Methods for summarizing interactive tasks, servers, electronic devices, and storage media

Abstract

This application discloses a method, server, electronic device, and storage medium for summarizing interactive tasks. The method includes: after reaching the i-th task node with a user, determining whether it is necessary to summarize the task details preceding the i-th task node; if so, engaging in one or more human-computer dialogues with the user to obtain the target summary content for summarizing the task details preceding the i-th task node from the user; performing semantic analysis on the target summary content to obtain a first score; and if the first score is less than a first threshold, sending a preset prompt statement to the user. In this application embodiment, prompt statements allow the user to summarize multiple times until the user can summarize completely, thus training and cultivating the user's summarizing ability; furthermore, prompt statements allow the user to summarize according to strict logic, thus training and cultivating the user's logical rigor.

Description

Technical Field

[0001] This application relates to the field of human-computer dialogue technology, specifically to a method for summarizing interactive tasks, a server, an electronic device, and a storage medium. Background Technology

[0002] Currently, with the development of artificial intelligence, speech processing technology (including speech analysis, speech recognition, semantic understanding, etc.) has been greatly improved. It is precisely because of this improvement in speech processing technology that more and more human-computer dialogue products have emerged. For example, the sales script marketing systems and customer service systems of various e-commerce platforms.

[0003] In addition to the aforementioned marketing systems, interactive tasks based on human-computer dialogue are gradually emerging. These interactive tasks primarily involve pre-setting a task scenario and configuring corresponding nodes within that scenario. Then, at specific nodes, the machine and the human engage in dialogue, achieving human-computer interaction. For example, currently, during interactive tasks, prompts are displayed at specific nodes, asking the user to summarize the previous events. Then, the user's voice is collected, and speech processing technology is used to recognize the user's speech to obtain the summarized content.

[0004] Currently, interactive tasks merely prompt users to summarize the plot, obtaining their summarized content, but no further processing is performed on this summary, resulting in a waste of user-summarized data. Therefore, how to efficiently utilize user-summarized content in interactive tasks is a pressing technical problem that needs to be solved. Summary of the Invention

[0005] This application provides a method for summarizing interactive tasks, a server, an electronic device, and a storage medium, thereby improving the utilization rate of summarized content for users.

[0006] In a first aspect, embodiments of this application provide a method for summarizing interactive tasks. The method is applied to a server, and the interactive task includes multiple task episodes and multiple task nodes, wherein the multiple task episodes and multiple task nodes correspond one-to-one, and are arranged in chronological order. The method includes:

[0007] After reaching the i-th task node with the user, determine whether it is necessary to summarize the task segments located before the i-th task node, wherein the i-th task node is any one of the plurality of task nodes;

[0008] If so, one or more human-computer dialogues are conducted with the user terminal to obtain the target summary content of the user's summary of the task plots located before the i-th task node from the user terminal, wherein the target summary content includes the first content to be summarized corresponding to the i-th task node, and the first content to be summarized is the content summarized by the user among all the content that needs to be summarized at the i-th task node.

[0009] Semantic analysis is performed on the target summary content to obtain a first score, wherein the first score is used to characterize the logical rigor of the user when summarizing the target summary content;

[0010] If the first score is less than the first threshold, a preset prompt statement is sent to the user terminal. The preset prompt statement includes a target statement for expressing the first content to be summarized. The preset prompt statement is used to prompt the user to imitate the target statement in order to fully summarize the first content to be summarized.

[0011] Secondly, embodiments of this application provide a server, characterized in that the server is used to implement interactive tasks, the multiple task plots and multiple task nodes, wherein the multiple task plots and the multiple task nodes correspond one-to-one, and the multiple task plots and the multiple task nodes are arranged in chronological order; the server includes a transceiver unit and a processing unit.

[0012] The processing unit is configured to, after interacting with the user terminal at the i-th task node, determine whether it is necessary to summarize the task segments preceding the i-th task node, wherein the i-th task node is any one of the plurality of task nodes; if so, control the transceiver unit to conduct one or more human-computer interactions with the user terminal to obtain from the user terminal the target summary content for summarizing the task segments preceding the i-th task node, wherein the target summary content includes the first content to be summarized corresponding to the i-th task node, and the first content to be summarized is the content summarized by the user among all the content that needs to be summarized at the i-th task node.

[0013] Semantic analysis is performed on the target summary content to obtain a first score, wherein the first score is used to characterize the logical rigor of the user when summarizing the target summary content;

[0014] The transceiver unit is configured to send a preset prompt statement to the user terminal if the first score is less than a first threshold. The preset prompt statement includes a target statement for expressing the first content to be summarized. The preset prompt statement is used to prompt the user to imitate the target statement in order to fully summarize the first content to be summarized.

[0015] Thirdly, embodiments of this application provide an electronic device, including: a processor and a memory, the processor being connected to the memory, the memory being used to store a computer program, and the processor being used to execute the computer program stored in the memory, so that the electronic device performs the method as described in the first aspect.

[0016] Fourthly, embodiments of this application provide a computer-readable storage medium storing a computer program that causes a computer to perform the method described in the first aspect.

[0017] Fifthly, embodiments of this application provide a computer program product, the computer program product including a non-transitory computer-readable storage medium storing a computer program, the computer being operable to perform the method as described in the first aspect.

[0018] Implementing the embodiments of this application has the following beneficial effects:

[0019] As can be seen in this embodiment, an interactive task is pre-set. For any task node in the interactive task, after interacting with the user, the server first determines whether the task details need to be summarized. If so, the server engages in one or more human-computer dialogues with the user to obtain the target summary content for complete summarization. In other words, if one human-computer dialogue cannot obtain the target summary content, multiple human-computer dialogues will be conducted to obtain the target summary content for complete summarization. This process of multiple human-computer dialogues allows the user to summarize content, thereby improving the user's summarization ability and increasing the utilization rate of the summarized content. Furthermore, after the user has completed the summary, the target summary content can be scored to assess the logical rigor of the user's summarization. When the score is less than a threshold, indicating that the user's summarization logic is not rigorous, the server will send a preset prompt statement to the user, prompting the user to imitate the preset prompt statement to summarize, i.e., to summarize according to rigorous logic, thereby improving the logical rigor of the user's content summarization and further increasing the utilization rate of the summarized content. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 A schematic diagram of an interactive system provided in an embodiment of this application;

[0022] Figure 2 A flowchart illustrating a method for summarizing interactive tasks provided in an embodiment of this application;

[0023] Figure 3 This is a schematic diagram illustrating the sending of a preset prompt statement to a user terminal, as provided in an embodiment of this application.

[0024] Figure 4 This is a schematic diagram illustrating the sending of a third prompt statement to a user terminal, as provided in an embodiment of this application.

[0025] Figure 5 This is a schematic diagram illustrating the sending of a fourth prompt statement to a user terminal, as provided in an embodiment of this application.

[0026] Figure 6 A flowchart illustrating a method for aggregation by a server, provided as an embodiment of this application;

[0027] Figure 7 A functional unit block diagram of a server provided in an embodiment of this application;

[0028] Figure 8 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0029] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0030] The terms "first," "second," "third," and "fourth," etc., used in the specification, claims, and accompanying drawings of this application are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or apparatuses.

[0031] In this document, the term "embodiment" means that a particular feature, result, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0032] To facilitate understanding of the technical solution of this application, the relevant technologies and definitions involved in this application will be explained first.

[0033] First, it should be noted that the interactive task in this application may also be referred to as an interactive product, interactive work, interactive project, etc. For the sake of consistency, this application will refer to it as an interactive task. In practical applications, the interactive task can be an interactive story, interactive game, interactive voice, interactive novel, interactive script, etc. For ease of distinction, when the interactive task is an interactive story, interactive novel, or interactive script, the task plot involved in this application can be called a story plot, and the task node can be called a plot node; when the interactive task is an interactive game, the task plot can be called a game scene, game plot, or game level, and the task node can be called a game node; when the interactive task is interactive voice, the task plot can be called a dialogue plot, and the task node can be called a dialogue node. This application mainly uses an interactive story as an example to illustrate the process of summarizing interactive tasks, but it does not limit the type of interactive task. The summarization process for other types of interactive tasks is similar and will not be described further.

[0034] It should be noted that when the interactive task is an interactive story, the multiple storylines within that task can constitute multiple main storylines. That is, each main storyline includes one or more of these multiple storylines. Any two main storylines may or may not be related, meaning they are independent of each other; this application does not impose any limitations on this. Furthermore, each storyline corresponds to a story node; therefore, reading a storyline can also be understood as reading a story node.

[0035] First, it should be noted that the multiple task nodes in this application can be divided into two types of task nodes: one type requires the user to summarize the previously occurred task events, and the other type introduces the task events.

[0036] For example, an interactive task might have four task segments, or four task nodes. The first task node corresponds to the task segment of discussing where to go with A; the second task node corresponds to discussing when to go with A; the third task node corresponds to discussing what to bring with A; and the fourth task node corresponds to requesting the user to relay the discussion with A to B. The function of the fourth node is to require the user to summarize the task segments from the first three nodes. Therefore, after the fourth task node is completed, the user must summarize the task segments from the first three nodes. The first three nodes are task segments that introduce the task segments.

[0037] For ease of description, in this application, the task node that requires the user to summarize the task details among multiple task nodes is referred to as the target task node.

[0038] It should be noted that when requiring users to summarize previous task plots for a target task node, the summary can be of all plots preceding that target task node, or it can be of only a portion of the plots preceding that target task node. The summary can be set according to actual needs, and this application does not limit it.

[0039] Furthermore, for any given target task node, all content that needs to be summarized at that target task node is pre-configured. For ease of description, this application refers to all content that needs to be summarized at that target task node as the second content to be summarized. Continuing with the example above, for the fourth node, the second content to be summarized at the fourth node can be configured to include: "who to play with," "where to play," "when to go," and "what to bring when going." Considering the user's actual summarizing ability, this application can divide all content that needs to be summarized at any given target task node into two parts: one part is called the content summarized by the user, which is called the first content to be summarized for ease of description, and the other part is called the content summarized by the server, which is called the second content to be summarized for ease of description. The content summarized by the user requires the collection of the user's voice information through the user terminal to complete the summarization of this part. The content summarized by the server is completed by the server directly sending a prompt statement to the user terminal after the task node ends.

[0040] It should be understood that if a user has strong actual aggregation capabilities and can aggregate all the second content to be aggregated for a certain target task node, then the content to be aggregated by the server at that target task node will be blank, that is, the server does not need to aggregate.

[0041] It should be noted that the content summarized by users at the same target task node can be different for different users. In other words, the content summarized by the server at the same target task node can also be different. Specifically, the first and third summaries for each user at any target task node can be determined by the user's age and historical interaction records. How to determine the first and third summaries for a user at any target task node will be explained later, and will not be described in detail here.

[0042] See Figure 1 , Figure 1 This is a schematic diagram of an interactive system provided in an embodiment of this application. The interactive system includes a user terminal 10 and a server 20. The user engages in multiple human-computer dialogues with the server 20 through the user terminal 10 to complete interactive tasks. These interactive tasks include multiple task scenarios and multiple task nodes, such as... Figure 1 The illustration shows storyline 1, storyline 2, ..., storyline n; plot node 1, plot node 2, ..., plot node n. Each of the multiple storylines and plot nodes corresponds one-to-one, and they are arranged in chronological order.

[0043] Optionally, server 20 is configured with interactive tasks. Client 10 can pre-download these interactive tasks from server 20 and cache them locally. This way, when a user interacts with server 20 via client 10 later, the interactive task can be retrieved from the local cache. Alternatively, server 20 can push interactive tasks to client 10 online, eliminating the need for local caching on the client.

[0044] like Figure 1As shown, after the server 20 and the user terminal 10 reach the i-th task node, that is, after the user terminal 10 finishes playing the storyline corresponding to the i-th task node, the server 20 determines whether it is necessary to summarize the tasklines before the i-th task node, that is, to determine whether the i-th task node is the target task node, where the i-th task node is any one of multiple task nodes; if so, the server 20 and the user terminal 10 conduct one or more human-computer dialogues to obtain the target summary content for the user to summarize the tasklines before the i-th task node from the user terminal 10, wherein the target summary content includes the first content to be summarized corresponding to the i-th task node, and the first content to be summarized is the content summarized by the user among all the content that needs to be summarized at the i-th task node. Then, server 20 performs semantic analysis on the target summary content to obtain a first score, wherein the first score is used to characterize the logical rigor of the user when summarizing the target summary content; if the first score is less than a first threshold, server 20 sends a preset prompt statement to user terminal 10, wherein the preset prompt statement includes a target statement for expressing the first content to be summarized, and the preset prompt statement is used to prompt the user to imitate the target statement in order to fully summarize the first content to be summarized.

[0045] It should be noted that the aforementioned user terminal 10 may include smartphones (such as Android phones, iOS phones, Windows Phones, etc.), tablet computers, PDAs, laptops, mobile internet devices (MIDs), or wearable devices. The aforementioned user terminal 10 is merely an example, not an exhaustive list, and includes, but is not limited to, the aforementioned user terminal 10. In practical applications, the aforementioned user terminal 10 may also include: intelligent in-vehicle terminals, computer equipment, etc. The aforementioned server 20 may be a cloud computing server, a Content Delivery Network (CDN) server, a Network Time Protocol (NTP) server, a Domain Name System (DNS) server, and other various types of servers.

[0046] See Figure 2 , Figure 2 This application provides a method for summarizing interactive tasks. This method is applied to the above-described embodiments. The method includes, but is not limited to, the following steps:

[0047] Step 201: After the i-th task node is reached with the user, the server determines whether it is necessary to summarize the task segments located before the i-th task node.

[0048] For example, after the server has interacted with the user to complete the interaction with the i-th task node, that is, after the server determines that the user has finished playing the plot corresponding to the i-th task node, it will determine whether it is necessary to summarize the plots preceding the i-th task node, that is, determine whether the i-th task node is the target task node. Here, the i-th task node can be any one of the plurality of task nodes.

[0049] Optionally, when configuring an interactive task to the server, it is pre-configured which task nodes in the interactive task require the user to summarize the task plots. That is, it is pre-configured which task nodes are target task nodes, and the identifiers (e.g., sequence numbers) of the target task nodes are recorded and cached. In this way, after the server and the user have completed the i-th task node, the server can determine whether the i-th task node is a target task node based on the identifier (e.g., sequence number) of the i-th task node, and thus determine whether it is necessary to summarize the task plots located before the i-th task node.

[0050] Step 202: If yes, the server engages in one or more human-computer dialogues with the user terminal to obtain the target summary content from the user terminal summarizing the task plots located before the i-th task node.

[0051] The target summary content includes the first summary content corresponding to the i-th task node, which is the content summarized by the user among all the content that needs to be summarized at the i-th task node.

[0052] For example, the server and the user engage in one or more human-computer interactions so that the user can fully summarize the content that needs to be summarized at the i-th task node.

[0053] Specifically, when the i-th task node is the target task node, the server first obtains the user's first voice information from the user's end. This first voice information is the user's voice summarizing the task plot preceding the i-th task node; that is, the voice information of the user's first summary of the task plot for the i-th task node. Then, the server analyzes this first voice information to obtain the user's first summary content. For example, the server can convert the voice information to text to obtain the first summary content. Then, the server compares this first summary content with the first content to be summarized to determine whether the user has fully summarized the content that needs to be summarized at the i-th task node. Optionally, it is determined whether the first summary content includes the first content to be summarized. If it does, it is determined that the user has fully summarized the content that needs to be summarized at the i-th task node; if the first summary content does not fully include the first content to be summarized, it is determined that the user has not fully summarized the content that needs to be summarized at the i-th task node.

[0054] For example, content element identification is performed on the first summary content to obtain the content elements of the first summary content. Specifically, content element identification on the first summary content can be understood as entity identification on the first summary content to obtain the entities in the first summary content, and using the entities as content elements of the first summary content. The entities involved in this application may include, but are not limited to, "geographical location," "person's name," "time," "item," "commodity," "price," etc. Then, the content elements of the first summary content are compared with the content elements of the first content to be summarized to determine whether the first summary content contains the first content to be summarized. For example, if the content elements of the first summary content contain the content elements of the first content to be summarized, it is determined that the first summary content contains the first content to be summarized, and the first summary content is used as the target summary content; if the content elements of the first summary content do not contain any of the content elements of the first content to be summarized, that is, the content elements of the first content to be summarized are not a subset of the content elements of the first summary content, it is determined that the first summary content does not completely contain the first content to be summarized.

[0055] In this embodiment, the first summary content includes the first content to be summarized, which can be understood as the first summary content completely including the first content to be summarized; or it can be understood as the first summary content including most of the content to be summarized. Specifically, the intersection between the content elements of the first summary content and the content elements of the first content to be summarized can be obtained; the ratio between the number of elements in the intersection and the number of content elements in the first content to be summarized can be obtained; when the ratio is greater than a preset value, it is determined that the first summary content includes the first content to be summarized. It should be understood that when the ratio is 1, it means that the first summary content completely includes the first content to be summarized; when the ratio is less than 1, it means that the first summary content includes most of the content to be summarized.

[0056] It should be noted that the content elements of the first content to be summarized may include content elements other than those specified at the i-th task node. In other words, the first content to be summarized by the user includes content other than that specified at the i-th task node. However, this application primarily uses the example where the content elements of the first content to be summarized do not include content elements other than those specified at the i-th task node.

[0057] Furthermore, if the first summary content does not completely include the first content to be summarized, that is, if the server determines that the user has not fully summarized the content that needs to be summarized at the i-th task node, the server sends a first prompt statement to the user. This first prompt statement includes a statement expressing the second summary content, which is the summary content of the first content to be summarized excluding the first summary content, i.e., the content that the user missed summarizing. The first prompt statement is used to prompt the user to perform a complete summary. Specifically, the server sends a first prompt statement to the user, including a prompt statement indicating the content the user has not summarized, so that the user can recall the second content to be summarized and perform a complete summary of the i-th task node.

[0058] Furthermore, after sending the first prompt statement to the user, the server can obtain the user's second voice information from the user. This second voice information is the user's voice information summarizing the task details preceding the i-th task node (i.e., the second summary). Similarly, the server performs speech recognition on the second voice information to obtain the second summary content. The method for obtaining the second summary content is similar to that for obtaining the first summary content and will not be described further. Then, the server determines whether the second summary content contains the first content to be summarized. The method for determining whether the second summary content contains the first content to be summarized is similar to that for determining whether the first summary content contains the first content to be summarized and will not be described further.

[0059] Furthermore, if the second summary content includes the first content to be summarized, then the second content to be summarized is used as the target content to be summarized. If the second summary content does not include the second content to be summarized, the server will send a new prompt statement to the user terminal again, and obtain new voice information from the user terminal, that is, the user's voice summarizing the task plots before the i-th task node; then, the server obtains new summary content from the new voice information; repeat the above process until the obtained new summary content includes the first content to be summarized, then the new summary content is used as the target summary content.

[0060] Step 203: The server performs semantic analysis on the target summary content to obtain the first score.

[0061] It is understood that if the target summary content includes the first content to be summarized, it means that the user is now performing a complete summary of the i-th task node. Then, semantic analysis can be performed on the target summary content, that is, the logical rigor between the target summary content is analyzed to obtain the first score. The first score is used to characterize the logical rigor of the user when summarizing the target summary content.

[0062] For example, syntactic analysis is performed on the target summary content to obtain its syntactic dependency tree; syntactic analysis is performed on the first content to be summarized to obtain its syntactic dependency tree; the minimum edit distance between the syntactic dependency tree of the target summary content and the syntactic dependency tree of the first content to be summarized is obtained; based on the minimum edit distance, the first score of the target summary content is determined, wherein the smaller the minimum edit distance, the more similar the target summary content is to the first content to be summarized, the stronger the logic of the target summary content, and the higher the first score.

[0063] Step 204: If the first score is less than the first threshold, the server sends a preset prompt statement to the user terminal.

[0064] For example, when the first score is less than the first threshold, it indicates that the logic of the target summary content compiled by the user is not rigorous enough (unqualified), such as... Figure 3 As shown, the server sends a preset prompt statement to the user, which is a pre-set prompt statement corresponding to the first content to be summarized. Therefore, the preset prompt statement includes a target statement for expressing the first content to be summarized. The preset prompt statement is used to prompt the user to imitate the target statement to fully summarize the first content to be summarized. For example, the preset prompt statement is: "The complete summary content corresponding to story node i includes: 'Content 1', 'Content 2', ... You can imitate the complete content to summarize again." It can be understood that the preset prompt statement contains the complete summary content, and the logic of the preset prompt statement in summarizing the content is relatively rigorous, that is, the first score of the complete summary content corresponding to the preset prompt statement is greater than a first threshold.

[0065] It should be noted that users of different ages have varying logical expression abilities, resulting in different levels of logical rigor in the summarized content. Therefore, the logical validity of the summarized content should be determined based on the user's age. For example, the server obtains the user's age. For instance, the server obtains the user's login information; based on this login information, it determines the user's age. Then, based on this age and the correspondence between age and logical thresholds, the server can determine a first threshold corresponding to the user.

[0066] Furthermore, after sending the preset prompt statement, the server can obtain third voice information from the user's end. This third voice information is a summary of the user's imitation of the preset prompt statement. It can be understood that the server can score the summary of this third voice information to obtain a first score corresponding to that third voice information. If the first score is less than a first threshold, the server can send the preset prompt statement again until logically sound voice information is obtained.

[0067] Interactive tasks are pre-set. For any task node within the interactive task, after interacting with the user at that node, the server first determines whether task content needs to be summarized. If so, the server engages in one or more human-computer dialogues with the user to obtain the target summary content for complete summarization. In other words, if one human-computer dialogue cannot obtain the target summary content, multiple dialogues will be conducted to obtain the target summary content for complete summarization. This process of multiple human-computer dialogues allows the user to summarize content, improving their summarization ability and thus increasing the utilization rate of the summarized content. Furthermore, after the user completes the summary, they can rate the target summary content to assess the logical rigor of their summarization. When the rating is below a threshold, indicating that the user's summary logic is not rigorous, the server will send a preset prompt to the user, prompting them to follow the preset prompt to summarize according to rigorous logic, thereby improving the logical rigor of the user's content summarization and further increasing the utilization rate of the summarized content.

[0068] In one embodiment of this application, Figure 2 The embodiment describes the process of prompting the user to perform a summary when the i-th task node is a target task node. When the i-th task node is not a target task node, that is, after performing the i-th task node, if it is not necessary to summarize the task segments preceding the i-th task node, the association between the i-th task node and subsequent target task nodes can be determined. If an association exists, a prompt is given after performing the i-th task node to remind the user to pay attention to the content that needs to be summarized in the task segments of the i-th task node. This helps the user to notice the content that needs to be summarized in advance, avoids the user missing any content, and improves the completeness and accuracy of the user's summary.

[0069] For example, if it is not necessary to summarize the task segments preceding the i-th task node, then q second candidate task nodes following the i-th task node are obtained, wherein the number of task nodes between any second candidate task node and the i-th task node is less than or equal to a third threshold. For example, starting from the i-th task node, task nodes whose intervals with the i-th task node are less than or equal to the third threshold are selected as second candidate task nodes, resulting in q second candidate task nodes.

[0070] It should be noted that in practical applications, all nodes after the i-th node can be considered as second candidate nodes. The reason for filtering task nodes whose intervals from the i-th task node are less than or equal to a third threshold is that if the interval between the second candidate node and the i-th task node is too large, even if the user is reminded at the i-th task node, by the time the user reaches the second candidate node, the large number of nodes between them will have caused the user to forget the task details of the i-th task node, rendering the reminder at the i-th task node ineffective. Therefore, this application filters task nodes whose intervals are less than or equal to the third threshold, ensuring that the distance between any second candidate task node and the i-th task node is not too far, thus guaranteeing that the reminder to the user at the i-th task node is effective.

[0071] Further, it is determined whether a second target task node exists among the q second candidate task nodes, that is, whether a task node requiring user task plot summarization exists among the q second candidate task nodes. If it exists, it is determined whether there is a correlation between the task plot corresponding to the i-th task node and the fourth content to be summarized corresponding to the second target task node. For example, the fourth content to be summarized corresponding to the second target task node is obtained, that is, the content to be summarized by the user among all the content that needs to be summarized at the second target task node. Keyword extraction is performed on the task plot of the i-th task node to obtain multiple keywords, and then it is determined whether the content elements of the fourth content to be summarized contain the multiple keywords; if they are contained, it is determined that there is a correlation between the task plot corresponding to the i-th task node and the fourth content to be summarized corresponding to the second target task node, that is, the fourth content to be summarized contains the task plot corresponding to the i-th task node.

[0072] Furthermore, in cases where there is a connection, such as Figure 4 As shown, before proceeding to the i-th task node, that is, before the user plays the task plot corresponding to the i-th task node, a third prompt statement is sent to the user. For example, the third prompt statement is: "There is content that needs to be summarized in plot node i. When reading the task plot of plot node i, you can pay more attention to this task plot." Therefore, the third prompt statement is used to indicate that there is content that needs to be summarized in the task plot corresponding to the i-th task node, thereby reminding the user to pay attention to the task plot corresponding to the i-th task node.

[0073] It should be understood that if there is no task node among the q second candidate task nodes that requires the user to summarize the task plot, or if there is no correlation between the task plot corresponding to the i-th task node and the fourth content to be summarized corresponding to the second target task node, then after the i-th task node is executed, that is, after the user has finished playing the task plot corresponding to the i-th task node, the (i+1)-th task node will be executed, that is, the user will be instructed to play the task plot corresponding to the (i+1)-th task node.

[0074] In one embodiment of this application, if there is a correlation between the task plot corresponding to the i-th task node and the fourth content to be summarized corresponding to the second target task node, then after performing the i-th task node, if... Figure 5 As shown, if an exit command is received from the user, where the exit command requests to exit the current interactive task (i.e., close the interactive task), the server first obtains the number of task nodes x between the i-th task node and the second target task node, and then sends a fourth prompt statement to the user. This fourth prompt statement prompts the user to exit the current interactive task after completing x task nodes. For example... Figure 5 As shown, the fourth prompt statement could be: "To avoid missing previous content when summarizing tasks, read x more task details before summarizing the tasks."

[0075] As can be seen, in this implementation, when the user's exit command is received after reading any task node, the interactive task will not be exited immediately. Instead, it will first determine whether there is a connection between the task node and the subsequent target task node. If there is, the user will be prompted to read the plot node of the target task node and summarize the information before exiting. This can prevent the user from forgetting the content that needs to be summarized after reading the interactive story for a long time.

[0076] See Figure 6 , Figure 6 This is a flowchart illustrating a method for aggregation by a server, provided as an embodiment of this application. In this embodiment, [the method is related to...]. Figure 2 The same content as the illustrated embodiments will not be repeated here. This method applies to the above. The method includes, but is not limited to, the following steps:

[0077] 601: The server obtains the user's age and historical interaction records.

[0078] For example, the server can obtain a user's age through their login information. Since this application uses an interactive story as an example, the historical interaction record can be understood as the user's historical reading record of that interactive story. This historical reading record includes the number of times the user has read the interactive story, the total number of times each target task node in the interactive story has been read, and so on. This application mainly uses the historical reading count as an example for explanation.

[0079] 602: The server determines the first number of users based on their age.

[0080] Wherein, the first quantity is the number of contents that can be summarized by the age.

[0081] For example, if the server is pre-configured with a mapping relationship between age ranges and the amount of content to be aggregated, the server determines the age range to which the user's age falls, and based on the age range and the mapping relationship, determines a first amount of content that the user can aggregate. It should be noted that this first amount is the theoretical amount that the user can aggregate; however, due to differences in the abilities of different users, the amount that a user can aggregate may be more or less than this first amount.

[0082] For example, the server can determine the user's ability coefficient, i.e. the user's actual aggregate ability, based on the mapping relationship between historical reading counts and ability coefficients.

[0083] 603: The server determines the user's actual aggregation capability based on the historical interaction records.

[0084] For example, a user's actual summarization ability can be represented by a capability coefficient y, where the value of the capability coefficient y ranges from 0 to 1. When the capability coefficient y is 1, it indicates that the user can summarize all the content to be summarized for any target node in the interactive task. Therefore, the user's actual summarization ability is used to characterize the user's ability to summarize all the content to be summarized for any target node in the interactive task.

[0085] 604: The server adjusts the first quantity based on the actual aggregation capacity to obtain the second quantity.

[0086] For example, the server multiplies the user's actual aggregation capability (capability coefficient) by the first quantity to obtain the second quantity, thus obtaining the quantity that the user can actually aggregate. Here, the second quantity is the quantity of content that the user can actually aggregate.

[0087] 605: The server determines the fourth quantity of the third content to be summarized based on the second and third quantities.

[0088] For example, the server first obtains the second content to be summarized at the i-th task node based on the identifier (e.g., sequence number). This second content to be summarized comprises all content that needs to be summarized at the i-th task node, and the quantity of this second content to be summarized is the third quantity. In other words, there is a third quantity of second content to be summarized at the i-th task node. Then, since the user can actually summarize the second quantity, there is a fourth quantity (the third quantity - the second quantity) of second content that the user cannot summarize. This portion of content needs to be summarized by the server; that is, the server needs to summarize the fourth quantity of content at the i-th task node. For ease of distinction, this application refers to this portion of content as the fourth quantity of third content to be summarized.

[0089] Therefore, in this application, the third quantity of the second content to be summarized at the i-th task node is split into the second quantity of the first content to be summarized and the fourth quantity of the third content to be summarized by the user.

[0090] 606: The server selects the fourth number of third items to be summarized from the third number of second items to be summarized.

[0091] For example, the task segment containing each second piece of content to be summarized is obtained, that is, it is determined in which task segment each piece of content to be summarized appears. Then, based on the task segment containing each piece of content to be summarized, it is determined whether the second number of pieces of content to be summarized appear in different task segments.

[0092] Optionally, if the second quantity of second items to be summarized appear in completely different task scenarios, then the appearance time of the task scenario containing each second item to be summarized is taken as the appearance time of each second item to be summarized. The fourth quantity of third items to be summarized is then selected from the third quantity of second items to be summarized in chronological order of their appearance times. It should be noted that the earlier the appearance time of the task scenario containing each second item to be summarized, the greater the probability that the user will forget the task scenario containing that second item to be summarized, thus making it more difficult for the user to summarize that second item to be summarized. Therefore, the second items to be summarized earlier can be delegated to the server for summarization, thereby reducing the difficulty for the user to summarize and improving the accuracy of the user's summary.

[0093] Optionally, if at least two of the second items to be summarized in the second quantity appear in the same task scenario, i.e., the second items to be summarized in the second quantity do not all appear in the same task scenario, then the second items to be summarized in the second quantity that appear in the same task scenario are grouped together to obtain multiple groups of items to be summarized, and the fourth quantity of third items to be summarized is selected from the multiple groups of items to be summarized. Therefore, the second items to be summarized in each group of items appear in the same task scenario.

[0094] For example, firstly, a second score is obtained for the second content to be summarized in each group of content to be summarized. This second score represents the probability that the user remembers the second content to be summarized after completing the task scenario containing that content. This probability reflects the user's reading richness of the task scenario containing that content to be summarized.

[0095] The following explanation uses the j-th second item to be summarized in the k-th group of items to be summarized as an example to illustrate how to obtain the second score of the j-th second item to be summarized. The method for determining the second score of other second items to be summarized is similar and will not be described again. Here, the k-th group of items to be summarized refers to any one of the multiple groups of items to be summarized, and the j-th second item to be summarized refers to any one of the second items to be summarized in the k-th group of items to be summarized.

[0096] For example, h first candidate task nodes are obtained from the task nodes preceding the i-th task node that are related to the j-th second content to be summarized. Optionally, the first task plot corresponding to each task node preceding the i-th task node is obtained; then the second task plot where the j-th second content to be summarized is located is obtained. Then, the correlation between the first task plot and the second task plot is obtained to obtain the correlation between each task node preceding the i-th task node and the j-th second content to be summarized, and the task nodes with a correlation greater than a preset threshold are taken as first candidate task nodes that are related to the j-th second content to be summarized, thus obtaining h first task nodes. Further, for each first candidate task node, it is determined whether user interaction is required at the first candidate task node, that is, whether the user participates in the interaction with the server; if so, the first candidate task node is taken as a first target task node, thereby selecting r first target task nodes from the h first candidate task nodes. Finally, based on the h first candidate task nodes and the r first target task nodes, the second score of the j-th second content to be summarized is determined, that is, based on the number of first candidate task nodes h and the number of first target task nodes r, the second score of the j-th second content to be summarized is determined.

[0097] For example, the server obtains the first weight corresponding to the first candidate task node and the second weight corresponding to the first target task node; based on the first and second weights, h and r are weighted respectively to obtain the target quantity. Finally, the second score of the j-th second content to be summarized is determined based on the target quantity.

[0098] It should be understood that the more task nodes associated with a particular piece of secondary content to be summarized, the more information a user obtains related to that content when reading the interactive task, resulting in a deeper impression of the summary. Furthermore, the more task nodes the user needs to interact with among these associated nodes, the deeper the user's impression of the secondary content to be summarized. Therefore, the higher the secondary score of the secondary content to be summarized, the stronger the user's memory of it; conversely, the lower the secondary score, the weaker the user's memory. Thus, secondary content with lower secondary scores can be summarized by the server, reducing the difficulty for users and improving the accuracy of their summaries.

[0099] Specifically, the occurrence time of the task plot corresponding to each group of content to be summarized is taken as the occurrence time of each group of content to be summarized. Following the order of occurrence time from front to back, the second content to be summarized with a second score less than a second threshold is selected from each group of content to be summarized as the third content to be summarized, thus obtaining the fourth number of third content to be summarized. That is, first, the multiple groups of content to be summarized are sorted according to the order of occurrence time from front to back; then, starting from the first group of content to be summarized, the second content to be summarized with a second score less than a second threshold is selected as the third content to be summarized; if the number of third content to be summarized selected from the first group of content to be summarized is less than the fourth number, the selection continues from the second group of content to be summarized, and so on, until the fourth number of third content to be summarized is selected.

[0100] For example, if a task node requires summarizing information including location, time, and items carried, and most of the preceding task events focus on discussing location and time with the user, with only a small portion discussing items, then the user will have a higher level of engagement with the summary information related to location and time than with items. However, if the target task node requires the user to relay previously determined information to B in the interactive story, the server can handle the item summary, allowing the user to then summarize the specific location and time based on the preceding task events.

[0101] 607: The server sends a second prompt statement to the client.

[0102] The second prompt statement includes a statement for expressing the third content to be summarized, and the second prompt statement is used to prompt the server to summarize the fourth quantity of the third content to be summarized.

[0103] For example, after the server determines the fourth number of third items to be summarized, it can send a second prompt statement to the user, so that the server can summarize the fourth number of third items to be summarized.

[0104] It should be understood that the server can send a second prompt statement to the user after the user has summarized the information, i.e., after obtaining the user's target summary. In this way, the server can supplement the user's summary, making it easier for the user to recall all the previous events. Alternatively, the server can send a second prompt statement to the user before summarizing the information. In this way, the server can first explain some of the information to be summarized, making it easier for the user to associate and recall the information they want to summarize based on this information, thus facilitating the user to complete the summary.

[0105] See Figure 7 , Figure 7 This application provides a functional unit block diagram of a server. The server 700 includes: a transceiver unit 701 and a processing unit 702, wherein:

[0106] Processing unit 702 is configured to, after interacting with the user terminal at the i-th task node, determine whether it is necessary to summarize the task segments preceding the i-th task node, wherein the i-th task node is any one of the plurality of task nodes; if so, control transceiver unit 701 to conduct one or more human-computer dialogues with the user terminal to obtain from the user terminal the target summary content for summarizing the task segments preceding the i-th task node, wherein the target summary content includes a first content to be summarized corresponding to the i-th task node, the first content to be summarized being the content summarized by the user among all the content that needs to be summarized at the i-th task node; perform semantic analysis on the target summary content to obtain a first score, wherein the first score is used to characterize the logical rigor of the user when summarizing the target summary content.

[0107] The transceiver unit 701 is configured to send a preset prompt statement to the user terminal if the first score is less than a first threshold. The preset prompt statement includes a target statement for expressing the first content to be summarized. The preset prompt statement is used to prompt the user to imitate the target statement in order to fully summarize the first content to be summarized.

[0108] In one embodiment of this application, in order to obtain target summary content from the user terminal for summarizing task plots located before the i-th task node by engaging in one or more human-computer dialogues with the user terminal, the processing unit 702 is specifically used for:

[0109] The user's first voice information is obtained from the user terminal, wherein the first voice information is the user's voice summarizing the task plots located before the i-th task node;

[0110] Based on the first voice information, determine the user's first summary content;

[0111] Determine whether the first summary content includes the first content to be summarized;

[0112] If so, then the first summary content shall be used as the target summary content;

[0113] If not, the control transceiver unit 701 sends a first prompt statement to the user terminal, wherein the first prompt statement includes a statement for expressing the second summary content, the second summary content being the summary content of the first content to be summarized excluding the first summary content, and the first prompt statement is used to prompt the user to perform a complete summary.

[0114] The control transceiver unit 701 obtains the user's second voice information from the user terminal, wherein the second voice information is the user's voice summarizing the task details before the i-th task node;

[0115] Based on the second voice information, the second summary content is obtained;

[0116] Determine whether the second summary content includes the first summary content;

[0117] If yes, then the second summary content is used as the target summary content; if no, then the transceiver unit 701 is controlled to send a new prompt statement to the user terminal, and the transceiver unit 701 is controlled to obtain new voice information of the user from the user terminal to obtain new summary content, until the new summary content includes the first summary content, and the new summary content is used as the target summary content.

[0118] In one embodiment of this application, before engaging in one or more human-computer dialogues with the user terminal to obtain the target summary content from the user terminal summarizing the task plots located before the i-th task node, the processing unit 702 is further configured to:

[0119] Obtain the user's age and historical interaction records;

[0120] Based on the user's age, a first number of users is determined, wherein the first number is the number of contents that can be summarized by the age;

[0121] Based on the historical interaction records, determine the user's actual aggregation capability;

[0122] The first quantity is adjusted based on the actual aggregation capability to obtain a second quantity, wherein the second quantity is the amount of content that the user can actually aggregate;

[0123] Based on the second quantity and the third quantity, a fourth quantity of the third content to be summarized is determined, wherein the third quantity is the quantity of the second content to be summarized, the second content to be summarized is all the content that needs to be summarized at the i-th task node, and the first content to be summarized is the content of the second content to be summarized excluding the third content to be summarized.

[0124] The fourth number of third items to be summarized is selected from the third number of second items to be summarized.

[0125] The control transceiver unit 701 sends a second prompt statement to the user terminal, wherein the second prompt statement includes a statement for expressing the third content to be summarized, and the second prompt statement is used to prompt the server to summarize the fourth quantity of the third content to be summarized.

[0126] In one embodiment of this application, in selecting the fourth number of third items to be summarized from the third number of second items to be summarized, the processing unit 702 is specifically configured to:

[0127] Determine the task segment containing each of the second items to be summarized;

[0128] Based on the task scenario in which each second item to be summarized is located, determine whether the second number of second items to be summarized appear in different task scenarios;

[0129] If the second number of second items to be summarized appear in completely different task plots, then the occurrence time of the task plot in which each second item to be summarized appears shall be taken as the occurrence time of each second item to be summarized, and the fourth number of third items to be summarized shall be selected from the third number of second items to be summarized in chronological order of the occurrence time of each second item to be summarized.

[0130] If at least two of the second items to be summarized in the second quantity appear in the same task plot, then the second items to be summarized in the second quantity that appear in the same task plot are assigned to a group to obtain multiple groups of items to be summarized, and the fourth quantity of third items to be summarized is selected from the multiple groups of items to be summarized.

[0131] In one embodiment of this application, in selecting the fourth number of third items to be summarized from the plurality of sets of items to be summarized, the processing unit 702 is specifically used for:

[0132] For the j-th second content to be summarized in the k-th group of content to be summarized, obtain the second score of the j-th second content to be summarized. The second score is used to characterize the probability that the user remembers the j-th second content to be summarized after performing the task plot in which the j-th second content to be summarized is located. The k-th group of content to be summarized is any one of the multiple groups of content to be summarized, and the j-th second content to be summarized is any one of the second content to be summarized in the k-th group of content to be summarized.

[0133] Based on the second score of the j-th second content to be summarized, determine the second score of each second content to be summarized in each group of content to be summarized.

[0134] The occurrence time of the task plot corresponding to each group of content to be summarized is taken as the occurrence time of each group of content to be summarized.

[0135] According to the order of appearance time of each group of content to be summarized, the second content to be summarized with a second score less than the second threshold is selected from each group of content to be summarized as the third content to be summarized, thus obtaining the fourth number of third content to be summarized.

[0136] In one embodiment of this application, in obtaining the second score of the j-th second content to be summarized, the processing unit 702 is specifically used for:

[0137] Obtain h first candidate task nodes that are related to the j-th second content to be summarized from the task nodes located before the i-th task node;

[0138] From the h first candidate task nodes, r first target task nodes are selected, wherein the r first target task nodes are the task nodes that require user interaction from the h first candidate task nodes;

[0139] Based on the h first candidate task nodes and the r first target task nodes, determine the second score of the j-th second content to be summarized.

[0140] In one embodiment of this application, the processing unit 702 is further configured to:

[0141] If it is not necessary to summarize the task plots located before the i-th task node, then obtain q second candidate task nodes located after the i-th task node, wherein the number of task nodes between any second candidate task node and the i-th task node is less than or equal to a third threshold.

[0142] Determine whether there is a second target task node among the q second candidate task nodes, wherein the second target task node is the task node among the q second candidate task nodes that requires the user to summarize the task details;

[0143] If they exist, then determine whether there is a correlation between the task plot corresponding to the i-th task node and the fourth content to be summarized corresponding to the second target task node, wherein the fourth content to be summarized corresponding to the second target task node is the content summarized by the user among all the content that needs to be summarized at the second target task node;

[0144] If so, before proceeding to the i-th task node, the control transceiver unit 701 sends a third prompt statement to the user terminal, wherein the third prompt statement is used to prompt that there is content that needs to be summarized in the task plot corresponding to the i-th task node;

[0145] After completing the i-th task node, if an exit command is received from the user terminal, the number of task nodes between the i-th task node and the second target task node is obtained, and the transceiver unit 701 is controlled to send a fourth prompt statement to the user terminal. The exit command is used to request to exit the current interactive task, and the fourth prompt statement is used to prompt the user to complete x task nodes before exiting the current interactive task, so that the user completes the second target task node before exiting the current interactive task, where x is the number of task nodes between the i-th task node and the second target task node.

[0146] See Figure 8 , Figure 8 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Figure 8 As shown, the electronic device 800 includes a transceiver 801, a processor 802, and a memory 803. These are connected via a bus 804. The memory 803 stores computer programs and data, and can transfer data stored in the memory 803 to the processor 802.

[0147] Processor 802 is used to read the computer program in memory 803 and perform the following operations:

[0148] After reaching the i-th task node with the user terminal, it is determined whether it is necessary to summarize the task segments preceding the i-th task node, where the i-th task node is any one of the plurality of task nodes. If so, the transceiver 801 is controlled to conduct one or more human-computer interactions with the user terminal to obtain the target summary content for summarizing the task segments preceding the i-th task node from the user terminal. The target summary content includes the first content to be summarized corresponding to the i-th task node, which is the content summarized by the user among all the content that needs to be summarized at the i-th task node. Semantic analysis is performed on the target summary content to obtain a first score, whereby the first score is used to characterize the logical rigor of the user's summary of the target summary content.

[0149] If the first score is less than the first threshold, the transceiver 801 sends a preset prompt statement to the user terminal. The preset prompt statement includes a target statement for expressing the first content to be summarized. The preset prompt statement is used to prompt the user to imitate the target statement in order to fully summarize the first content to be summarized.

[0150] In one embodiment of this application, in order to obtain target summary content from the user terminal for summarizing task plots located before the i-th task node by engaging in one or more human-computer dialogues with the user terminal, the processor 802 is specifically configured to perform the following operations:

[0151] The user's first voice information is obtained from the user terminal, wherein the first voice information is the user's voice summarizing the task plots located before the i-th task node;

[0152] Based on the first voice information, determine the user's first summary content;

[0153] Determine whether the first summary content includes the first content to be summarized;

[0154] If so, then the first summary content shall be used as the target summary content;

[0155] If not, the transceiver 801 sends a first prompt statement to the user terminal, wherein the first prompt statement includes a statement for expressing the second summary content, the second summary content being the summary content of the first content to be summarized excluding the first summary content, and the first prompt statement is used to prompt the user to perform a complete summary.

[0156] The transceiver 801 controls the acquisition of the user's second voice information from the user terminal, wherein the second voice information is the user's voice summarizing the task details before the i-th task node;

[0157] Based on the second voice information, the second summary content is obtained;

[0158] Determine whether the second summary content includes the first summary content;

[0159] If yes, then the second summary content is used as the target summary content; if no, then the transceiver 801 is controlled to send a new prompt statement to the user terminal, and the transceiver 801 is controlled to obtain new voice information of the user from the user terminal to obtain new summary content, until the new summary content includes the first summary content, and the new summary content is used as the target summary content.

[0160] In one embodiment of this application, before engaging in one or more human-computer interactions with the user terminal to obtain the target summary content from the user terminal summarizing the task plots located before the i-th task node, the processor 802 is further configured to perform the following operations:

[0161] Obtain the user's age and historical interaction records;

[0162] Based on the user's age, a first number of users is determined, wherein the first number is the number of contents that can be summarized by the age;

[0163] Based on the historical interaction records, determine the user's actual aggregation capability;

[0164] The first quantity is adjusted based on the actual aggregation capability to obtain a second quantity, wherein the second quantity is the amount of content that the user can actually aggregate;

[0165] Based on the second quantity and the third quantity, a fourth quantity of the third content to be summarized is determined, wherein the third quantity is the quantity of the second content to be summarized, the second content to be summarized is all the content that needs to be summarized at the i-th task node, and the first content to be summarized is the content of the second content to be summarized excluding the third content to be summarized.

[0166] The fourth number of third items to be summarized is selected from the third number of second items to be summarized.

[0167] The transceiver 801 sends a second prompt statement to the user terminal, wherein the second prompt statement includes a statement for expressing the third content to be summarized, and the second prompt statement is used to prompt the server to summarize the fourth quantity of the third content to be summarized.

[0168] In one embodiment of this application, in selecting the fourth number of third items to be summarized from the third number of second items to be summarized, the processor 802 is specifically configured to perform the following operations:

[0169] Determine the task segment containing each of the second items to be summarized;

[0170] Based on the task scenario in which each second item to be summarized is located, determine whether the second number of second items to be summarized appear in different task scenarios;

[0171] If the second number of second items to be summarized appear in completely different task plots, then the occurrence time of the task plot in which each second item to be summarized appears shall be taken as the occurrence time of each second item to be summarized, and the fourth number of third items to be summarized shall be selected from the third number of second items to be summarized in chronological order of the occurrence time of each second item to be summarized.

[0172] If at least two of the second items to be summarized in the second quantity appear in the same task plot, then the second items to be summarized in the second quantity that appear in the same task plot are assigned to a group to obtain multiple groups of items to be summarized, and the fourth quantity of third items to be summarized is selected from the multiple groups of items to be summarized.

[0173] In one embodiment of this application, in selecting the fourth number of third items to be summarized from the plurality of sets of items to be summarized, the processor 802 is specifically configured to perform the following operations:

[0174] For the j-th second content to be summarized in the k-th group of content to be summarized, obtain the second score of the j-th second content to be summarized. The second score is used to characterize the probability that the user remembers the j-th second content to be summarized after performing the task plot in which the j-th second content to be summarized is located. The k-th group of content to be summarized is any one of the multiple groups of content to be summarized, and the j-th second content to be summarized is any one of the second content to be summarized in the k-th group of content to be summarized.

[0175] Based on the second score of the j-th second content to be summarized, determine the second score of each second content to be summarized in each group of content to be summarized.

[0176] The occurrence time of the task plot corresponding to each group of content to be summarized is taken as the occurrence time of each group of content to be summarized.

[0177] According to the order of appearance time of each group of content to be summarized, the second content to be summarized with a second score less than the second threshold is selected from each group of content to be summarized as the third content to be summarized, thus obtaining the fourth number of third content to be summarized.

[0178] In one embodiment of this application, in obtaining the second score of the j-th second content to be summarized, the processor 802 is specifically configured to perform the following operations:

[0179] Obtain h first candidate task nodes that are related to the j-th second content to be summarized from the task nodes located before the i-th task node;

[0180] From the h first candidate task nodes, r first target task nodes are selected, wherein the r first target task nodes are the task nodes that require user interaction from the h first candidate task nodes;

[0181] Based on the h first candidate task nodes and the r first target task nodes, determine the second score of the j-th second content to be summarized.

[0182] In one embodiment of this application, the processor 802 is further configured to perform the following operations:

[0183] If it is not necessary to summarize the task plots located before the i-th task node, then obtain q second candidate task nodes located after the i-th task node, wherein the number of task nodes between any second candidate task node and the i-th task node is less than or equal to a third threshold.

[0184] Determine whether there is a second target task node among the q second candidate task nodes, wherein the second target task node is the task node among the q second candidate task nodes that requires the user to summarize the task details;

[0185] If they exist, then determine whether there is a correlation between the task plot corresponding to the i-th task node and the fourth content to be summarized corresponding to the second target task node, wherein the fourth content to be summarized corresponding to the second target task node is the content summarized by the user among all the content that needs to be summarized at the second target task node;

[0186] If so, before proceeding to the i-th task node, the transceiver 801 is controlled to send a third prompt statement to the user terminal, wherein the third prompt statement is used to prompt that there is content that needs to be summarized in the task plot corresponding to the i-th task node;

[0187] After completing the i-th task node, if an exit command is received from the user terminal, the number of task nodes between the i-th task node and the second target task node is obtained, and the transceiver 801 is controlled to send a fourth prompt statement to the user terminal. The exit command is used to request to exit the current interactive task, and the fourth prompt statement is used to prompt the user to complete x task nodes before exiting the current interactive task, so that the user completes the second target task node before exiting the current interactive task, where x is the number of task nodes between the i-th task node and the second target task node.

[0188] Specifically, the aforementioned electronic device 800 can be the aforementioned server 700, and the aforementioned transceiver 801 can be... Figure 7 The transceiver unit 701 of the server 700 in the embodiment described above, and the processor 802 can be... Figure 7 The processing unit 702 of the server 700 in the embodiment described above.

[0189] This application also provides a computer-readable storage medium storing a computer program that is executed by a processor to implement some or all of the steps of a summarizing method for any of the interactive tasks described in the above method embodiments.

[0190] This application also provides a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program operable to cause a computer to perform some or all of the steps of a summary method for any of the interactive tasks described in the above method embodiments.

[0191] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, as some steps may be performed in other orders or simultaneously according to this application. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are all optional embodiments, and the actions and modules involved are not necessarily essential to this application.

[0192] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0193] In the several embodiments provided in this application, it should be understood that the disclosed apparatus can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical or other forms.

[0194] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0195] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software program module.

[0196] If the integrated unit is implemented as a software program module and sold or used as an independent product, it can be stored in a computer-readable storage device (CMD). Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a memory and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned memory includes various media capable of storing program code, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.

[0197] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, which may include: flash drive, read-only memory (ROM), random access memory (RAM), disk or optical disk, etc.

[0198] The embodiments of this application have been described in detail above. Specific examples have been used to illustrate the principles and implementation methods of this application. The description of the above embodiments is only for the purpose of helping to understand the method and core ideas of this application. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this application. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for summarizing interactive tasks, characterized in that, The method is applied to a server, and the interactive task includes multiple task episodes and multiple task nodes, wherein the multiple task episodes and multiple task nodes correspond one-to-one, and the multiple task episodes and multiple task nodes are arranged in chronological order; the method includes: After reaching the i-th task node with the user, determine whether it is necessary to summarize the task segments located before the i-th task node, wherein the i-th task node is any one of the plurality of task nodes; If so, one or more human-computer dialogues are conducted with the user terminal to obtain the target summary content of the user's summary of the task plots located before the i-th task node from the user terminal, wherein the target summary content includes the first content to be summarized corresponding to the i-th task node, and the first content to be summarized is the content summarized by the user among all the content that needs to be summarized at the i-th task node. Semantic analysis is performed on the target summary content to obtain a first score, wherein the first score is used to characterize the logical rigor of the user when summarizing the target summary content; If the first score is less than the first threshold, a preset prompt statement is sent to the user terminal. The preset prompt statement includes a target statement for expressing the first content to be summarized. The preset prompt statement is used to prompt the user to imitate the target statement in order to fully summarize the first content to be summarized.

2. The method according to claim 1, characterized in that, The step of engaging in one or more human-computer dialogues with the user terminal to obtain target summary content from the user terminal summarizing the task details preceding the i-th task node includes: The user's first voice information is obtained from the user terminal, wherein the first voice information is the user's voice summarizing the task plots located before the i-th task node; Based on the first voice information, determine the user's first summary content; Determine whether the first summary content includes the first content to be summarized; If so, then the first summary content shall be used as the target summary content; If not, send a first prompt statement to the user terminal, wherein the first prompt statement includes a statement for expressing the second summary content, the second summary content being the summary content of the first content to be summarized excluding the first summary content, and the first prompt statement is used to prompt the user to perform a complete summary; The user's second voice information is obtained from the user terminal, wherein the second voice information is the user's voice summarizing the task plots located before the i-th task node; Based on the second voice information, the second summary content is obtained; Determine whether the second summary content includes the first summary content; If yes, then the second summary content is used as the target summary content; if not, then a new prompt statement is sent to the user terminal, and new voice information of the user is obtained from the user terminal to obtain new summary content, until the new summary content includes the first summary content, and the new summary content is used as the target summary content.

3. The method according to claim 1 or 2, characterized in that, Before engaging in one or more human-computer dialogues with the user terminal to obtain the target summary content from the user terminal summarizing the task plots located before the i-th task node, the method further includes: Obtain the user's age and historical interaction records; Based on the user's age, a first number of users is determined, wherein the first number is the number of contents that can be summarized by the age; Based on the historical interaction records, determine the user's actual aggregation capability; The first quantity is adjusted based on the actual aggregation capability to obtain a second quantity, wherein the second quantity is the amount of content that the user can actually aggregate; Based on the second quantity and the third quantity, a fourth quantity of the third content to be summarized is determined, wherein the third quantity is the quantity of the second content to be summarized, the second content to be summarized is all the content that needs to be summarized at the i-th task node, and the first content to be summarized is the content of the second content to be summarized excluding the third content to be summarized. The fourth number of third items to be summarized is selected from the third number of second items to be summarized. A second prompt statement is sent to the user terminal, wherein the second prompt statement includes a statement for expressing the third content to be summarized, and the second prompt statement is used to prompt the server to summarize the fourth quantity of the third content to be summarized.

4. The method according to claim 3, characterized in that, The step of filtering the fourth number of third items to be summarized from the third number of second items to be summarized includes: Determine the task segment containing each of the second items to be summarized; Based on the task scenario in which each second item to be summarized is located, determine whether the second number of second items to be summarized appear in different task scenarios; If the second number of second items to be summarized appear in completely different task plots, then the occurrence time of the task plot in which each second item to be summarized appears shall be taken as the occurrence time of each second item to be summarized, and the fourth number of third items to be summarized shall be selected from the third number of second items to be summarized in chronological order of the occurrence time of each second item to be summarized. If at least two of the second items to be summarized in the second quantity appear in the same task plot, then the second items to be summarized in the second quantity that appear in the same task plot are assigned to a group to obtain multiple groups of items to be summarized, and the fourth quantity of third items to be summarized is selected from the multiple groups of items to be summarized.

5. The method according to claim 4, characterized in that, The step of selecting the fourth number of third items to be summarized from the multiple sets of items to be summarized includes: For the j-th second content to be summarized in the k-th group of content to be summarized, obtain the second score of the j-th second content to be summarized. The second score is used to characterize the probability that the user remembers the j-th second content to be summarized after performing the task plot in which the j-th second content to be summarized is located. The k-th group of content to be summarized is any one of the multiple groups of content to be summarized, and the j-th second content to be summarized is any one of the second content to be summarized in the k-th group of content to be summarized. Based on the second score of the j-th second content to be summarized, determine the second score of each second content to be summarized in each group of content to be summarized. The occurrence time of the task plot corresponding to each group of content to be summarized is taken as the occurrence time of each group of content to be summarized. According to the order of appearance time of each group of content to be summarized, the second content to be summarized with a second score less than the second threshold is selected from each group of content to be summarized as the third content to be summarized, thus obtaining the fourth number of third content to be summarized.

6. The method according to claim 5, characterized in that, The process of obtaining the second score for the j-th second content to be summarized includes: Obtain h first candidate task nodes that are related to the j-th second content to be summarized from the task nodes located before the i-th task node; From the h first candidate task nodes, r first target task nodes are selected, wherein the r first target task nodes are the task nodes that require user interaction from the h first candidate task nodes; Based on the h first candidate task nodes and the r first target task nodes, determine the second score of the j-th second content to be summarized.

7. The method according to claim 6, characterized in that, The method further includes: If it is not necessary to summarize the task plots located before the i-th task node, then obtain q second candidate task nodes located after the i-th task node, wherein the number of task nodes between any second candidate task node and the i-th task node is less than or equal to a third threshold. Determine whether there is a second target task node among the q second candidate task nodes, wherein the second target task node is the task node among the q second candidate task nodes that requires the user to summarize the task details; If they exist, then determine whether there is a correlation between the task plot corresponding to the i-th task node and the fourth content to be summarized corresponding to the second target task node, wherein the fourth content to be summarized corresponding to the second target task node is the content summarized by the user among all the content that needs to be summarized at the second target task node; If so, before proceeding to the i-th task node, a third prompt statement is sent to the user terminal, wherein the third prompt statement is used to prompt that there is content that needs to be summarized in the task plot corresponding to the i-th task node; After completing the i-th task node, if an exit command is received from the user terminal, the number of task nodes between the i-th task node and the second target task node is obtained, and a fourth prompt statement is sent to the user terminal. The exit command is used to request to exit the current interactive task, and the fourth prompt statement is used to prompt the user to complete x task nodes before exiting the current interactive task, so that the user completes the second target task node before exiting the current interactive task, where x is the number of task nodes between the i-th task node and the second target task node.

8. A server, characterized in that, The server is used to implement interactive tasks, which include multiple task plots and multiple task nodes, wherein the multiple task plots and multiple task nodes correspond one-to-one and are arranged in chronological order; the server includes a transceiver unit and a processing unit. The processing unit is configured to, after interacting with the user terminal at the i-th task node, determine whether it is necessary to summarize the task segments preceding the i-th task node, wherein the i-th task node is any one of the plurality of task nodes; if so, control the transceiver unit to conduct one or more human-computer interactions with the user terminal to obtain from the user terminal the target summary content for summarizing the task segments preceding the i-th task node, wherein the target summary content includes the first content to be summarized corresponding to the i-th task node, and the first content to be summarized is the content summarized by the user among all the content that needs to be summarized at the i-th task node. Semantic analysis is performed on the target summary content to obtain a first score, wherein the first score is used to characterize the logical rigor of the user when summarizing the target summary content; The transceiver unit is configured to send a preset prompt statement to the user terminal if the first score is less than a first threshold. The preset prompt statement includes a target statement for expressing the first content to be summarized. The preset prompt statement is used to prompt the user to imitate the target statement in order to fully summarize the first content to be summarized.

9. An electronic device, characterized in that, include: A processor and a memory, the processor being connected to the memory, the memory being used to store a computer program, and the processor being used to execute the computer program stored in the memory to cause the electronic device to perform the method as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that is executed by a processor to implement the method as described in any one of claims 1-7.

Images