Data analysis method and device based on multi-round questions and answers, equipment, medium and product

Abstract

The invention discloses a data analysis method and device based on multiple rounds of questions and answers, equipment, a medium and a product, and relates to the field of data processing technologies, artificial intelligence technologies, large model technologies and large language models.The data analysis method comprises the steps that a data analysis page is displayed, and the data analysis page comprises a dialogue area, the dialogue area displays at least one first question and answer pair, the first question and answer pair has a corresponding grouping label, a first sub-area of the dialogue area is used for displaying the grouping label, and the grouping label is determined based on a semantic analysis result of the first question and answer pair; and in response to a selection instruction for the grouping tag in the first sub-region, determining a first grouping tag, and displaying a first question and answer pair corresponding to the first grouping tag on the dialogue region. The problem of low historical content query efficiency in data analysis application can be solved.

Description

Technical Field

[0001] The present disclosure relates to the fields of data processing technology, artificial intelligence technology, large model technology, and large language model technology, and particularly relates to a data analysis method, apparatus, device, medium, and product based on multi-round question and answer. Background Art

[0002] Currently, mainstream data analysis applications mainly use language models to answer users' questions. During a conversation, users may continuously diverge their questions or drill down on the same question. After multiple rounds of conversation, the vertical conversation flow becomes very long. If a user wants to locate historical content, they need to scroll up the page and continuously search for the required content, resulting in a low efficiency of querying historical content. Summary of the Invention

[0003] In view of this, the present disclosure provides a data analysis method, apparatus, electronic device, storage medium, and program product based on multi-round question and answer to solve the problem of low efficiency in querying historical content in data analysis applications.

[0004] In a first aspect, the present disclosure provides a data analysis method based on multi-round question and answer, the method comprising:

[0005] Display a data analysis page, the data analysis page including a conversation area, the conversation area displaying at least one first question-and-answer pair, the first question-and-answer pair having a corresponding grouping label, a first sub-area of the conversation area being used to display the grouping label, the grouping label being determined based on a semantic analysis result of the first question-and-answer pair;

[0006] In response to a selection instruction for the grouping label in the first sub-area, determine a first grouping label and display, on the conversation area, the first question-and-answer pair corresponding to the first grouping label.

[0007] In a second aspect, the present disclosure provides a data analysis apparatus based on multi-round question and answer, the apparatus comprising:

[0008] A first display module, configured to display a data analysis page, the data analysis page including a conversation area, the conversation area displaying at least one first question-and-answer pair, the first question-and-answer pair having a corresponding grouping label, a first sub-area of the conversation area being used to display the grouping label, the grouping label being determined based on a semantic analysis result of the first question-and-answer pair;

[0009] A second display module, configured to, in response to a selection instruction for the grouping label in the first sub-area, determine a first grouping label and display, on the conversation area, the first question-and-answer pair corresponding to the first grouping label.

[0010] In a third aspect, the present disclosure provides an electronic device, including: a memory and a processor, which are communicatively connected to each other. The memory stores computer instructions, and the processor executes the computer instructions to perform the data analysis method based on multi-round Q&A according to the first aspect or any corresponding embodiment thereof.

[0011] In a fourth aspect, the present disclosure provides a computer-readable storage medium, on which computer instructions are stored. The computer instructions are used to cause a computer to perform the data analysis method based on multi-round Q&A according to the first aspect or any corresponding embodiment thereof.

[0012] In a fifth aspect, the present disclosure provides a computer program product, including computer instructions, which are used to cause a computer to perform the data analysis method based on multi-round Q&A according to the first aspect or any corresponding embodiment thereof.

[0013] For the data analysis method based on multi-round Q&A provided by the embodiments of the present disclosure, corresponding grouping labels are configured for each first Q&A pair. While the first Q&A pair is displayed in the dialogue area of the data analysis page, the grouping label of the first Q&A pair is also displayed in the first sub-area in the dialogue area. Therefore, the user only needs to select the grouping label corresponding to the historical dialogue content to be searched in the first sub-area, and the first Q&A pairs corresponding to the grouping label can be displayed in the dialogue area, so as to quickly search for the required historical dialogue content, thereby improving the search efficiency of the historical dialogue content.

[0014] Regarding the beneficial effects of the data analysis device, electronic device, storage medium, and program product based on multi-round Q&A, they correspond to the beneficial effects of the data analysis method based on multi-round Q&A, and will not be elaborated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the specific embodiments of the present disclosure or the technical solutions in the prior art, the following will briefly introduce the drawings required for the description of the specific embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present disclosure. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0016] Figure 1 is a flowchart of a data analysis method based on multi-round Q&A according to an embodiment of the present disclosure;

[0017] Figure 2 is a schematic diagram of an initial page according to an embodiment of the present disclosure;

[0018] Figure 3 is a schematic diagram of a first dialogue area according to an embodiment of the present disclosure;

[0019] Figure 4 is a schematic diagram of a first sub-region according to an embodiment of the present disclosure;

[0020] Figure 5 is a schematic diagram of a second type of dialogue area according to an embodiment of the present disclosure;

[0021] Figure 6 is a schematic diagram of the display of description information and the number of question-and-answer pairs according to an embodiment of the present disclosure;

[0022] Figure 7 is a schematic diagram of an attribute page according to an embodiment of the present disclosure;

[0023] Figure 8 is a schematic diagram of a third type of dialogue area according to an embodiment of the present disclosure;

[0024] Figure 9 is a schematic diagram of a fourth type of dialogue area according to an embodiment of the present disclosure;

[0025] Figure 10 is a schematic diagram of a drawing board area according to an embodiment of the present disclosure;

[0026] Figure 11 is a schematic diagram of another drawing board area according to an embodiment of the present disclosure;

[0027] Figure 12 is a schematic diagram of historical analysis content and its title according to an embodiment of the present disclosure;

[0028] Figure 13 is a structural block diagram of a data analysis device based on multi-round question-and-answer according to an embodiment of the present disclosure;

[0029] Figure 14 is a structural block diagram of an electronic device according to an embodiment of the present disclosure. Detailed implementation manners

[0030] To make the objectives, technical solutions, and advantages of the embodiments of the present disclosure clearer, the technical solutions in the embodiments of the present disclosure will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present disclosure. Apparently, the described embodiments are some, but not all, of the embodiments of the present disclosure. All other embodiments obtained by those skilled in the art based on the embodiments of the present disclosure without creative efforts shall fall within the protection scope of the present disclosure.

[0031] It can be understood that before using the technical solutions disclosed in the embodiments of the present disclosure, the types, usage scopes, usage scenarios, etc. of the personal information involved in the present disclosure should be informed to users and user authorization should be obtained in an appropriate manner in accordance with relevant laws and regulations.

[0032] For example, when a user's active request is received, a prompt message is sent to the user to clearly prompt the user that the operation requested by the user will require obtaining and using the user's personal information. Thus, the user can autonomously choose whether to provide personal information to software or hardware such as an electronic device, an application, a server, or a storage medium that performs the operations of the present disclosure's technical solution based on the prompt message.

[0033] As an optional but non-limiting implementation manner, the manner of sending a prompt message to the user in response to receiving the user's active request may be, for example, in the form of a pop-up window. The prompt message may be presented in text in the pop-up window. In addition, the pop-up window may also carry a selection control for the user to choose "agree" or "disagree" to provide personal information to the electronic device.

[0034] It can be understood that the above notification and user authorization process is only illustrative and does not limit the implementation manner of the present disclosure. Other manners that comply with relevant laws and regulations can also be applied to the implementation manner of the present disclosure.

[0035] It can be understood that the data involved in the present technical solution (including but not limited to the data itself, the acquisition or use of the data) should comply with the requirements of corresponding laws, regulations, and related regulations.

[0036] Currently, mainstream data analysis applications mainly use language models to answer users' questions. During a conversation, a user may continuously diverge in their questions or drill down on the same question. After multiple rounds of conversation, the vertical conversation flow will become very long. If the user wants to locate historical content, they need to scroll up the page and continuously search for the required content, resulting in a low efficiency of querying historical content.

[0037] Moreover, when the user analyzes using the answer content of the language model, the user needs to copy the answer content. Then, paste the copied content into a document of a third-party tool for analysis. The analysis path is long, and during the analysis process, third-party tools need to be used crosswise.

[0038] In view of this, according to an embodiment of the present disclosure, an embodiment of a data analysis method based on multi-round question and answer is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in a different order than here.

[0039] In this embodiment, a data analysis method based on multi-round question and answer is provided, which can be used for data analysis applications. Figure 1It is a schematic flowchart of a data analysis method based on multi-round Q&A according to an embodiment of the present disclosure. As Figure 1 shown, the process includes the following steps:

[0040] Step S101, display a data analysis page. The data analysis page includes a dialogue area, and at least one first Q&A pair is displayed in the dialogue area. The first Q&A pair has a corresponding grouping label, and a first sub-area of the dialogue area is used to display the grouping label, and the grouping label is determined based on the semantic analysis result of the first Q&A pair.

[0041] Among them, the first Q&A pair includes a question and the answer content of the question.

[0042] Optionally, the first first Q&A pair can be obtained through the following steps: display an initial page; in response to the first question determined on the initial page, display a data analysis page, and display the first first Q&A pair on the dialogue area of the data analysis page. The first first Q&A pair includes the first question determined on the initial page and its answer content. Among them, the first question can be determined in response to the question entered in the input control on the initial page. Or, one or more recommended questions are displayed on the initial page, and in response to the selection instruction for the recommended question, the first question is determined.

[0043] Exemplarily, as Figure 2 shown, the user can enter Question 1 in the input control on the initial page, or select one of the recommended Question 1 and recommended Question 2 displayed on the initial page as the first question. The data analysis application displays a data analysis page in response to the first question determined on the initial page, and displays the first first Q&A pair on the dialogue area of the data analysis page. For example Figure 3 shown, Question 1 and the answer content of Question 1.

[0044] In addition, when using the initial page to determine the first question, an upload control is also displayed on the initial page. The upload control is used to upload a target file as the data source for data analysis. The first question can be analyzed through the target file to obtain the answer content of the first question.

[0045] In another alternative embodiment, the first first Q&A pair can also be obtained through the following steps: in response to the first question determined in the dialogue area, display the first first Q&A pair on the dialogue area. The first first Q&A pair includes the first question and its answer content. Among them, the first question can be determined in response to the question entered in the input control in the dialogue area. Or, one or more recommended questions are displayed in the dialogue area, and in response to the selection instruction for the recommended question, the first question is determined.

[0046] For other first question-and-answer pairs, they are obtained through the following steps: In response to the next question determined in the conversation area, display the next first question-and-answer pair in the conversation area, where the next first question-and-answer pair includes the next question and its answer content. Among them, the next question can be determined in response to the question entered in the input control in the conversation area. Or, corresponding to each first question-and-answer pair area in the conversation area, there are corresponding recommended questions, and in response to the selection instruction for the recommended question, the next question is determined.

[0047] It should be noted that the conversation area has a corresponding first language model, and the first language model is used to analyze each question to obtain the answer content of each question, so as to obtain the first question-and-answer pair.

[0048] In addition, the first language model can also be used to perform semantic analysis on each first question-and-answer pair to obtain the semantic analysis results of each first question-and-answer pair. Based on the semantic analysis results of the first question-and-answer pair, the grouping labels of each first question-and-answer pair are determined.

[0049] Specifically, the first language model performs semantic analysis on each first question-and-answer pair based on the questions and context information in each first question-and-answer pair to obtain the semantic analysis results of each first question-and-answer pair.

[0050] Exemplarily, as Figure 4 shown, Figure 4 there are two first question-and-answer pairs shown. One first question-and-answer pair includes question 1 and the answer content of question 1. The other first question-and-answer pair includes question 2 and the answer content of question 2. The grouping labels corresponding to the two first question-and-answer pairs are displayed in the first sub-region in the conversation area.

[0051] Optionally, in practical applications, in order to quickly locate the initial question (i.e., the first question), the grouping label of the first first question-and-answer pair is different from that of other question-and-answer pairs.

[0052] Step S102, in response to the selection instruction for the grouping label in the first sub-region, determine the first grouping label, and display the first question-and-answer pair corresponding to the first grouping label in the conversation area.

[0053] In practical applications, if non-text grouping labels are used, the description information of the grouping label can be displayed when the interaction point is at the corresponding position of the grouping label, so that the user can determine the grouping label of the first question-and-answer pair to be searched according to the description information.

[0054] The first question-and-answer pair corresponding to the first grouping label displayed in the conversation area is not limited to the current conversation turn. The first question-and-answer pair corresponding to the first grouping label can be displayed across turns, and continue the conversation under the first question-and-answer pair corresponding to the first grouping label. After restoring the global display (i.e., canceling the selection of the first grouping label), the conversation turns are still displayed in the order of question-and-answer time.

[0055] Exemplarily, as Figure 5 shown, assume that Figure 5 the second grouping label in the first sub-region shown corresponds to two first question-and-answer pairs. For example, the first question-and-answer pair composed of question 2 and its answer content, and the first question-and-answer pair composed of question 3 and its answer content. Then, when the second grouping label in the first sub-region is selected, question 2 and the first question-and-answer pair corresponding to question 3 are displayed.

[0056] For the data analysis method based on multi-round question and answer provided in this embodiment, corresponding grouping labels are configured for each first question-and-answer pair. While the first question-and-answer pair is displayed in the dialogue area of the data analysis page, the grouping label of the first question-and-answer pair is also displayed in the first sub-region in the dialogue area. Therefore, the user only needs to select the grouping label corresponding to the historical dialogue content to be searched in the first sub-region, and the first question-and-answer pair corresponding to the grouping label can be displayed in the dialogue area, so as to quickly search for the required historical dialogue content, thereby improving the search efficiency of the historical dialogue content.

[0057] In some optional implementation manners, the data analysis method based on multi-round question and answer of the present disclosure further includes: if the current interaction point is located at the corresponding position of any grouping label in the first sub-region, the number of question-and-answer pairs and the description information corresponding to the grouping label are displayed.

[0058] Wherein, the description information includes the dialogue topic characterized by the grouping label, and the dialogue topic is related to the question of the first question-and-answer pair under the grouping label.

[0059] Optionally, the dialogue topic of the grouping label is determined based on the semantic analysis result of the first question-and-answer pair under the grouping label. Specifically, the first language model is used to perform topic analysis on the first question-and-answer pair under the grouping label to obtain the dialogue topic corresponding to the grouping label.

[0060] It can be understood that when new question-and-answer pairs are continuously added subsequently, new grouping labels may be generated. If the question-and-answer pairs under the grouping label include the first question-and-answer pair, the dialogue topic corresponding to the grouping label is related to the question of the first first question-and-answer pair under the grouping label. If the question-and-answer pairs under the grouping label only include subsequently added question-and-answer pairs (such as the second question-and-answer pair), the dialogue topic corresponding to the grouping label is related to the question of the first newly added question-and-answer pair under the grouping label.

[0061] In addition, the user can also define the description information of the grouping label by himself / herself, and the generation method of the description information is not limited herein.

[0062] Exemplarily, as Figure 6As shown, when the current interaction point is at the corresponding position of the second grouping label in the first sub-region, the description information corresponding to the second grouping label and the number of Q&A pairs are displayed (see Figure 6 2) shown in

[0063] For the data analysis method based on multi-round Q&A provided in this embodiment, when the current interaction point is at the corresponding position of any grouping label in the first sub-region, the number of Q&A pairs and the description information corresponding to the grouping label are displayed. Therefore, it is convenient to quickly locate the grouping label of the historical conversation content to be searched.

[0064] In some alternative embodiments, grouping labels are displayed in the second sub-regions corresponding to each first Q&A pair in the conversation area. The data analysis method based on multi-round Q&A of the present disclosure further includes:

[0065] Step a1, in response to an interaction operation on the grouping label in any second sub-region, display an attribute page of the grouping label, and the attribute page is used to configure the display attributes of the grouping label.

[0066] The second sub-region corresponding to the first Q&A pair is the corresponding area of the question in the first Q&A pair. For example, it is the upper right area of the question in the first Q&A pair.

[0067] See Figure 7 , the attribute page includes configuration items for the display attributes of the grouping label. In response to a configuration operation on the configuration item, the display attributes of the grouping label are determined. Among them, the display attributes include at least one of the existence situation, color, and shape of the grouping label, and the display attributes can also be adjusted according to the actual situation.

[0068] In addition, the attribute page further includes existing grouping labels. In response to the selection of an existing grouping label in the attribute page, the display attributes of the grouping label of the first Q&A pair are determined according to the selected existing grouping label.

[0069] Step a2, in response to the display attributes determined in the attribute page, update the grouping label of the corresponding first Q&A pair.

[0070] Specifically, if the color of the grouping label changes, the color of the grouping label of the first Q&A pair is updated. If the shape of the grouping label changes, the shape of the grouping label of the first Q&A pair is updated. If the existence situation of the grouping label is to cancel the grouping label, the grouping label of the first Q&A pair is cancelled.

[0071] Understandably, the user can customize the color, shape of the grouped label or ungroup the label by configuring the display attributes of the grouped label in the property page. The first question-and-answer pair of the same grouped label will be divided into a group. If the grouped label is ungrouped, the corresponding first question-and-answer pair is considered ungrouped, indicating that the first question-and-answer pair is invalid content and will not be screened subsequently.

[0072] The data analysis method based on multi-round question and answer provided in this embodiment configures the display attributes of the grouped label of the first question-and-answer pair by using the property page. Therefore, the grouped label of the first question-and-answer pair can be flexibly adjusted, improving the flexibility of grouping the first question-and-answer pair.

[0073] In some optional embodiments, recommended questions of each first question-and-answer pair are displayed in the third sub-region corresponding to the first question-and-answer pair in the dialogue area. The data analysis method based on multi-round question and answer of the present disclosure further includes:

[0074] Step b1, in response to a selection instruction for a recommended question of any first question-and-answer pair, generate an analysis instruction and hide the selected recommended question in the corresponding third sub-region. The analysis instruction is used to analyze the selected recommended question to obtain the answer content of the selected recommended question, so as to obtain a second question-and-answer pair and the grouped label of the second question-and-answer pair. The grouped label of the second question-and-answer pair is determined based on the semantic analysis result of the second question-and-answer pair.

[0075] The third sub-region corresponding to the first question-and-answer pair is the corresponding region of the answer content in the first question-and-answer pair. For example, the region below the answer content in the first question-and-answer pair.

[0076] It should be noted that the second question-and-answer pair includes the selected recommended question and its answer content.

[0077] Specifically, use the first language model to analyze the selected recommended question to obtain the answer content of the selected recommended question, so as to obtain a second question-and-answer pair. And use the first language model to perform semantic analysis on the second question-and-answer pair to obtain the semantic analysis result of the second question-and-answer pair. Determine the grouped label of the second question-and-answer pair based on the semantic analysis result of the second question-and-answer pair.

[0078] It should be noted that at the end of the previous round of question and answer, the unselected recommended questions will not disappear, and the user can still select the recommended questions of the previous question-and-answer pair for follow-up questions to obtain new question-and-answer pairs.

[0079] Exemplarily, assume that the conversation area displays the first Q&A pair corresponding to question 1 and question 2, the recommended questions of the first Q&A pair corresponding to question 1 include recommended question 11 and recommended question 12, and the recommended questions of the first Q&A pair corresponding to question 2 include recommended question 21 and recommended question 22. The user can select any one of the recommended questions 11, 12, 21, and 22 as the next question according to the actual situation. Refer to Figure 8 , assume that recommended question 22 is selected, then an analysis instruction for recommended question 22 is generated to obtain the answer content of recommended question 22. The recommended question 22 and its answer content are displayed in the conversation area. At the same time, recommended question 22 is hidden in the third sub-area corresponding to question 2.

[0080] And recommended question 41 is also displayed in the corresponding area of the answer content of recommended question 22. The user can select recommended question 41 as the question for the next Q&A to obtain a new Q&A pair.

[0081] Step b2, display the second Q&A pair in the conversation area and display the grouping label of the second Q&A pair in the first sub-area.

[0082] It should be noted that as Figure 8 shown, if the first language model determines that the selected recommended question is a drill-down question, a new grouping label will be generated as the grouping label of the second Q&A pair. At this time, the grouping label of the second Q&A pair is displayed in the first sub-area. As Figure 9 shown, if the grouping label of the second Q&A pair is the same as any one of the grouping labels in the first sub-area, there is no need to repeat displaying the same grouping label. That is, only when the grouping label of the second Q&A pair is different from the grouping label in the first sub-area, the grouping label of the second Q&A pair is added and displayed in the first sub-area. If the grouping label of the second Q&A pair already exists in the first sub-area, there is no need to add the grouping label of the second Q&A pair to the first sub-area again.

[0083] In the data analysis method based on multi-round Q&A provided in this embodiment, the recommended questions of the first Q&A pair are displayed in the third sub-area corresponding to the first Q&A pair. Therefore, it is possible to use the recommended questions for follow-up questions, improving the data analysis efficiency. And after selecting a recommended question for follow-up, the selected recommended question is hidden. Therefore, it is possible to avoid repeating the same question and reduce ineffective questions.

[0084] In some alternative embodiments, the data analysis method based on multi-turn Q&A of the present disclosure further includes: in response to a target question input in an input control in a dialogue area, generating an analysis instruction for the target question, where the analysis instruction is used to analyze the target question to obtain an answer content for the target question, so as to obtain a third Q&A pair and a grouping label for the third Q&A pair, and the grouping label for the third Q&A pair is determined based on the semantic analysis result of the third Q&A pair. Display the third Q&A pair in the dialogue area, and display the grouping label for the third Q&A pair in a first sub-region.

[0085] In some alternative embodiments, the data analysis method based on multi-turn Q&A of the present disclosure further includes: in response to a deselection instruction for a first grouping label, displaying the first Q&A pair and the second Q&A pair in the dialogue area in the order of Q&A time.

[0086] In practical applications, the first grouping label in the first sub-region can be selected by clicking. When the first grouping label is clicked again, the first grouping label is deselected, and a deselection instruction for the first grouping label is generated.

[0087] After deselecting the first grouping label, the Q&A pairs of the current conversation are displayed in the order of Q&A time, so as to return to the normal conversation end. Among them, the Q&A pairs of the current conversation include the first Q&A pair and the second Q&A pair.

[0088] For the data analysis method based on multi-turn Q&A provided in this embodiment, after deselecting the first grouping label, each Q&A pair is displayed in the order of Q&A time. Therefore, the normal display of the Q&A pairs can be ensured.

[0089] In some alternative embodiments, a source identifier of the answer content in the Q&A pair is displayed in the corresponding area of the first Q&A pair and / or the second Q&A pair, so as to facilitate tracing the source of the answer content.

[0090] In some alternative embodiments, a new conversation control (see “+ New Conversation” in the attached drawing) is displayed in the dialogue area, and the new conversation control is used to start a new conversation.

[0091] In some alternative embodiments, a drawing board area is further displayed on the data analysis page, and the drawing board area is used to display the content of interest in the dialogue area and the association relationship between the content of interest.

[0092] Optionally, a first data unit corresponding to the question in the first first Q&A pair is displayed in the drawing board area.

[0093] As Figure 10 shown, assuming that the question in the first first Q&A pair is question 1, then the first data unit corresponding to question 1 is displayed in the drawing board area.

[0094] Optionally, in response to an adjustment instruction for the area size of the drawing board area or the dialogue area, adjust the area sizes of the drawing board area and the dialogue area.

[0095] In practical applications, the drawing board area can be used as the main canvas of the data analysis page, and the dialogue area is embedded in the drawing board area in the form of a panel, so that users can complete the whole process of data analysis on the same data analysis page, avoiding cross-using third-party tools for data analysis.

[0096] See Figure 10 As shown by the double arrow, the area sizes of the dialogue area and the drawing board area can be adjusted by stretching the panel of the dialogue area.

[0097] The data analysis method based on multi-round Q&A provided in this embodiment configures a drawing board area in the data analysis panel to display the content of interest in the dialogue area and the association relationship between the content of interest. Therefore, it is possible to use the drawing board area to retain the content of interest in the dialogue area, edit and organize the logical relationship between the content of interest for data analysis. Therefore, there is no need to copy the dialogue content to a document of a third-party tool for analysis, effectively shortening the analysis path and avoiding cross-using different data analysis tools to meet different data analysis needs.

[0098] In some alternative embodiments, a first analysis content is displayed in the drawing board area, and the first analysis content includes at least one first data unit having an association relationship. The first data unit is generated based on the first content of interest in the dialogue area. It should be noted that the first content of interest includes at least one of the question in the first Q&A pair and the historical selected content of the target Q&A pair in the dialogue area. In addition, the first analysis content may also include data units corresponding to other content, such as data units of copied pictures, texts, etc., which are not limited herein.

[0099] Furthermore, the data analysis method based on multi-round Q&A of the present disclosure further includes:

[0100] Step c1, in response to the selected content of the target Q&A pair in the dialogue area, determine the second content of interest. The target Q&A pair includes at least one of the first Q&A pair and the second Q&A pair, and the second Q&A pair is obtained based on the recommended question of the first Q&A pair.

[0101] Exemplarily, see Figure 11 , assuming that question 2 is "Introduce the content of Meeting A", and the answer content to this question is "Meeting A: Held from January 1st to 10th, 2025, and the meeting includes the computing power base and application implementation, etc.". If the user selects "Meeting A: Held from January 1st to 10th, 2025", then this content is used as the second content of interest.

[0102] Step c2, in response to the operation of moving the second content of interest to the drawing board area, display a second data unit at the position corresponding to the second content of interest in the drawing board area. The second data unit includes the second content of interest and a unit title, and the unit title is related to the question in the target question-and-answer pair.

[0103] Specifically, the second content of interest can be copied and moved to the drawing board area by means of dragging, copying, etc. It should be noted that when the second content of interest is moved to the drawing board area, the selected content in the dialogue area remains unchanged.

[0104] Optionally, the unit title is the question in the target question-and-answer pair, or the unit title is obtained by analyzing the question in the target question-and-answer pair.

[0105] Step c3, associate the second data unit with the first data unit in the drawing board area to obtain second analysis content.

[0106] Specifically, the drawing board area corresponds to a second language model, where the second language model can be the same as or different from the first language model. Use the second language model to associate the second data unit with the first data unit according to the distance between the second data unit and the first data unit in the first analysis content. Or, use the second language model to associate the second data unit with the first data unit according to the correlation between the second data unit and the first data unit in the first analysis content. Or, in response to the connection instruction between the second data unit and any first data unit, associate the second data unit with the corresponding first data unit to obtain second analysis content.

[0107] Optionally, the first analysis content and the second analysis content are mind maps.

[0108] Exemplarily, referring to Figure 11 , the second data unit can be connected to the first data unit corresponding to "Question 1" by a connecting line to establish an association relationship between the second data unit and the first data unit, thereby forming second analysis content.

[0109] In the data analysis method based on multi-round question and answer provided in this embodiment, when the second content of interest in the dialogue area is moved to the drawing board area, a second data unit corresponding to the second content of interest is generated in the drawing board area. Therefore, the content in the dialogue area can be explored in depth using the drawing board area. At the same time, the unit title of the second data unit is related to the question in the target question-and-answer pair. Therefore, the source of the second content of interest can be traced using the unit title, better understanding the context of the second content of interest, and improving the accuracy of data analysis.

[0110] In some alternative embodiments, the above step c3 includes:

[0111] Step c31: In the drawing board area, based on the distance between the first data unit and the second data unit, determine the first data unit associated with the second data unit.

[0112] Specifically, based on the distances between the second data unit and each first data unit, determine the first data unit with the shortest distance as the one associated with the second data unit.

[0113] Step c32: Associate the second data unit with the determined associated first data unit to obtain the second analysis content.

[0114] Specifically, establish a connection line between the second data unit and the determined associated first data unit to associate the second data unit with the corresponding first data unit and obtain the second analysis content.

[0115] Alternatively, if the second data unit is within the position range of any first data unit, then associate the first data unit with the second data unit.

[0116] The data analysis method based on multi-round Q&A provided in this embodiment automatically establishes the association relationship between the second data unit and the first data unit through distance. Therefore, it can improve the association efficiency between the second data unit and the first data unit, thereby improving the data analysis efficiency.

[0117] In some alternative embodiments, the data analysis method based on multi-round Q&A of the present disclosure further includes: in response to an editing operation on a target data unit in the drawing board area, update the target data unit, where the target data unit is the first data unit or the second data unit.

[0118] Specifically, the unit title or content of the target data unit (i.e., the content of interest recorded in the target data unit) can be edited, or the connection relationship between the target data unit and other data units can be edited to update the target data unit.

[0119] In addition, downstream data units of the target data unit can also be established according to actual needs to obtain new analysis content.

[0120] For the data analysis method based on multi-round Q&A provided in this embodiment, the target data unit is editable. Therefore, it can flexibly adjust the target data unit in the drawing board area according to the analysis idea to improve the flexibility of data analysis.

[0121] In some alternative embodiments, a collapse control is also displayed on the data analysis page, and the collapse control is used to trigger the display or collapse of the dialogue area.

[0122] In some alternative embodiments, the association relationship is a hierarchical attribution relationship, and the first data unit at the top layer is related to the question of the first first Q&A pair in the current conversation.

[0123] Optionally, the first data unit at the top layer includes the question in the first first Q&A pair, or includes the content analyzed based on the question in the first first Q&A pair.

[0124] Taking the first analysis content as a mind map as an example, the first data unit at the top layer is the central node of the first analysis content. After determining the question of the first first Q&A pair in the conversation area / initial page, synchronously in the drawing board area, the first data unit at the top layer is generated based on the question of the first first Q&A pair.

[0125] In the data analysis method based on multi-round Q&A provided in this embodiment, the current conversation usually starts around the first question. Therefore, the first question in the current conversation can often represent the core topic of the current conversation. Based on this question, the first data unit at the top layer is generated. Therefore, the core topic or analysis purpose of this analysis can be represented by the first data unit at the top layer.

[0126] In some alternative embodiments, a first control is also displayed in the drawing board area, and the first control is used to trigger the display of historical analysis content and the title of the historical analysis content.

[0127] Specifically, in response to an interaction operation on the first control, the historical analysis content and the title of the historical analysis content are displayed in the drawing board area or on a separate historical page.

[0128] In practical applications, editable units can be generated for each historical analysis content and its title. For example, Figure 12 the editable unit composed of the historical analysis content 1 and its title shown, and the editable unit composed of the historical analysis content 2 and its title. The corresponding historical analysis content can be accessed through interaction with the editable unit.

[0129] It should be noted that if there is a large amount of historical analysis content, only part of the historical analysis content can be displayed.

[0130] The title of the historical analysis content is related to the data unit at the top layer in the historical analysis content, or related to the first question in the conversation corresponding to the historical analysis content, or the title of the historical analysis content is user-defined.

[0131] The historical analysis content includes the analysis content historically displayed in the drawing board area.

[0132] The data analysis method based on multi-round question and answer provided in this embodiment sets a first control in the drawing board area to trigger the display of historical analysis content and the title of the historical analysis content. Therefore, the historical analysis content can be quickly queried.

[0133] In some alternative embodiments, while displaying the historical analysis content, a new control can also be displayed. The new control is used to create new conversations and / or analysis content.

[0134] In some alternative embodiments, a second control is also displayed in the drawing board area. The second control is used to export the content in the drawing board area.

[0135] Specifically, in response to an interaction operation on the second control, the content in the drawing board area is exported. For example, the second analysis content. Among them, the content in the drawing board area can be exported in formats such as documents and mind maps.

[0136] The data analysis method based on multi-round question and answer provided in this embodiment sets a second control in the drawing board area to export the content in the drawing board area. Therefore, it is convenient for data sharing and transmission to meet diverse data analysis needs.

[0137] As a specific application example, a data analysis application is installed on a computer. The data analysis application is used to execute the data analysis method based on multi-round question and answer of the present disclosure. A data analysis page is displayed in the data analysis application. The user can input questions in the conversation area of the data analysis page to perform multi-dimensional exploration in the form of a conversation, generating one or more question-and-answer pairs. Grouping labels of each question-and-answer pair are displayed in the first sub-area of the conversation area. If the user needs to query a certain content in the historical conversation, the user can click on the grouping label corresponding to the content in the first sub-area to filter out irrelevant question-and-answer pairs, and only display the question-and-answer pairs under the selected grouping label in the conversation area to find the content to be queried. At the same time, the user can drag the interesting content in the conversation area to the drawing board area of the data analysis page, and conduct exploratory analysis on the interesting content in the drawing board area. For example, edit and organize the logical relationships between different interesting contents to obtain analysis content. In addition, the analysis content can also be exported for use in other research and investigation scenarios.

[0138] In this embodiment, a data analysis device based on multi-round question and answer is also provided. The device is used to implement the above embodiments and preferred implementation manners, and those that have been described will not be repeated. As used below, the term "module" can be a combination of software and / or hardware that realizes a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, implementation in hardware, or a combination of software and hardware is also possible and contemplated.

[0139] This embodiment provides a data analysis device based on multi-round question and answer, such asFigure 13 As shown in, it includes:

[0140] A first display module 1301, configured to display a data analysis page, the data analysis page includes a conversation area, at least one first question-and-answer pair is displayed in the conversation area, the first question-and-answer pair has a corresponding grouping label, and a first sub-area of the conversation area is used to display the grouping label, and the grouping label is determined based on the semantic analysis result of the first question-and-answer pair;

[0141] A second display module 1302, configured to determine a first grouping label in response to a selection instruction for the grouping label in the first sub-area, and display the first question-and-answer pair corresponding to the first grouping label on the conversation area.

[0142] In some alternative embodiments, the data analysis device of the present disclosure further includes:

[0143] A third display module, configured to display the number of question-and-answer pairs corresponding to the grouping label and the description information if the current interaction point is located at the corresponding position of any grouping label in the first sub-area.

[0144] In some alternative embodiments, grouping labels are displayed in a second sub-area corresponding to each first question-and-answer pair in the conversation area. The data analysis device of the present disclosure further includes:

[0145] A fourth display module, configured to display an attribute page of the grouping label in response to an interaction operation on the grouping label in any second sub-area, and the attribute page is used to configure the display attributes of the grouping label;

[0146] A label update module, configured to update the grouping label of the corresponding first question-and-answer pair in response to the display attributes determined on the attribute page.

[0147] In some alternative embodiments, recommended questions of each first question-and-answer pair are displayed in a third sub-area corresponding to the first question-and-answer pair in the conversation area. The data analysis device of the present disclosure further includes:

[0148] A question selection module, configured to generate an analysis instruction and hide the selected recommended question in the corresponding third sub-area in response to a selection instruction for the recommended question of any first question-and-answer pair, and the analysis instruction is used to analyze the selected recommended question to obtain the answer content of the selected recommended question, so as to obtain a second question-and-answer pair and the grouping label of the second question-and-answer pair, and the grouping label of the second question-and-answer pair is determined based on the semantic analysis result of the second question-and-answer pair;

[0149] A fifth display module, configured to display the second question-and-answer pair on the conversation area and display the grouping label of the second question-and-answer pair in the first sub-area.

[0150] In some alternative embodiments, the data analysis device of the present disclosure further includes:

[0151] A sixth display module, configured to respond to a deselected instruction for the first grouping label and display the first Q&A pair and the second Q&A pair in the conversation area according to the Q&A time.

[0152] In some alternative embodiments, the data analysis page further displays a drawing board area, which is configured to display the content of interest in the conversation area and the association relationship between the content of interest.

[0153] In some alternative embodiments, a first analysis content is displayed in the drawing board area, and the first analysis content includes at least one first data unit having an association relationship, and the first data unit is generated based on the first content of interest in the conversation area. The data analysis device of the present disclosure further includes:

[0154] A content selection module, configured to determine a second content of interest in response to the selected content of the target Q&A pair in the conversation area, where the target Q&A pair includes at least one of the first Q&A pair and the second Q&A pair, and the second Q&A pair is a recommended question based on the first Q&A pair;

[0155] A content movement module, configured to display a second data unit at a position corresponding to the second content of interest in the drawing board area in response to an operation of moving the second content of interest to the drawing board area, where the second data unit includes the second content of interest and a unit title, and the unit title is related to the question in the target Q&A pair;

[0156] A unit association module, configured to associate the second data unit with the first data unit in the drawing board area to obtain a second analysis content.

[0157] In some alternative embodiments, the unit association module includes:

[0158] An operation unit, configured to determine a first data unit associated with the second data unit based on the distance between the first data unit and the second data unit in the drawing board area;

[0159] An association unit, configured to associate the second data unit with the determined first data unit to obtain a second analysis content.

[0160] In some alternative embodiments, the data analysis device of the present disclosure further includes:

[0161] A data editing module, configured to update the target data unit in response to an editing operation on the target data unit in the drawing board area, where the target data unit is the first data unit or the second data unit.

[0162] In some alternative embodiments, the association relationship is a hierarchical attribution relationship, and the first data unit at the top layer is related to the question of the first first question-and-answer pair in this conversation.

[0163] In some alternative embodiments, a first control is further displayed in the drawing board area, and the first control is used to trigger the display of historical analysis content and the title of the historical analysis content.

[0164] In some alternative embodiments, a second control is further displayed in the drawing board area, and the second control is used to export the content in the drawing board area.

[0165] The further function descriptions of the above-mentioned various modules and units are the same as those in the corresponding above-mentioned embodiments, and will not be elaborated here.

[0166] The data analysis device based on multi-round question and answer in this embodiment is presented in the form of functional units. Here, the unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and a memory that execute one or more software or fixed programs, and / or other devices that can provide the above functions.

[0167] The embodiments of the present disclosure further provide an electronic device having the above-mentioned Figure 13 shown data analysis device based on multi-round question and answer.

[0168] Please refer to Figure 14 , Figure 14 is a structural block diagram of an electronic device provided by an alternative embodiment of the present disclosure. As Figure 14 shown, the electronic device includes: one or more processors 1401, a memory 1402, and interfaces for connecting various components, including a high-speed interface and a low-speed interface. Each component communicates with each other using different buses and can be installed on a common main board or installed in other ways as needed. The processor can process instructions executed within the electronic device, including instructions stored in the memory or on the memory to display graphical information of the GUI on an external input / output device (such as a display device coupled to the interface). In some alternative embodiments, if necessary, multiple processors and / or multiple buses can be used together with multiple memories and multiple memories. Similarly, multiple electronic devices can be connected, and each device provides some necessary operations (for example, as a server array, a set of blade servers, or a multi-processor system). Figure 14 In

[0169] The processor 1401 can be a central processing unit, a network processor, or a combination thereof. Among them, the processor 1401 can further include a hardware chip. The above-mentioned hardware chip can be an application-specific integrated circuit, a programmable logic device, or a combination thereof. The above-mentioned programmable logic device can be a complex programmable logic device, a field-programmable gate array, a generic array logic, or any combination thereof.

[0170] Among them, the memory 1402 stores instructions executable by at least one processor 1401, so that at least one processor 1401 executes the method shown in the above embodiments.

[0171] The memory 1402 can include a program storage area and a data storage area. Among them, the program storage area can store an operating system and application programs required for at least one function; the data storage area can store data created according to the use of the electronic device, etc. In addition, the memory 1402 can include a high-speed random access memory, and can also include a non-transitory memory, such as at least one disk storage device, a flash memory device, or other non-transitory solid-state storage devices. In some alternative embodiments, the memory 1402 can optionally include a memory remotely set relative to the processor 1401, and these remote memories can be connected to the electronic device through a network. Examples of the above-mentioned network include but are not limited to the Internet, an enterprise intranet, a local area network, a mobile communication network, and combinations thereof.

[0172] The memory 1402 can include a volatile memory, such as a random access memory; the memory can also include a non-volatile memory, such as a flash memory, a hard disk, or a solid-state hard disk; the memory 1402 can also include a combination of the above types of memories.

[0173] The electronic device further includes an input device 1403 and an output device 1404. The processor 1401, the memory 1402, the input device 1403, and the output device 1404 can be connected through a bus or other means. Figure 14 Taking connection through a bus as an example.

[0174] The input device 1403 can receive input digital or character information, and generate key signal inputs related to the user settings and function controls of the electronic device, such as a touch screen, a keypad, a mouse, a trackpad, a touchpad, a pointing stick, one or more mouse buttons, a trackball, a joystick, etc. The output device 1404 can include a display device, an auxiliary lighting device (such as an LED), and a tactile feedback device (such as a vibration motor), etc. The above-mentioned display device includes but is not limited to a liquid crystal display, a light-emitting diode, a display, and a plasma display. In some alternative embodiments, the display device can be a touch screen.

[0175] Embodiments of the present disclosure also provide a computer-readable storage medium. The methods according to the embodiments of the present disclosure can be implemented in hardware, firmware, or be implemented as computer code that can be recorded on a storage medium, or be implemented as computer code that is originally stored in a remote storage medium or a non-transitory machine-readable storage medium and downloaded through a network and will be stored in a local storage medium, so that the methods described herein can be processed by such software stored on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only memory, a random access memory, a flash memory, a hard disk, or a solid-state drive, etc.; further, the storage medium can also include a combination of the above-mentioned types of memories. It can be understood that a computer, a processor, a microprocessor controller, or programmable hardware includes a storage component that can store or receive software or computer code. When the software or computer code is accessed and executed by the computer, the processor, or the hardware, the methods shown in the above embodiments are implemented.

[0176] A part of the present disclosure can be applied as a computer program product, such as computer program instructions. When executed by a computer, through the operation of the computer, the methods and / or technical solutions according to the present disclosure can be invoked or provided. Those skilled in the art should understand that the forms of existence of computer program instructions in a computer-readable medium include, but are not limited to, source files, executable files, installation package files, etc. Correspondingly, the ways in which computer program instructions are executed by a computer include, but are not limited to: the computer directly executes the instructions, or the computer compiles the instructions and then executes the corresponding compiled program, or the computer reads and executes the instructions, or the computer reads and installs the instructions and then executes the corresponding installed program. Herein, the computer-readable medium can be any available computer-readable storage medium or communication medium accessible to the computer.

[0177] Although the embodiments of the present disclosure have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of the present disclosure, and such modifications and variations all fall within the scope defined by the appended claims.

Claims

1. A data analysis method based on multi-round question-answering, characterized in that: The method comprises: Displaying a data analysis page, the data analysis page comprising a dialogue area, the dialogue area displaying at least one first question-answer pair, the first question-answer pair having a corresponding grouping label, a first sub-area of ​​the dialogue area being used to display the grouping label, the grouping label being determined based on a semantic analysis result of the first question-answer pair; In response to a selection instruction for a grouping label in the first sub-area, a first grouping label is determined, and a first question-answer pair corresponding to the first grouping label is displayed in the conversation area.

2. The data analysis method according to claim 1, characterized in that: The method further comprises: If the current interaction point is located at a corresponding position of any grouping label in the first sub-area, the number of question-answer pairs and description information corresponding to the grouping label are displayed.

3. The data analysis method according to claim 1, characterized in that: The second sub-area corresponding to each of the first question-answer pairs in the conversation area displays the group label; the method further includes: In response to an interactive operation on any grouping label in the second sub-area, displaying a property page of the grouping label, the property page being used to configure display properties of the grouping label; In response to the display attribute determined on the attribute page, the grouping label of the corresponding first question-answer pair is updated.

4. The data analysis method according to claim 1, characterized in that: The third sub-area corresponding to each of the first question-answer pairs in the conversation area displays a recommended question of the first question-answer pair; the method further includes: In response to a selection instruction for a recommended question of any of the first question-answer pairs, an analysis instruction is generated and the selected recommended question is hidden in the corresponding third sub-region, the analysis instruction is used to analyze the selected recommended question to obtain an answer content of the selected recommended question, so as to obtain a second question-answer pair and a grouping label of the second question-answer pair, wherein the grouping label of the second question-answer pair is determined based on a semantic analysis result of the second question-answer pair; The second question-answer pair is displayed in the conversation area, and a grouping label of the second question-answer pair is displayed in the first sub-area.

5. The data analysis method according to claim 4, characterized in that: The method further comprises: In response to a deselection instruction for the first grouping label, the first question and answer pair and the second question and answer pair are displayed in the conversation area according to the question and answer time.

6. The data analysis method according to claim 1, characterized in that: The data analysis page also displays a drawing board area, and the drawing board area is used to display the interesting contents in the conversation area and the association relationship between the interesting contents.

7. The data analysis method according to claim 6, characterized in that: The drawing board area displays first analysis content, the first analysis content includes at least one first data unit with an associated relationship, and the first data unit is generated based on the first content of interest in the conversation area; the method further includes: In response to selected content of a target question-answer pair in the conversation area, determining second content of interest, the target question-answer pair comprising at least one of the first question-answer pair and a second question-answer pair, the second question-answer pair being obtained based on a recommended question of the first question-answer pair; In response to the operation of moving the second interesting content to the artboard area, displaying a second data unit at a position corresponding to the second interesting content in the artboard area, wherein the second data unit includes the second interesting content and a unit title, and the unit title is related to the question in the target question-answer pair; The second data unit is associated with the first data unit in the drawing board area to obtain second analysis content.

8. The data analysis method according to claim 7, characterized in that: The associating the second data unit with the first data unit in the drawing board area to obtain second analysis content includes: In the artboard area, based on the distance between the first data unit and the second data unit, determining a first data unit to be associated with the second data unit; The second data unit is associated with the first data unit that is determined to be associated to obtain the second analysis content.

9. The data analysis method according to claim 7, characterized in that: The method further comprises: In response to an editing operation on a target data unit in the artboard area, the target data unit is updated, and the target data unit is the first data unit or the second data unit.

10. The data analysis method according to claim 7, characterized in that: The association relationship is a hierarchical relationship, and the first data unit at the top level is related to the first question of the first question-answer pair in this conversation.

11. The data analysis method according to claim 6, characterized in that: The drawing board area also displays a first control, which is used to trigger the display of historical analysis content and the title of the historical analysis content.

12. The data analysis method according to claim 6, characterized in that: The artboard area also displays a second control, which is used to export the content in the artboard area.

13. A data analysis device based on multi-round question and answer, characterized in that: The device comprises: A first display module, used to display a data analysis page, the data analysis page includes a dialogue area, the dialogue area displays at least one first question-answer pair, the first question-answer pair has a corresponding grouping label, and a first sub-area of ​​the dialogue area is used to display the grouping label, and the grouping label is determined based on a semantic analysis result of the first question-answer pair; The second display module is used to determine a first grouping label in response to a selection instruction for the grouping label in the first sub-area, and display a first question-answer pair corresponding to the first grouping label in the conversation area.

14. An electronic device, characterized in that: include: A memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the data analysis method based on multi-round questioning as described in any one of claims 1 to 12 by executing the computer instructions.

15. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a computer to execute the data analysis method based on multi-round questioning and answering according to any one of claims 1 to 12.

16. A computer program product, characterized in that The method comprises computer instructions for causing a computer to execute the data analysis method based on multi-round questioning and answering according to any one of claims 1 to 12.

Images