[0079] In S204, response information for the current conversation information is determined according to the target storage object.

[0080] Illustratively, this step can be referred to above in step S103, and therefore will not be described again.

[0081] In an embodiment of the present disclosure, a user can obtain current conversation information and a conversation history database in real time when engaging in human-machine interaction. The conversation history database can include multiple history storage objects, each of which can represent analysis results such as the user's conversation history information, timestamps of the conversation history information, and semantic information. This achieves generating and storing individual history storage objects for each conversation history information. According to the analysis results of the history storage objects, a history storage object associated with the current conversation information is identified as a target storage object from the conversation history database. A response is then sent to the user based on the target storage object. Generating a history storage object reduces the need to cut off long conversation histories, improves memory capacity, and enables more accurate retrieval of history information associated with the current conversation, thereby improving the accuracy and efficiency of human-machine interaction and improving the user's interaction experience.

[0082] FIG. 3 is a flowchart of a human-machine interaction method based on conversation history provided by an embodiment of the present disclosure.

[0083] In this embodiment, the step of determining response information for the current conversation information according to the target memory object includes the step of inputting the target memory object and the current conversation information into a predetermined large-scale language model to obtain response information for the current conversation information, and the predetermined large-scale language model is a neural network model and is used to converse with the user during human-machine interaction.

[0084] As shown in FIG. 3, the method includes the following steps S301 to S303.

[0085] In S301, the user's current conversation information and conversation history database are obtained, and the conversation history database includes multiple history storage objects, which represent the user's conversation history information and the analysis results of the conversation history information, and the analysis results include timestamps and semantic information of the conversation history information.

[0086] Illustratively, this step can be referred to above in step S101, and therefore will not be described again.

[0087] In S302, a history storage object associated with the current conversation information is determined as a target storage object from the conversation history database.

[0088] Illustratively, this step can be referred to above in step S102, and therefore will not be described again.

[0089] In S303, the target memory object and the current conversation information are input into a pre-defined large-scale language model to obtain response information of the current conversation information, and the pre-defined large-scale language model is a neural network model, which is used to converse with the user during human-machine interaction.

[0090] For example, assuming that an LLM model is pre-configured, after a target memory object is obtained, both the target memory object and current conversation information can be input into the pre-configured LLM model, and response information for the current conversation information can be output by the LLM model. The LLM model is a pre-configured and trained neural network model that can be used to converse with a user and respond to questions from the user during human-machine interaction. In this embodiment, the model structure of the LLM is not particularly limited.

[0091] For example, a prompt for the LLM model can be generated based on the target memory object and the current conversation information. The LLM can perform a semantic analysis of the prompt and obtain response information for the current conversation information by referring to the context of the current conversation information. For example, if the current conversation information is "What are the directions to Housing Complex A?" and it is assumed that Housing Complex A exists in City B and Housing Complex A also exists in City C, and the conversation history information of the target memory object mentions Housing Complex A in City B, the response information can be directions to Housing Complex A in City B.

[0092] In this embodiment, the large-scale language model can combine current conversation information with the target memory object to automatically generate answer information, and using the target memory object, the LLM can more accurately understand the user's intention, improve the quality of the response, and enhance the user experience.

[0093] In an embodiment of the present disclosure, a user can obtain current conversation information and a conversation history database in real time when engaging in human-machine interaction. The conversation history database can include multiple history storage objects, each of which can represent analysis results such as the user's conversation history information, timestamps of the conversation history information, and semantic information. This achieves generating and storing individual history storage objects for each conversation history information. According to the analysis results of the history storage objects, a history storage object associated with the current conversation information is identified as a target storage object from the conversation history database. A response is then sent to the user based on the target storage object. Generating a history storage object reduces the need to cut off long conversation histories, improves memory capacity, and enables more accurate retrieval of history information associated with the current conversation, thereby improving the accuracy and efficiency of human-machine interaction and improving the user's interaction experience.

[0094] 4 is a structural block diagram of a human-machine interaction device based on conversation history provided by an embodiment of the present disclosure. For convenience of explanation, only parts related to the embodiment of the present disclosure are shown. Referring to FIG. 4, the human-machine interaction device 400 based on conversation history includes an acquiring unit 401, a target determining unit 402 and a response determining unit 403.

[0095] The acquisition unit 401 is used to acquire a user's current conversation information and a conversation history database, the conversation history database including a plurality of history storage objects, the history storage objects representing the user's conversation history information and an analysis result of the conversation history information, the analysis result including timestamps and semantic information of the conversation history information; The target determination unit 402 is used to determine a history storage object associated with the current conversation information from the conversation history database as a target storage object; The response determination unit 403 is used for determining response information of the current conversation information according to the target storage object.

[0096] FIG. 5 is a structural block diagram of a human-machine interaction device based on conversation history provided by an embodiment of the present disclosure. As shown in FIG. 5, the human-machine interaction device 500 based on conversation history includes an acquisition unit 501, a target determination unit 502 and a response determination unit 503, and the target determination unit 502 includes a first determination module 5021 and a second determination module 5022.

[0097] a first determination module 5021 for determining a correlation value between each history storage object and the current conversation information in the conversation history database, the correlation value representing the degree of correlation between the history storage object and the current conversation information; The second determination module 5022 is used to identify the target storage object from the conversation history database according to the association value corresponding to each history storage object.

[0098] In one example, the first determination module 5021 includes a first determination sub-module, a second determination sub-module, and a third determination sub-module; a first determining submodule for determining a distance value between the history storage object and the current conversation information according to a timestamp of the history storage object, the distance value representing the temporal proximity between the history storage object and the current conversation information; a second determination sub-module for determining a correlation value between the history storage object and the current conversation information according to the semantic information of the history storage object, the correlation value representing the degree of semantic association between the conversation history information represented by the history storage object and the current conversation information; The third determination sub-module is used for determining the association value between the history storage object and the current conversation information according to the distance value and correlation value corresponding to the history storage object.

[0099] In one example, the semantic information includes a sentiment tag that represents a sentiment expressed by the conversation history information represented by the history storage object, and the second determination submodule specifically: performing a sentiment analysis on the current conversation information to obtain sentiment information of the current conversation information; determining a similarity between sentiment information of the current conversation information and sentiment tags of the history storage object; and determining a correlation value between the history storage object and the current conversation information according to the similarity.

[0100] In one example, the third determination submodule specifically: determining a lexical structure of the conversation history information represented by the history storage object, the lexical structure representing the grammar and word order of the conversation history information; determining an importance value of the history storage object based on the phrase structure, the importance value representing the grammatical consistency and contextual importance of the conversation history information represented by the history storage object; and determining a relevance value between the history storage object and the current conversation information according to a distance value, a correlation value, and an importance value corresponding to the history storage object.

[0101] In one example, a frequency determination unit used to determine the frequency of occurrence in the conversation history database of the conversation history information represented by the history storage object; The method further includes an importance value adjusting unit, which is used to adjust the importance value of the history storage object according to the number of occurrences.

[0102] In one example, the third determination submodule specifically: A weighted processing of the distance value, correlation value, and importance value corresponding to the history storage object is performed, and the weighted processing result is used to determine the relevance value between the history storage object and the current conversation information.

[0103] In one example, a timestamp determining unit, used for identifying a sending time of the current conversation information and determining the sending time as a timestamp of the current conversation information; a semantic determination unit used for performing semantic analysis processing of the current conversation information to obtain semantic information of the current conversation information; a result determination unit, which is used for determining the timestamp of the current conversation information and the semantic information of the current conversation information as an analysis result of the current conversation information; The conversation history database further includes a storage unit used for determining the current conversation information and the analysis result of the current conversation information as a new history storage object and storing the new history storage object in the conversation history database.

[0104] In one example, a conversation identification unit in the conversation history database, used to identify conversation information immediately before the current conversation information; The device further includes an identifier adding unit, which is used to add a predetermined logical identifier to the current conversation information and the conversation information immediately preceding the current conversation information, and the predetermined logical identifier indicates that there is a logical structure and a time sequence between the two conversation information.

[0105] In one example, the response determination unit 503: The system further includes a model response module for inputting the target memory object and the current conversation information into a predetermined large-scale language model to obtain response information for the current conversation information, wherein the predetermined large-scale language model is a neural network model and is used to converse with a user during human-machine interaction.

[0106] In one example, a timestamp obtaining unit, used for obtaining timestamps of history storage objects in the conversation history database according to a preset information update period; The system further includes a database update unit, which is used to delete the history storage object from the conversation history database when it is determined that the length of storage time of the history storage object in the conversation history database exceeds a predetermined threshold value of time length according to the timestamp of the history storage object.

[0107] FIG. 6 is a structural block diagram of an electronic device provided by an embodiment of the present disclosure. As shown in FIG. 6, the electronic device 600 includes at least one processor 602 and a memory 601 communicatively connected to the at least one processor 602, wherein the memory stores instructions executable by the at least one processor 602, and the instructions are executed by the at least one processor 602, thereby enabling the at least one processor 602 to perform the human-machine interaction method based on conversation history according to the present disclosure.

[0108] The electronic device 600 further includes a receiver 603 and a transmitter 604. The receiver 603 is used to receive commands and data transmitted from other devices, and the transmitter 604 is used to transmit commands and data to external devices.

[0109] According to an embodiment of the present disclosure, the present disclosure further provides an electronic device, a readable storage medium, and a computer program.

[0110] According to an embodiment of the present disclosure, the present disclosure further provides a computer program, the computer program being stored in a readable storage medium, and at least one processor of the electronic device being able to read the computer program from the readable storage medium, and the at least one processor executing the computer program causing the electronic device to perform the means provided by any of the above embodiments.

[0111] 7 is a schematic block diagram of an exemplary electronic device 700 that can be used to implement embodiments of the present disclosure. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smartphones, wearable devices, and other similar computing devices. The components, their connections and relationships, and their functions illustrated herein are merely examples and are not intended to limit the description herein and / or the practice of the disclosure as claimed.

[0112] 7, device 700 includes a computing unit 701, which can perform various appropriate operations and processes based on a computer program stored in a read-only memory (ROM) 702 or loaded from a storage unit 708 into a random access memory (RAM) 703. The RAM 703 can further store various programs and data necessary for the operation of device 700. The computing unit 701, ROM 702, and RAM 703 are connected to each other via a bus 704. An input / output (I / O) interface 705 is also connected to the bus 704.

[0113] Several components in device 700 are connected to I / O interface 705, including input units 706 such as a keyboard and a mouse, output units 707 such as various types of monitors and speakers, storage units 708 such as magnetic disks and optical disks, and communication units 709 such as a network card, modem, wireless communication transceiver, etc. The communication units 709 enable device 700 to exchange information / data with other devices via computer networks such as the Internet and / or various telecommunication networks.

[0114] The computing unit 701 may be any of a variety of general-purpose and / or special-purpose processing components having processing or computational capabilities. Some examples of the computing unit 701 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various specialized artificial intelligence (AI) computing chips, various computing units that execute machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 701 performs each of the methods and processes described above, such as the human-machine interaction method based on a conversation history. For example, in some embodiments, the human-machine interaction method based on a conversation history may be embodied as a computer software program tangibly contained in a machine-readable medium, such as the storage unit 708. In some embodiments, some or all of the computer program may be loaded and / or installed into the device 700 via the ROM 702 and / or the communication unit 709. When loaded into the RAM 703 and executed by the computing unit 701, the computer program may perform one or more steps of the human-machine interaction method based on a conversation history described above. Alternatively, in other embodiments, the computing unit 701 may be configured to perform the conversation history based human-machine interaction method using any other suitable means (e.g., relying on firmware).

[0115] The various system and technical embodiments described herein may be implemented in digital electronic circuitry systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on a chip (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may be embodied in one or more computer programs that can be executed and / or interpreted by a programmable system that includes at least one programmable processor, which may be a special purpose or general purpose programmable processor, that can receive data and instructions from a storage system, at least one input device, and at least one output device, and that can transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0116] Program code for implementing the methods of the present disclosure can be programmed using any combination of one or more programming languages. These program codes can be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus, so that when the program code is executed by the processor or controller, the functions / operations shown in the flowcharts and / or block diagrams are performed. The program code can be entirely executed on a machine, partially executed on a machine, partially executed on a machine as a separate software package and partially executed on a remote machine, or entirely executed on a remote machine or server.

[0117] In the context of this disclosure, a machine-readable medium may be a tangible medium that can contain or store a program for use with, or in connection with, an instruction execution system, device, or apparatus. The machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. The machine-readable medium includes, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or apparatus, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include one or more wire-based electrical connections, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0118] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to a user, and a keyboard and pointing device (e.g., a mouse or trackball) by which a user can provide input to the computer. Other types of devices can also provide interaction with a user; for example, the feedback provided to the user can be any form of sensing feedback (e.g., visual feedback, auditory feedback, or haptic feedback) and can receive input from the user in any form (including acoustic, speech, and tactile input).

[0119] The systems and techniques described herein may be implemented in a computing system including a back-end component (e.g., a data server), a computing system including a middleware component (e.g., an application server), a computing system including a front-end component (e.g., a user computer having a graphical user interface or a web browser through which a user interacts with embodiments of the systems and techniques described herein), or a computing system including any combination of back-end, middleware, or front-end components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network (LAN), a wide area network (WAN), and the Internet.

[0120] The computer system may include a client terminal and a server. The client terminal and the server are generally remote from each other and typically interact via a communication network. The relationship between the client terminal and the server is created by a computer program running on a corresponding computer and having a client terminal-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in a cloud computing service system and solves the shortcomings of traditional physical hosts and virtual private servers (abbreviated as "VirtuaL Private Server" or "VPS"), such as difficult management and poor business scalability. The server may also be a server in a distributed system or a server combined with a blockchain.

[0121] It should be understood that steps can be rearranged, added, or deleted using the various forms of flow shown above. For example, the steps described in the present disclosure may be performed in parallel, sequentially, or in a different order, but this specification does not limit this as long as the technical solution disclosed in the present disclosure can achieve the desired results.

[0122] The above-described embodiments of the present invention do not limit the scope of protection of the present disclosure. Those skilled in the art can make various modifications, combinations, subcombinations, and substitutions based on design requirements and other factors. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present disclosure should be included within the scope of protection of the present disclosure.

Claims

1. acquiring a user's current conversation information and a conversation history database, the conversation history database including a plurality of history storage objects, the history storage objects representing the user's conversation history information and an analysis result of the conversation history information, the analysis result including a timestamp and semantic information of the conversation history information; determining a history storage object associated with the current conversation information from the conversation history database as a target storage object; determining response information for the current conversation information according to the target storage object.

2. The step of determining, from the conversation history database, a history storage object associated with the current conversation information as a target storage object includes: determining an association value between each history storage object and the current conversation information in the conversation history database, the association value representing a degree of association between the history storage object and the current conversation information; and identifying the target stored object from the conversation history database according to a relevance value corresponding to each history stored object.

3. determining an association value between each history storage object and the current conversation information in the conversation history database; determining a distance value between the history storage object and the current conversation information according to a timestamp of the history storage object, the distance value representing the temporal proximity between the history storage object and the current conversation information; determining a correlation value between the history storage object and the current conversation information according to semantic information of the history storage object, the correlation value representing a degree of semantic association between the conversation history information represented by the history storage object and the current conversation information; and determining a correlation value between the history storage object and the current conversation information according to a distance value and a correlation value corresponding to the history storage object.

4. The semantic information includes a sentiment tag that indicates a sentiment expressed by the conversation history information represented by the history storage object, and the step of determining a correlation value between the history storage object and the current conversation information according to the semantic information of the history storage object includes: performing a sentiment analysis on the current conversation information to obtain sentiment information of the current conversation information; determining a similarity between sentiment information of the current conversation information and sentiment tags of the history storage object; and determining a correlation value between the history storage object and the current conversation information according to the similarity.

5. determining a relevance value between the history storage object and the current conversation information according to a distance value and a correlation value corresponding to the history storage object, a step of determining a phrase structure of the conversation history information represented by the history storage object, the phrase structure representing the grammar and word order of the conversation history information; determining an importance value of the history storage object based on the phrase structure, the importance value representing the grammatical consistency and contextual importance of the conversation history information represented by the history storage object; 5. A human-machine interaction method based on conversation history according to claim 3 or 4, further comprising: determining a relevance value between the history storage object and the current conversation information according to a distance value, a correlation value, and an importance value corresponding to the history storage object.

6. determining the number of occurrences in the conversation history database of the conversation history information represented by the history storage object; The method for human-machine interaction based on conversation history according to claim 5 , further comprising the step of: adjusting an importance value of the history storage object according to the number of occurrences.

7. determining a relevance value between the history storage object and the current conversation information according to a distance value, a correlation value, and an importance value corresponding to the history storage object, 7. The human-machine interaction method based on conversation history according to claim 6, further comprising the step of performing weighted processing of the distance values, correlation values ​​and importance values ​​corresponding to the history storage object, and determining the weighted processing result as the relevance value between the history storage object and the current conversation information.

8. Identifying a transmission time of the current conversation information and determining the transmission time as a timestamp of the current conversation information; performing a semantic analysis process on the current conversation information to obtain semantic information of the current conversation information; determining a timestamp of the current conversation information and semantic information of the current conversation information as an analysis result of the current conversation information; The human-machine interaction method based on conversation history according to any one of claims 1 to 4, further comprising: determining the current conversation information and the analysis result of the current conversation information as a new history storage object and storing it in the conversation history database.

9. identifying conversation information immediately before the current conversation information in the conversation history database; 9. The human-machine interaction method based on conversation history according to claim 8, further comprising the step of assigning a predetermined logical identifier to the current conversation information and the conversation information immediately preceding the current conversation information, the predetermined logical identifier representing that there is a logical structure and a time sequence between the two conversation information.

10. determining response information for the current conversation information in response to the target storage object, 5. A human-machine interaction method based on conversation history according to claim 1, further comprising the step of inputting the target memory object and the current conversation information into a predetermined large-scale language model to obtain response information for the current conversation information, wherein the predetermined large-scale language model is a neural network model and is used to converse with a user during human-machine interaction.

11. acquiring a timestamp of a history storage object in the conversation history database according to a preset information update period; A human-machine interaction method based on conversation history as described in any one of claims 1 to 4, further comprising a step of deleting the history storage object from the conversation history database when it is determined that the length of storage time of the history storage object in the conversation history database exceeds a predetermined threshold length of time according to the timestamp of the history storage object.

12. an acquisition unit used to acquire a user's current conversation information and a conversation history database, the conversation history database including a plurality of history storage objects, the history storage objects representing the user's conversation history information and an analysis result of the conversation history information, the analysis result including timestamps and semantic information of the conversation history information; a target determination unit, used for determining a history storage object associated with the current conversation information from the conversation history database as a target storage object; a response determination unit used for determining response information of the current conversation information according to the target storage object.

13. The target determination unit: a first determination module used to determine a relevance value between each history storage object and the current conversation information in the conversation history database, the relevance value representing a degree of relevance between the history storage object and the current conversation information; A human-machine interaction device based on conversation history according to claim 12, further comprising: a second determination module used to identify the target storage object from the conversation history database according to a related value corresponding to each history storage object.

14. The first determination module: a first determination sub-module used to determine a distance value between the history storage object and the current conversation information according to a timestamp of the history storage object, the distance value representing the temporal proximity between the history storage object and the current conversation information; a second determination sub-module, used to determine a correlation value between the history storage object and the current conversation information according to the semantic information of the history storage object, wherein the correlation value represents the degree of semantic association between the conversation history information represented by the history storage object and the current conversation information; The human-machine interaction device based on conversation history according to claim 13, further comprising: a third determination sub-module used for determining a correlation value between the history storage object and the current conversation information according to a distance value and a correlation value corresponding to the history storage object.

15. The semantic information includes a sentiment tag that indicates a sentiment expressed by the conversation history information represented by the history storage object, and the second determination submodule specifically: performing a sentiment analysis on the current conversation information to obtain sentiment information of the current conversation information; determining a similarity between sentiment information of the current conversation information and sentiment tags of the history storage object; and determining a correlation value between the history storage object and the current conversation information according to the similarity.

16. The third determination sub-module specifically includes: determining a phrase structure of the conversation history information represented by the history storage object, the phrase structure representing the grammar and word order of the conversation history information; determining an importance value of the history storage object based on the phrase structure, the importance value representing the grammatical consistency and contextual importance of the conversation history information represented by the history storage object; and determining a relevance value between the history storage object and the current conversation information according to a distance value, a correlation value, and an importance value corresponding to the history storage object.

17. a frequency determination unit used to determine the frequency of occurrence of the conversation history information represented by the history storage object in the conversation history database; The conversation history-based human-machine interaction device according to claim 16, further comprising: an importance value adjusting unit, used for adjusting the importance value of the history storage object according to the number of occurrences.

18. The third determination sub-module specifically includes: The human-machine interaction device based on conversation history according to claim 17, wherein weighted processing is performed on the distance value, correlation value and importance value corresponding to the history storage object, and the weighted processing result is used to determine the association value between the history storage object and the current conversation information.

19. a timestamp determining unit, used for identifying a sending time of the current conversation information and determining the sending time as a timestamp of the current conversation information; a semantic determination unit used for performing semantic analysis processing of the current conversation information to obtain semantic information of the current conversation information; a result determination unit, which is used for determining the timestamp of the current conversation information and the semantic information of the current conversation information as an analysis result of the current conversation information; The human-machine interaction device based on conversation history according to any one of claims 12 to 15, further comprising: a storage unit used for determining the current conversation information and the analysis result of the current conversation information as a new history storage object and storing it in the conversation history database.

20. a conversation identification unit used to identify conversation information immediately before the current conversation information in the conversation history database; 20. The human-machine interaction device based on conversation history according to claim 19, further comprising: an identifier adding unit used to add a predetermined logical identifier to the current conversation information and the conversation information immediately before the current conversation information, wherein the predetermined logical identifier represents that there is a logical structure and a time sequence between the two conversation information.

21. The response determination unit: The human-machine interaction device based on conversation history according to any one of claims 12 to 15, further comprising: a model response module used for inputting the target memory object and the current conversation information into a predetermined large-scale language model to obtain response information for the current conversation information, wherein the predetermined large-scale language model is a neural network model and is used to converse with a user during human-machine interaction.

22. a timestamp obtaining unit, used for obtaining timestamps of history storage objects in the conversation history database according to a preset information update period; A human-machine interaction device based on conversation history according to any one of claims 12 to 15, further comprising: a database update unit used for deleting the history storage object from the conversation history database when it is determined that the length of storage time of the history storage object in the conversation history database exceeds a predetermined time length threshold according to the timestamp of the history storage object.

23. at least one processor; a memory communicatively coupled to the at least one processor, An electronic device in which instructions executable by the at least one processor are stored in the memory, and when the instructions are executed by the at least one processor, the at least one processor is able to execute the human-machine interaction method based on conversation history described in any one of claims 1 to 4.

24. A non-transitory computer-readable storage medium having stored thereon computer instructions, the computer instructions being used to cause a computer to perform the method for human-machine interaction based on conversation history according to any one of claims 1 to 4.

25. A computer program, which, when executed by a processor, implements the steps of the method for human-machine interaction based on conversation history according to any one of claims 1 to 4.

Images