Question and answer information tracing method, device, equipment, computer-readable medium and product

Abstract

The embodiments of the present disclosure disclose a method, device, equipment, computer-readable medium and product for tracing the source of question and answer information. A specific implementation of the method includes: generating question and answer information according to question information; displaying the question and answer information at a preset position corresponding to the question information in the question and answer window; in response to detecting a selection operation acting on any question and answer content in the question and answer information, determining at least one unit question and answer information included in the question and answer content as a target unit question and answer information sequence; for each target unit question and answer information, determining each matching object information corresponding to the target unit question and answer information as a matching object information set; dividing each matching object information set into a matching object information group set; determining a target matching object information group; and displaying a reference source corresponding to the target matching object information group in a tracing window corresponding to the question and answer window. This implementation enables users to quickly verify the authenticity of answers, accurately locate specific locations, and improve user experience.

Description

Technical Field

[0001] Embodiments of the present disclosure relate to the field of computer technology, and in particular to question and answer information tracing methods, devices, equipment, computer-readable media, and products. Background Art

[0002] Intelligent question answering system is an advanced form of information retrieval system, which can answer questions asked by users in natural language accurately and concisely. At present, when answering questions asked by users, the usual methods are: directly giving answers to user questions, or directly generating references to answers (for example, similar to references in papers).

[0003] However, the inventors have discovered that when the above-mentioned method is used to give answers, the following technical problems often occur: the method of directly giving answers cannot verify the reference information of the answers generated by the system, resulting in the user being unable to know the authenticity of the answers given by the system. When the authenticity of some answers needs to be verified, additional searches through search engines are still required, resulting in a poor user experience; the method of directly generating references to answers has a large granularity of references and cannot accurately locate specific locations.

[0004] The above information disclosed in this Background section is only for enhancement of understanding of the background of the inventive concept and therefore it may contain information that does not form the prior art that is already known in this country to a person of ordinary skill in the art. Summary of the invention

[0005] The content of this disclosure is used to introduce concepts in a brief form, which will be described in detail in the detailed implementation section below. The content of this disclosure is not intended to identify the key features or essential features of the technical solution claimed for protection, nor is it intended to limit the scope of the technical solution claimed for protection.

[0006] Some embodiments of the present disclosure propose question and answer information tracing methods, devices, electronic devices, computer-readable media and computer program products to solve one or more of the technical problems mentioned in the above background technology section.

[0007] In a first aspect, some embodiments of the present disclosure provide a method for tracing the source of question and answer information, the method comprising: generating question and answer information corresponding to the question information according to the question information input by the user, wherein the question and answer information comprises a sequence of unit question and answer information, each unit question and answer information corresponds to a matching value vector, each matching value in the matching value vector corresponds to matching object information, the matching object information comprises a matching object and reference source information corresponding to the matching object, the reference source information comprises a reference source identifier and position information of the matching object in the reference source corresponding to the reference source identifier, and each matching value in the matching value vector is arranged in descending order; displaying the question and answer information at a preset position corresponding to the question information in a question and answer window; in response to detecting a selection operation acting on any question and answer content in the question and answer information, confirming at least one unit question and answer information included in the question and answer content A target unit question and answer information sequence is defined, wherein the question and answer content includes at least one unit question and answer information; for each target unit question and answer information in the target unit question and answer information sequence, each matching object information corresponding to the target unit question and answer information is determined as a matching object information set; according to each reference source information included in each obtained matching object information set, each matching object information set is divided into a matching object information group set, wherein each matching object information in each matching object information group corresponds to the same reference source identifier; a target matching object information group is determined; and a reference source corresponding to the target matching object information group is displayed in a traceability window corresponding to the question and answer window, wherein the reference source corresponds to the reference source identifier corresponding to the target matching object information group, and in the displayed reference source, each matching object included in the target matching object information group is in a highlighted state.

[0008] In a second aspect, some embodiments of the present disclosure provide a question-and-answer information tracing device, the device comprising: a generating unit, configured to generate question-and-answer information corresponding to the question information according to the question information input by the user, wherein the question-and-answer information comprises a sequence of unit question-and-answer information, each unit question-and-answer information corresponds to a matching value vector, each matching value in the matching value vector corresponds to matching object information, the matching object information comprises a matching object and reference source information corresponding to the matching object, the reference source information comprises a reference source identifier and position information of the matching object in the reference source corresponding to the reference source identifier, and each matching value in the matching value vector is arranged in descending order; a first display unit, configured to display the question-and-answer information in a preset position corresponding to the question information in a question-and-answer window; a first determining unit, configured to determine at least one unit question-and-answer information included in the question-and-answer content as a target unit in response to detecting a selection operation acting on any question-and-answer content in the question-and-answer information. A question and answer information sequence, wherein the question and answer content includes at least one unit question and answer information; a second determination unit is configured to determine, for each target unit question and answer information in the target unit question and answer information sequence, each matching object information corresponding to the target unit question and answer information as a matching object information set; a division unit is configured to divide the above-obtained matching object information sets into matching object information group sets according to each reference source information included in each matching object information set, wherein each matching object information in each matching object information group corresponds to the same reference source identifier; a third determination unit is configured to determine the target matching object information group; a second display unit is configured to display the reference source corresponding to the target matching object information group in a traceability window corresponding to the question and answer window, wherein the reference source corresponds to the reference source identifier corresponding to the target matching object information group, and in the displayed reference source, each matching object included in the target matching object information group is in a highlighted state.

[0009] In a third aspect, some embodiments of the present disclosure provide an electronic device comprising: one or more processors; a storage device on which one or more programs are stored, and when the one or more programs are executed by the one or more processors, the one or more processors implement the method described in any implementation manner of the above-mentioned first aspect.

[0010] In a fourth aspect, some embodiments of the present disclosure provide a computer-readable medium having a computer program stored thereon, wherein when the program is executed by a processor, the method described in any implementation manner of the above-mentioned first aspect is implemented.

[0011] In a fifth aspect, some embodiments of the present disclosure provide a computer program product, including a computer program, which implements the method described in any implementation manner of the above-mentioned first aspect when executed by a processor.

[0012] The above-mentioned various embodiments of the present disclosure have the following beneficial effects: through the question and answer information tracing method of some embodiments of the present disclosure, users can quickly verify the authenticity of the answer, and can accurately locate the specific location, thereby improving the user experience. Specifically, the reasons why users cannot know the authenticity of the answers given by the system and the user experience is poor are: the way of directly giving answers cannot verify the reference information of the answers generated by the system, resulting in users being unable to know the authenticity of the answers given by the system. When the authenticity of some answers needs to be verified, additional searches through search engines are still required, resulting in poor user experience; the way of directly generating references to answers has a large granularity of references and cannot be accurately located to a specific location. Based on this, the question and answer information tracing method of some embodiments of the present disclosure first generates question and answer information corresponding to the question information based on the question information input by the user. Wherein, the above-mentioned question and answer information includes a sequence of unit question and answer information, each unit question and answer information corresponds to a matching value vector, each matching value in the above-mentioned matching value vector corresponds to matching object information, the matching object information includes a matching object and reference source information corresponding to the above-mentioned matching object, the reference source information includes a reference source identifier and the position information of the matching object in the reference source corresponding to the reference source identifier, and each matching value in the above-mentioned matching value vector is arranged in descending order. Thus, answers can be automatically generated according to user questions, and the matching value vector corresponding to each unit question and answer information is recorded in the process of generating answers, and each matching value corresponds to matching object information, which can provide a reference for tracing and finding evidence. Then, the above-mentioned question and answer information is displayed in the preset position corresponding to the above-mentioned question information in the question and answer window. Thus, the answer can be displayed on the screen for the user to view in accordance with the question asked. Then, in response to detecting a selection operation acting on any question and answer content in the above-mentioned question and answer information, at least one unit question and answer information included in the above-mentioned question and answer content is determined as a target unit question and answer information sequence, wherein the above-mentioned question and answer content includes at least one unit question and answer information. Thus, it is possible to determine the partial question and answer content selected by the user, and to determine the fine-grained unit question and answer information included in the question and answer content selected by the user. Secondly, for each target unit question and answer information in the above-mentioned target unit question and answer information sequence, the respective matching object information corresponding to the above-mentioned target unit question and answer information is determined as a matching object information set. Thus, it is possible to determine the respective matching object information matched when generating each fine-grained target unit question and answer information. Then, according to the respective reference source information included in the obtained respective matching object information sets, the above-mentioned respective matching object information sets are divided into matching object information group sets, wherein the respective matching object information in each matching object information group corresponds to the same reference source identifier. Thus, the respective matching object information belonging to the same reference source can be divided into the same group. Then, the target matching object information group is determined. Thus, the determined target matching object information group can be the respective matching object information in the reference source to be displayed.Finally, the reference source corresponding to the target matching object information group is displayed in the source tracing window corresponding to the above-mentioned question and answer window, wherein the above-mentioned reference source corresponds to the reference source identifier corresponding to the target matching object information group, and in the displayed reference source, each matching object included in the target matching object information group is in a highlighted state. Thus, the matching object that matches the question and answer content selected by the user can be highlighted in the source tracing window, so that the user can directly view the source of the question and answer content in the source tracing window without the need for additional retrieval through a search engine, which results in a poor user experience. Thus, the user can quickly verify the authenticity of the answer and can accurately locate the specific location, thereby improving the user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] The above and other features, advantages and aspects of the embodiments of the present disclosure will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. Throughout the accompanying drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic and that components and elements are not necessarily drawn to scale.

[0014] Figure 1 is a flow chart of some embodiments of the question and answer information tracing method according to the present disclosure;

[0015] Figure 2 It is a schematic diagram of the structure of some embodiments of the question and answer information tracing device according to the present disclosure;

[0016] Figure 3 It is a schematic diagram of the structure of an electronic device suitable for implementing some embodiments of the present disclosure. DETAILED DESCRIPTION

[0017] Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be construed as being limited to the embodiments set forth herein. On the contrary, these embodiments are provided to provide a more thorough and complete understanding of the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are only for exemplary purposes and are not intended to limit the scope of protection of the present disclosure.

[0018] It should also be noted that, for ease of description, only the parts related to the invention are shown in the drawings. In the absence of conflict, the embodiments and features in the embodiments of the present disclosure can be combined with each other.

[0019] It should be noted that the concepts such as "first" and "second" mentioned in the present disclosure are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.

[0020] It should be noted that the modifications of "one" and "plurality" mentioned in the present disclosure are illustrative rather than restrictive, and those skilled in the art should understand that unless otherwise clearly indicated in the context, it should be understood as "one or more".

[0021] The names of the messages or information exchanged between multiple devices in the embodiments of the present disclosure are only used for illustrative purposes and are not used to limit the scope of these messages or information.

[0022] With regard to the collection, storage, and use of user personal information (such as question information) involved in this disclosure, before performing the corresponding operations, the relevant organizations or individuals shall fulfill their obligations, including conducting personal information security impact assessments, fulfilling the obligation to inform the personal information subject, and obtaining the authorization and consent of the personal information subject in advance.

[0023] The present disclosure will be described in detail below with reference to the accompanying drawings and in conjunction with embodiments.

[0024] Figure 1 The process 100 of some embodiments of the question and answer information tracing method according to the present disclosure is shown. The question and answer information tracing method comprises the following steps:

[0025] Step 101: Generate question-answer information corresponding to the question information according to the question information input by the user.

[0026] In some embodiments, the execution subject (such as a computing device) of the question and answer information tracing method can generate question and answer information corresponding to the question information according to the question information input by the user. The question information can be a question input by the user. The question and answer information can be an answer to the question information. The question and answer information includes a sequence of unit question and answer information. The unit question and answer information can be an indivisible character or string. For example, the unit question and answer information can be a text, a word, or a numerical value. Each unit question and answer information corresponds to a matching value vector. The matching value vector can be a vector composed of the matching values ​​of each matching object and the unit question and answer information. The matching values ​​in the matching value vector are arranged in descending order. Each matching value in the matching value vector corresponds to matching object information. The matching object information can be information related to the matching object itself and its reference source. The matching object can be an object in the reference source that matches the unit question and answer information. The type of the matching object can be but is not limited to: text, word, numerical value, image. The matching object information can include the matching object and the reference source information corresponding to the matching object. The reference source information can be information related to the reference source to which the matching object belongs. The reference source information includes the reference source identifier and the location information of the matching object in the reference source corresponding to the reference source identifier. The reference source identifier can represent the reference source. The type of the reference source can be, but is not limited to: a document, a web page. The format of the document can include, but is not limited to: Word, PDF, Excel, txt.

[0027] In practice, the above-mentioned execution entity may send the question information to the server, so that the server generates question and answer information corresponding to the above-mentioned question information through the intelligent question and answer model. The above-mentioned intelligent question and answer model may be a large language model. For example, the above-mentioned large language model may be a generative AI model built based on a Transformer network. The above-mentioned execution entity may be a server or a terminal device. The user may input question information through the question and answer window of the web application, or may input question information through the question and answer window of the local application. The question and answer window may be a window for displaying the question information input by the user and the question and answer information replied by the system. The question information and question and answer information displayed in the question and answer window may be presented in the form of a chat page.

[0028] Optionally, the question and answer information includes text information. Each unit of question and answer information included in the question and answer content is a text character.

[0029] In some optional implementations of some embodiments, question and answer information corresponding to the question information may be generated according to the question information input by the user through the following steps:

[0030] The first step is to generate a reference fragment information set based on the above question information. Each reference fragment information in the reference fragment information set corresponds to a reference source identifier and fragment location information. The above reference fragment information set includes reference fragment information of corresponding text types. In practice, each reference fragment information associated with the above question information can be retrieved through intelligent search technology. The reference fragment information can be part of the content intercepted from a reference source and related to the question information. The fragment location information can be the location information of the reference fragment information in the reference source. For example, when the reference fragment information is a text segment, the location information can be a paragraph identifier.

[0031] The second step is to generate question-answer information corresponding to the question information based on the question information and the reference segment information set. In practice, the question information and the reference segment information set can be input into an intelligent question-answering model to obtain question-answer information corresponding to the question information. The intelligent question-answering model can be a large model. For example, the large model can be a generative AI model built based on a Transformer network.

[0032] Step 3: For each text character included in the question and answer information, perform the following steps:

[0033] The first sub-step is, in response to determining that the text character is not the first character in the question and answer information, generating an initial matching value vector based on the question information, the reference fragment information set and the previous text character of the text character in the question and answer information. The matching objects corresponding to the matching values ​​in the initial matching value vector include matching words. In practice, when the text character is the first character in the question and answer information, the previous text character of the text character in the question and answer information may be a null value.

[0034] The second sub-step is to sort the matching values ​​in the above initial matching value vector in descending order to obtain a matching value vector in descending order.

[0035] The third sub-step is to determine the preset initial value as the cumulative matching value, wherein the preset initial value may be 0.

[0036] The fourth sub-step is to determine the first matching value in the above descending matching value vector as the target matching value.

[0037] The fifth sub-step is to execute the following loop steps based on the cumulative matching value and the target matching value:

[0038] First, the sum of the cumulative matching value and the target matching value is determined as the cumulative matching value to update the cumulative matching value.

[0039] Second, in response to determining that the updated cumulative match value is greater than the preset match threshold, each match value before the target match value in the above descending match value vector and the target match value are concatenated into a screening match value vector. For example, the preset match threshold may be 0.9. Thus, each match value whose cumulative sum exceeds the preset match threshold in the descending match value vector can be screened out.

[0040] Third, in response to determining that the updated cumulative matching value is less than or equal to the preset matching threshold, the next matching value of the target matching value in the descending matching value vector is determined as the target matching value to update the target matching value, and the above-mentioned loop steps are executed again based on the updated cumulative matching value and the updated target matching value.

[0041] The sixth sub-step is to generate a matching value vector and each matching object information corresponding to the above text character according to the spliced ​​screening matching value vector. Thus, the matching value vector and each corresponding matching object information can be determined by using each screening matching value.

[0042] Optionally, the question and answer information may be obtained by inputting the question information into a pre-trained intelligent question and answer model. The intelligent question and answer model may be a generative AI model built based on a Transformer network. The intelligent question and answer model includes various decoding layers. Each decoding layer includes various attention layers. Each attention layer is used to generate an attention value vector for the corresponding text character based on the question information, the reference fragment information set and the text character. The attention value vector may be a vector of attention scores generated for the question information, the reference fragment information set and all the text characters preceding the text character when generating a text character, which may represent all the information sources required to generate the text character.

[0043] Optionally, the execution subject may also determine the question and answer information corresponding to the question information sent by the received server as the question and answer information to be restored. Then, the pre-trained intelligent question and answer model stored locally may be used to restore the question and answer information to be restored for the question information. During the restoration process, the matching value vector of each text character in the question and answer information to be restored and the information of each matching object may be recorded. In this way, the application scenarios of question and answer traceability can be expanded. Even if the intelligent question and answer model stored on the server does not have refined question and answer traceability, fine-grained question and answer traceability can be achieved locally.

[0044] In some optional implementations of some embodiments, an initial matching value vector may be generated based on the question information, the reference segment information set, and the previous text character of the text character in the question and answer information through the following steps:

[0045] In the first step, a decoding layer corresponding to a preset number of layers among the above decoding layers is determined as a target decoding layer. The preset number of layers may be pre-set.

[0046] In the second step, each attention layer included in the above target decoding layer is determined as a target attention layer set.

[0047] The third step is to generate an initial matching value vector based on each attention value vector corresponding to the above text character generated by each target attention layer in the above target attention layer set. In practice, the mean vector corresponding to each attention value vector can be determined as the initial matching value vector.

[0048] In some optional implementations of some embodiments, the following steps may be used to generate a matching value vector corresponding to the above text characters and each matching object information according to the concatenated screening matching value vectors:

[0049] In the first step, the number of each matching value included in the above-mentioned screening matching value vector is determined as the matching number.

[0050] In the second step, in response to determining that the number of matches is greater than a preset number, the preset number of match values ​​are extracted from the screened match value vector as the match value vector corresponding to the text character according to a preset extraction method. The preset number may be 10. The preset extraction method may be a method of extracting from front to back in sequence. Thus, the number of matching objects corresponding to the screened text characters can be limited to within the preset number.

[0051] Step 3: For each matching value in the matching value vector, perform the following steps:

[0052] The first sub-step is to determine the matching object corresponding to the above matching value.

[0053] The second sub-step is to determine the reference segment information to which the determined matching object belongs as the target reference segment information.

[0054] The third sub-step is to determine the reference source identifier of the reference source corresponding to the target reference fragment information.

[0055] The fourth sub-step is to determine the position information of the determined matching object in the reference source corresponding to the reference source identifier. When the matching object is text, the position information may include a paragraph number and a character arrangement number corresponding to the paragraph number, which is used to locate the position of the matching object in the reference source.

[0056] The fifth sub-step is to determine the determined reference source identification and location information as reference source information.

[0057] The sixth sub-step is to determine the determined matching object and reference source information as the matching object information corresponding to the above matching value. Thus, each determined matching object information can be used as each piece of evidence information for generating the text character.

[0058] Optionally, the reference segment information set may include reference segment information of a corresponding image type. The reference segment information of a corresponding image type may be an image. Specifically, the reference segment information of a corresponding image type may be a whole image.

[0059] In some optional implementations of some embodiments, question and answer information corresponding to the question information may be generated according to the question information and the reference segment information set through the following steps:

[0060] In the first step, for each reference segment information of the corresponding image type included in the reference segment information set, the following steps are performed:

[0061] The first sub-step is to segment the reference fragment information to obtain various image blocks. In practice, the reference fragment information can be segmented according to preset row and column information to obtain various image blocks. For example, the preset row and column information can be 3 rows and 3 columns, and the segmented image blocks can be 3 rows and 3 columns.

[0062] In the second sub-step, for each of the above-mentioned image blocks, the following steps are performed:

[0063] First, feature extraction is performed on the image block to obtain a feature vector of the image block. In practice, feature extraction can be performed on the image block through a convolutional neural network to obtain a feature vector of the image block.

[0064] Second, determine the image position of the image block in the reference segment information. The image position can be represented by row and column numbers.

[0065] Third, the image position is coded to obtain an image position coding vector. For example, the coding method may be binary coding.

[0066] Fourth, the image block feature vector and the image position encoding vector are determined as image block information.

[0067] The third sub-step is to determine the determined information of each image block as reference segment information to update the reference segment information.

[0068] The second step is to generate question-answer information corresponding to the question information based on the question information and the updated reference segment information set. In practice, the specific implementation method of generating question-answer information corresponding to the question information based on the question information and the updated reference segment information set can refer to the specific implementation method of "generating question-answer information corresponding to the question information based on the question information and the reference segment information set", which will not be repeated here. Therefore, the image can also be used as evidence information of the text characters that generate the question-answer information.

[0069] In some optional implementations of some embodiments, the reference source corresponding to the target matching object information group may be displayed in the source tracing window corresponding to the question and answer window through the following steps:

[0070] In response to determining that each matching object in the target matching object information group includes a matching object of a corresponding image block type, for each matching object of the corresponding image block type, the following steps are performed:

[0071] The first step is to determine the reference segment information corresponding to the above matching object as the target reference segment information.

[0072] In the second step, the image position of the image block corresponding to the matching object in the target reference segment information is determined as the positioning image position.

[0073] The third step is to display a highlight mark at the location of the positioning image of the target reference segment information in the reference source displayed in the source tracing window, so that the image block corresponding to the matching object in the target reference segment information is in a highlighted state. The highlight mark may be a rectangular frame.

[0074] The reference fragment information corresponding to the image type and its related content are an inventive point of the embodiment of the present disclosure, which solves the technical problem that "when the evidence of the text characters in the generated answer is an image, the user is even more unable to verify the authenticity of the answer, and compared with using a search engine to retrieve text, self-retrieval of images is more complicated, and the user's experience in using the intelligent question-answering system is poor." The factors that lead to the poor experience of users when using the intelligent question-answering system are often as follows: when the evidence of the text characters in the generated answer is an image, the user is even more unable to verify the authenticity of the answer, and compared with using a search engine to retrieve text, self-retrieval of images is more complicated. If the above factors are solved, the effect of improving the user's experience when using the intelligent question-answering system can be achieved. In order to achieve this effect, when the reference fragment information corresponds to the image type, the present disclosure divides the image into individual image blocks, extracts the image features, and then inputs them into the intelligent question-answering model for text character matching. Because the matching dimension is based on the image block, the evidence of the generated text characters can be located from the image block, which is a finer-grained dimension, so that when the evidence of the text characters in the generated answer is an image, the authenticity of the answer can also be verified. There is no need for users to retrieve images by themselves, which improves the user's experience when using the intelligent question-answering system.

[0075] Step 102: display the question and answer information at a preset position corresponding to the question information in the question and answer window.

[0076] In some embodiments, the execution entity may display the question and answer information in a preset position corresponding to the question information in the question and answer window. The preset position may be the lower left corner. In practice, the execution entity may display the question and answer information word by word at the preset position.

[0077] Step 103: in response to detecting a selection operation acting on any question and answer content in the question and answer information, determining at least one unit question and answer information included in the question and answer content as a target unit question and answer information sequence.

[0078] In some embodiments, the execution subject may determine at least one unit question and answer information included in the question and answer content as a target unit question and answer information sequence in response to detecting a selection operation acting on any question and answer content in the question and answer information. The selection operation may be, but is not limited to, a point selection operation, a box selection operation, and a swipe operation. For example, the point selection operation may be a double-click operation acting on a position in the question and answer information, and the word, word or value at the position may be selected by the double-click operation. The swipe operation may be an operation of selecting the unit question and answer information falling within the rectangular area by pulling a rectangular area. The swipe operation may be an operation of swiping each character in front of a character in the question and answer information in sequence. In practice, the execution subject may sequentially combine the various unit question and answer information included in the question and answer content into a target unit question and answer information sequence. When the question and answer content selected by the user includes only one unit question and answer information, the target unit question and answer information sequence includes only one target unit question and answer information.

[0079] Step 104: for each target unit question and answer information in the target unit question and answer information sequence, each piece of matching object information corresponding to the target unit question and answer information is determined as a matching object information set.

[0080] In some embodiments, the execution subject may determine, for each target unit question and answer information in the target unit question and answer information sequence, each matching object information corresponding to the target unit question and answer information as a matching object information set. In practice, the execution subject may determine the matching value vector corresponding to the target unit question and answer information as a target matching value vector. Then, each matching object information corresponding to the target matching value vector may be determined as a matching object information set.

[0081] Step 105: divide each matching object information set into matching object information group sets according to each reference source information included in each obtained matching object information set.

[0082] In some embodiments, the execution subject may divide the above-mentioned matching object information sets into matching object information group sets according to the reference source information included in each matching object information set. The matching object information in each matching object information group corresponds to the same reference source identifier. In practice, the execution subject may divide the matching object information with the same corresponding reference source identifier in the above-mentioned matching object information sets into the same matching object information group to obtain a matching object information group set.

[0083] Step 106: Determine the target matching object information group.

[0084] In some embodiments, the execution entity may determine a target matching object information group. In practice, the execution entity may determine the first matching object information group in the matching object information group set as the target matching object information group.

[0085] In some optional implementations of some embodiments, the execution subject may determine the target matching object information group through the following steps:

[0086] In the first step, the first matching object information group in the above matching object information group set is determined as the target matching object information group.

[0087] In the second step, in response to detecting the reference source switching operation, according to the target matching object information group, the matching object information group corresponding to the reference source switching operation in the above matching object information group set is determined as the target matching object information group to update the target matching object information group. Among them, the reference source switching operation can be a selection operation acting on the reference source switching control. The reference source switching control can include a reference source up-cut control and a reference source down-cut control. The reference source up-cut control is used to switch to the previous reference source. The reference source down-cut control is used to switch to the next reference source. The reference source switching control can be displayed in the traceability window. In practice, the above-mentioned execution subject can determine a matching object information group after the target matching object information group in the matching object information group set as the target matching object information group in response to determining that the above-mentioned reference source switching operation is a selection operation acting on the reference source down-cut control, so as to update the target matching object information group. It can be understood that after the target matching object information group is updated, the corresponding reference source is also updated accordingly.

[0088] The third step is to switch the reference source displayed in the traceability window to the reference source corresponding to the updated target matching object information group, wherein each matching object included in the updated target matching object information group in the switched reference source is in a highlighted state.

[0089] Step 107: Display the reference source corresponding to the target matching object information group in the source tracing window corresponding to the question and answer window.

[0090] In some embodiments, the execution subject may display the reference source corresponding to the target matching object information group in the tracing window corresponding to the question-and-answer window. The reference source corresponds to the reference source identifier corresponding to the target matching object information group, and in the displayed reference source, each matching object included in the target matching object information group is in a highlighted state. For example, the matching object in a highlighted state may be displayed in a highlighted manner or in a box-selected manner.

[0091] In some optional implementations of some embodiments, the execution subject may display the reference source corresponding to the target matching object information group in the source tracing window corresponding to the question-and-answer window through the following steps:

[0092] The first step is to determine whether there is a matching object included in the target matching object information group in the current tracing page displayed in the above tracing window.

[0093] In the second step, in response to determining that there is a matching object included in the target matching object information group in the current tracing page displayed in the above-mentioned tracing window, for each matching object included in the target matching object information group existing in the current tracing page, the above-mentioned matching object is highlighted. Among them, the matching object switching control is also displayed in the above-mentioned current tracing page. The matching object switching control can be used to switch the selected matching object. The matching object switching control can include. The matching object switching control can include a matching object up-cut control and a matching object down-cut control. The matching object up-cut control is used to switch to the previous matching object. The matching object down-cut control is used to switch to the next matching object. The matching object switching control can be displayed in the tracing window. In the initial state, the selected matching object is the first matching object included in the target matching object information group.

[0094] The third step is to determine the switching matching object according to the target matching object information group and the currently selected matching object in response to detecting the selection operation corresponding to the matching object switching control. In practice, the execution subject may determine the next matching object of the currently selected matching object in the target matching object information group as the switching matching object in response to determining that the selection operation corresponding to the matching object switching control is a selection operation acting on the matching object down-cut control.

[0095] The fourth step is to locate the current source tracing page to the page position corresponding to the switching matching object. In practice, the execution subject can scroll the current source tracing page to the page position corresponding to the switching matching object. The current source tracing page can be a web page or a document page. Thus, the matching object switching control can facilitate users to view various matching objects in a reference source.

[0096] Optionally, in response to determining that the question-and-answer content includes at least one sentence, for each of the at least one sentence, the execution subject may perform the following steps:

[0097] The first step is to determine each unit question and answer information included in the above sentence as a unit question and answer information set. Here, a sentence can be understood as a character string ending with a period.

[0098] In the second step, each location information included in each matching object information corresponding to the above unit question and answer information set is determined as a traceability location information set.

[0099] The third step is to determine each paragraph identifier corresponding to each source tracing location information in the source tracing location information set, wherein the paragraph identifier among the above paragraph identifiers is unique.

[0100] The fourth step is to extract the paragraph contents of each paragraph corresponding to each paragraph identifier from the reference source corresponding to the matching object information.

[0101] The fifth step is to input the paragraph content and the corresponding sentence into a pre-trained traceability relationship recognition model for each of the paragraph contents mentioned above, and obtain the traceability relationship between the paragraph content and the sentence. The traceability relationship recognition model can be a neural network model that takes the text in the question and answer information and the text in the reference source as input, and takes the traceability relationship between the two texts as output. The traceability relationship can be, but is not limited to, support, contradiction, and irrelevance. Support can represent that the text in the reference source can support the generation of the text in the question and answer information. Contradiction can represent that the text in the reference source and the text in the question and answer information have completely opposite meanings. Irrelevant can represent that there is no connection between the generation of the text in the reference source and the text in the question and answer information. For example, the traceability relationship recognition model can be a natural language inference model (Natural Language Inference, NLI).

[0102] Optionally, in the displayed reference source, the highlight state of each matching object included in the target matching object information group corresponds to the target tracing relationship corresponding to the above matching object, and the above target tracing relationship is the tracing relationship between the paragraph content where the above matching object is located and the sentence where the target unit question and answer information corresponding to the above matching object is located. Specifically, when the target tracing relationship is supportive, the highlight state of the matching object can be a green highlighted state. When the target tracing relationship is contradictory, the highlight state of the matching object can be a red highlighted state. When the target tracing relationship is irrelevant, the highlight state of the matching object can be a gray highlighted state. As a result, users can view the relationship between the tracing content and the question and answer content to confirm the supporting relationship of the tracing content to the question and answer content.

[0103] The above-mentioned various embodiments of the present disclosure have the following beneficial effects: through the question and answer information tracing method of some embodiments of the present disclosure, users can quickly verify the authenticity of the answer, and can accurately locate the specific location, thereby improving the user experience. Specifically, the reasons why users cannot know the authenticity of the answers given by the system and the user experience is poor are: the way of directly giving answers cannot verify the reference information of the answers generated by the system, resulting in users being unable to know the authenticity of the answers given by the system. When the authenticity of some answers needs to be verified, additional searches through search engines are still required, resulting in poor user experience; the way of directly generating references to answers has a large granularity of references and cannot be accurately located to a specific location. Based on this, the question and answer information tracing method of some embodiments of the present disclosure first generates question and answer information corresponding to the question information based on the question information input by the user. Wherein, the above-mentioned question and answer information includes a sequence of unit question and answer information, each unit question and answer information corresponds to a matching value vector, each matching value in the above-mentioned matching value vector corresponds to matching object information, the matching object information includes a matching object and reference source information corresponding to the above-mentioned matching object, the reference source information includes a reference source identifier and the position information of the matching object in the reference source corresponding to the reference source identifier, and each matching value in the above-mentioned matching value vector is arranged in descending order. Thus, answers can be automatically generated according to user questions, and the matching value vector corresponding to each unit question and answer information is recorded in the process of generating answers, and each matching value corresponds to matching object information, which can provide a reference for tracing and finding evidence. Then, the above-mentioned question and answer information is displayed in the preset position corresponding to the above-mentioned question information in the question and answer window. Thus, the answer can be displayed on the screen for the user to view in accordance with the question asked. Then, in response to detecting a selection operation acting on any question and answer content in the above-mentioned question and answer information, at least one unit question and answer information included in the above-mentioned question and answer content is determined as a target unit question and answer information sequence, wherein the above-mentioned question and answer content includes at least one unit question and answer information. Thus, it is possible to determine the partial question and answer content selected by the user, and to determine the fine-grained unit question and answer information included in the question and answer content selected by the user. Secondly, for each target unit question and answer information in the above-mentioned target unit question and answer information sequence, the respective matching object information corresponding to the above-mentioned target unit question and answer information is determined as a matching object information set. Thus, it is possible to determine the respective matching object information matched when generating each fine-grained target unit question and answer information. Then, according to the respective reference source information included in the obtained respective matching object information sets, the above-mentioned respective matching object information sets are divided into matching object information group sets, wherein the respective matching object information in each matching object information group corresponds to the same reference source identifier. Thus, the respective matching object information belonging to the same reference source can be divided into the same group. Then, the target matching object information group is determined. Thus, the determined target matching object information group can be the respective matching object information in the reference source to be displayed.Finally, the reference source corresponding to the target matching object information group is displayed in the source tracing window corresponding to the above-mentioned question and answer window, wherein the above-mentioned reference source corresponds to the reference source identifier corresponding to the target matching object information group, and in the displayed reference source, each matching object included in the target matching object information group is in a highlighted state. Thus, the matching object that matches the question and answer content selected by the user can be highlighted in the source tracing window, so that the user can directly view the source of the question and answer content in the source tracing window without the need for additional retrieval through a search engine, which results in a poor user experience. Thus, the user can quickly verify the authenticity of the answer and can accurately locate the specific location, thereby improving the user experience.

[0104] Further references Figure 2 As an implementation of the methods shown in the above figures, the present disclosure provides some embodiments of a question-answer information tracing device. These device embodiments are similar to Figure 1 Corresponding to the method embodiments shown, the device can be specifically applied to various electronic devices.

[0105] like Figure 2As shown, the question and answer information tracing device 200 of some embodiments includes: a generating unit 201, a first display unit 202, a first determining unit 203, a second determining unit 204, a dividing unit 205, a third determining unit 206 and a second display unit 207. Wherein, the generation unit 201 is configured to generate question and answer information corresponding to the question information according to the question information input by the user, wherein the question and answer information includes a unit question and answer information sequence, each unit question and answer information corresponds to a matching value vector, each matching value in the matching value vector corresponds to matching object information, the matching object information includes a matching object and reference source information corresponding to the matching object, the reference source information includes a reference source identifier and position information of the matching object in the reference source corresponding to the reference source identifier, and each matching value in the matching value vector is arranged in descending order; the first display unit 202 is configured to display the question and answer information in a preset position corresponding to the question information in the question and answer window; the first determination unit 203 is configured to determine at least one unit question and answer information included in the question and answer content as a target unit question and answer information sequence in response to detecting a selection operation acting on any question and answer content in the question and answer information, wherein the question and answer content It includes at least one unit question and answer information; the second determination unit 204 is configured to determine, for each target unit question and answer information in the above target unit question and answer information sequence, each matching object information corresponding to the above target unit question and answer information as a matching object information set; the division unit 205 is configured to divide the above matching object information sets into matching object information group sets according to each reference source information included in the obtained matching object information sets, wherein each matching object information in each matching object information group corresponds to the same reference source identifier; the third determination unit 206 is configured to determine the target matching object information group; the second display unit 207 is configured to display the reference source corresponding to the target matching object information group in the traceability window corresponding to the above question and answer window, wherein the above reference source corresponds to the reference source identifier corresponding to the target matching object information group, and in the displayed reference source, each matching object included in the target matching object information group is in a highlighted state.

[0106] It is understood that the units described in the device 200 are similar to those described in the reference Figure 1 Therefore, the operations, features and beneficial effects described above for the method are also applicable to the device 200 and the units included therein, and will not be described in detail here.

[0107] Reference below Figure 3 , which shows a structural schematic diagram of an electronic device 300 suitable for implementing some embodiments of the present disclosure. Figure 3 The electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0108] like Figure 3 As shown, the electronic device 300 may include a processing device 301 (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 302 or a program loaded from a storage device 308 into a random access memory (RAM) 303. In the RAM 303, various programs and data required for the operation of the electronic device 300 are also stored. The processing device 301, the ROM 302, and the RAM 303 are connected to each other via a bus 304. An input / output (I / O) interface 305 is also connected to the bus 304.

[0109] Typically, the following devices may be connected to the I / O interface 305: input devices 306 including, for example, a touch screen, a touch pad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; output devices 307 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; storage devices 308 including, for example, a magnetic tape, a hard disk, etc.; and communication devices 309. The communication devices 309 may allow the electronic device 300 to communicate with other devices wirelessly or by wire to exchange data. Although Figure 3 The electronic device 300 is shown with various devices, but it should be understood that it is not required to implement or possess all the devices shown. More or fewer devices may be implemented or possessed instead. Figure 3 Each block shown in the figure may represent one device, or may represent multiple devices as required.

[0110] In particular, according to some embodiments of the present disclosure, the process described above with reference to the flowchart can be implemented as a computer software program. For example, some embodiments of the present disclosure include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In some such embodiments, the computer program can be downloaded and installed from the network through the communication device 309, or installed from the storage device 308, or installed from the ROM 302. When the computer program is executed by the processing device 301, the above-mentioned functions defined in the method of some embodiments of the present disclosure are executed.

[0111] It should be noted that the computer-readable medium recorded in some embodiments of the present disclosure may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In some embodiments of the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In some embodiments of the present disclosure, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The computer readable signal medium may also be any computer readable medium other than a computer readable storage medium, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer readable medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0112] In some embodiments, the client and the server may communicate using any currently known or future developed network protocol such as HTTP (HyperText Transfer Protocol), and may be interconnected with any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network ("LAN"), a wide area network ("WAN"), an internet (e.g., the Internet), and a peer-to-peer network (e.g., an ad hoc peer-to-peer network), as well as any currently known or future developed network.

[0113] The computer-readable medium may be included in the electronic device; or it may exist independently without being installed in the electronic device. The computer-readable medium carries one or more programs. When the one or more programs are executed by the electronic device, the electronic device: generates question-and-answer information corresponding to the question information according to the question information input by the user, wherein the question-and-answer information includes a sequence of unit question-and-answer information, each unit question-and-answer information corresponds to a matching value vector, each matching value in the matching value vector corresponds to matching object information, the matching object information includes a matching object and reference source information corresponding to the matching object, the reference source information includes a reference source identifier and position information of the matching object in the reference source corresponding to the reference source identifier, and each matching value in the matching value vector is arranged in descending order; displays the question-and-answer information in a preset position corresponding to the question information in the question-and-answer window; and in response to detecting a selection operation acting on any question-and-answer content in the question-and-answer information, displays at least one of the question-and-answer contents included in the question-and-answer content. Unit question and answer information is determined as a target unit question and answer information sequence, wherein the above question and answer content includes at least one unit question and answer information; for each target unit question and answer information in the above target unit question and answer information sequence, each matching object information corresponding to the above target unit question and answer information is determined as a matching object information set; according to each reference source information included in each obtained matching object information set, the above each matching object information set is divided into a matching object information group set, wherein each matching object information in each matching object information group corresponds to the same reference source identifier; determine the target matching object information group; display the reference source corresponding to the target matching object information group in the tracing window corresponding to the above question and answer window, wherein the above reference source corresponds to the reference source identifier corresponding to the target matching object information group, and in the displayed reference source, each matching object included in the target matching object information group is in a highlighted state.

[0114] Computer program code for performing the operations of some embodiments of the present disclosure may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0115] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present disclosure. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some implementations as replacements, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0116] The units described in some embodiments of the present disclosure may be implemented by software or by hardware. The described units may also be provided in a processor, for example, may be described as: a processor including a generation unit, a first display unit, a first determination unit, a second determination unit, a division unit, a third determination unit, and a second display unit. The names of these units do not, in some cases, constitute limitations on the units themselves, for example, the generation unit may also be described as "a unit that generates question and answer information corresponding to the question information input by the user".

[0117] The functions described above herein may be performed at least in part by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), and the like.

[0118] Some embodiments of the present disclosure further provide a computer program product, including a computer program, which implements any of the above-mentioned question and answer information tracing methods when executed by a processor.

[0119] The above descriptions are only some preferred embodiments of the present disclosure and an explanation of the technical principles used. Those skilled in the art should understand that the scope of the invention involved in the embodiments of the present disclosure is not limited to the technical solutions formed by a specific combination of the above-mentioned technical features, but should also cover other technical solutions formed by any combination of the above-mentioned technical features or their equivalent features without departing from the above-mentioned inventive concept. For example, the above-mentioned features are replaced with the technical features with similar functions disclosed in the embodiments of the present disclosure (but not limited to) and the technical solutions formed.

Claims

1. A question-and-answer information tracing method, comprising: According to the question information input by the user, question and answer information corresponding to the question information is generated, wherein the question and answer information includes a sequence of unit question and answer information, each unit question and answer information corresponds to a matching value vector, each matching value in the matching value vector corresponds to matching object information, the matching object information includes a matching object and reference source information corresponding to the matching object, the reference source information includes a reference source identifier and position information of the matching object in a reference source corresponding to the reference source identifier, and each matching value in the matching value vector is arranged in descending order; Displaying the question and answer information in a preset position corresponding to the question information in the question and answer window; In response to detecting a selection operation acting on any question and answer content in the question and answer information, determining at least one unit question and answer information included in the question and answer content as a target unit question and answer information sequence, wherein the question and answer content includes at least one unit question and answer information; For each target unit question and answer information in the target unit question and answer information sequence, determining each matching object information corresponding to the target unit question and answer information as a matching object information set; According to each reference source information included in each obtained matching object information set, the respective matching object information sets are divided into matching object information group sets, wherein each matching object information in each matching object information group corresponds to the same reference source identifier; Determine a target matching object information group; The reference source corresponding to the target matching object information group is displayed in the traceability window corresponding to the question and answer window, wherein the reference source corresponds to the reference source identifier corresponding to the target matching object information group, and in the displayed reference source, each matching object included in the target matching object information group is in a highlighted state.

2. The method according to claim 1, wherein: The question and answer information includes text information, and each unit of question and answer information included in the question and answer content is a text character; And generating question-answer information corresponding to the question information according to the question information input by the user includes: Generate a reference segment information set according to the question information, wherein each reference segment information in the reference segment information set corresponds to a reference source identifier and segment position information, and the reference segment information set includes reference segment information corresponding to a text type; generating question-answer information corresponding to the question information according to the question information and the reference segment information set; For each text character included in the question and answer information, perform the following steps: In response to determining that the text character is not the first character in the question and answer information, generating an initial matching value vector based on the question information, the reference segment information set, and a previous text character of the text character in the question and answer information, wherein the matching objects corresponding to the matching values ​​in the initial matching value vector include matching words; Sorting each matching value in the initial matching value vector in descending order to obtain a matching value vector in descending order; Determine the preset initial value as the cumulative matching value; Determine the first matching value in the descending matching value vector as the target matching value; Based on the cumulative match value and the target match value, the following loop steps are performed: Determine the sum of the cumulative matching value and the target matching value as the cumulative matching value, so as to update the cumulative matching value; In response to determining that the updated cumulative match value is greater than a preset match threshold, concatenating each match value before the target match value in the descending match value vector and the target match value into a screening match value vector; In response to determining that the updated cumulative matching value is less than or equal to the preset matching threshold, determining the next matching value of the target matching value in the descending matching value vector as the target matching value to update the target matching value, and executing the looping step again based on the updated cumulative matching value and the updated target matching value; According to the spliced ​​screening matching value vectors, a matching value vector corresponding to the text character and each matching object information is generated.

3. The method according to claim 2, wherein: The step of generating a matching value vector corresponding to the text character and each matching object information according to the spliced ​​screening matching value vectors includes: Determine the number of each matching value included in the screening matching value vector as the matching number; In response to determining that the number of matches is greater than a preset number, extracting the preset number of match values ​​from the screening match value vector as a match value vector corresponding to the text character according to a preset extraction method; For each matching value in the matching value vector, perform the following steps: Determine a matching object corresponding to the matching value; Determining the reference segment information to which the determined matching object belongs as target reference segment information; Determine a reference source identifier of a reference source corresponding to the target reference segment information; Determine the position information of the determined matching object in the reference source corresponding to the reference source identifier; Determining the determined reference source identification and location information as reference source information; The determined matching object and reference source information are determined as matching object information corresponding to the matching value.

4. The method according to claim 2, wherein: The question-answer information is obtained by inputting the question information into a pre-trained intelligent question-answer model, wherein the intelligent question-answer model includes various decoding layers, each decoding layer includes various attention layers, and each attention layer is used to generate an attention value vector corresponding to a text character according to the question information, a reference segment information set, and a text character; And the generating an initial matching value vector based on the question information, the reference segment information set and the previous text character of the text character in the question and answer information comprises: Determining a decoding layer corresponding to a preset number of layers among the decoding layers as a target decoding layer; Determine each attention layer included in the target decoding layer as a target attention layer set; An initial matching value vector is generated based on each attention value vector corresponding to the text character generated by each target attention layer in the target attention layer set.

5. The method according to claim 1, wherein: The step of determining the target matching object information group includes: Determine the first matching object information group in the matching object information group set as the target matching object information group; In response to detecting a reference source switching operation, according to the target matching object information group, determining the matching object information group in the matching object information group set corresponding to the reference source switching operation as the target matching object information group, so as to update the target matching object information group; The reference source displayed in the tracing window is switched to a reference source corresponding to the updated target matching object information group, wherein in the switched reference source, each matching object included in the updated target matching object information group is in a highlighted state.

6. The method according to claim 1, wherein: The displaying of the reference source corresponding to the target matching object information group in the source tracing window corresponding to the question-and-answer window includes: Determine whether there is a matching object included in the target matching object information group in the current tracing page displayed in the tracing window; In response to determining that a matching object included in the target matching object information group exists in the current tracing page displayed in the tracing window, for each matching object included in the target matching object information group existing in the current tracing page, the matching object is highlighted, wherein a matching object switching control is also displayed in the current tracing page; In response to detecting a selection operation corresponding to the matching object switching control, determining a switching matching object according to the target matching object information group and the currently selected matching object; The current tracing page is positioned to the page position corresponding to the switching matching object.

7. A question-and-answer information tracing device, comprising: A generating unit is configured to generate question-answer information corresponding to the question information according to the question information input by the user, wherein the question-answer information includes a sequence of unit question-answer information, each unit question-answer information corresponds to a matching value vector, each matching value in the matching value vector corresponds to matching object information, the matching object information includes a matching object and reference source information corresponding to the matching object, the reference source information includes a reference source identifier and position information of the matching object in the reference source corresponding to the reference source identifier, and each matching value in the matching value vector is arranged in descending order; A first display unit is configured to display the question and answer information in a preset position corresponding to the question information in the question and answer window; A first determining unit is configured to, in response to detecting a selection operation acting on any question and answer content in the question and answer information, determine at least one unit question and answer information included in the question and answer content as a target unit question and answer information sequence, wherein the question and answer content includes at least one unit question and answer information; a second determining unit configured to determine, for each target unit question and answer information in the target unit question and answer information sequence, each piece of matching object information corresponding to the target unit question and answer information as a matching object information set; a dividing unit configured to divide each obtained matching object information set into matching object information group sets according to each reference source information included in each obtained matching object information set, wherein each matching object information in each matching object information group corresponds to the same reference source identifier; A third determining unit is configured to determine a target matching object information group; The second display unit is configured to display a reference source corresponding to the target matching object information group in a tracing window corresponding to the question and answer window, wherein the reference source corresponds to a reference source identifier corresponding to the target matching object information group, and in the displayed reference source, each matching object included in the target matching object information group is in a highlighted state.

8. An electronic device comprising: one or more processors; a storage device having one or more programs stored thereon, When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 6.

9. A computer readable medium having a computer program stored thereon, wherein: When the computer program is executed by a processor, the method according to any one of claims 1 to 6 is implemented.

10. A computer program product, comprising a computer program, which, when executed by a processor, implements the method according to any one of claims 1 to 6.

Images