Question and answer information processing method, device and electronic device based on authority label

Abstract

The embodiments of the present disclosure disclose a method, device and electronic device for processing question and answer information based on permission labels. A specific implementation of the method includes: obtaining a question and answer information set; extracting question and answer information features from the question and answer content information included in the question and answer information; merging heterogeneous question and answer information from the question and answer information in the question and answer information set; for a question and answer information group, in response to determining that the number of question and answer information included in the question and answer information group is greater than the target number, generating an updated question and answer information label for the question and answer information group based on the question and answer information source and question and answer information label included in the question and answer information group; updating the index of the question and answer information index tree corresponding to the virtual question and answer information library based on the updated question and answer information label and the question and answer information group; in response to determining that the number of question and answer information included in the question and answer information group is equal to the target number, updating the index of the question and answer information index tree based on the question and answer information group. This implementation improves storage resource utilization.

Description

Technical Field

[0001] Embodiments of the present disclosure relate to the field of computer technology, and in particular to a method, device, and electronic device for processing question and answer information based on permission labels. Background Art

[0002] With the development of machine learning related technologies, question and answer information extraction and application based on machine learning are increasingly used (for example, question and answer information application in the field of regulations). However, for multiple question and answer information databases, when there are duplicate files for question and answer information extraction, it often leads to repeated extraction and storage of question and answer information, which may cause a waste of computing resources and storage resources.

[0003] 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 to a person of ordinary skill in the art. Summary of the invention

[0004] 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.

[0005] Some embodiments of the present disclosure propose a question-and-answer information processing method, device, and electronic device based on permission labels to solve one or more of the technical problems mentioned in the above background technology section.

[0006] In a first aspect, some embodiments of the present disclosure provide a method for processing question and answer information based on permission labels, the method comprising: obtaining a question and answer information set, wherein the question and answer information in the question and answer information set corresponds to different question and answer information bases that are data-isolated from each other, and the question and answer information comprises: a question and answer information source, question and answer content information, and a question and answer information label, wherein the question and answer information label represents the access rights of the question and answer information; performing question and answer information feature extraction on the question and answer content information included in each question and answer information in the question and answer information set to generate question and answer information features; merging heterogeneous question and answer information in the question and answer information set according to the question and answer information features corresponding to the question and answer information to generate a question and answer information group set; for the above For each question and answer information group in the question and answer information group set, the following first processing step is performed: in response to determining that the number of question and answer information included in the above question and answer information group is greater than the target number, based on the question and answer information source and question and answer information tag included in the question and answer information in the above question and answer information group, an updated question and answer information tag for the above question and answer information group is generated; based on the above updated question and answer information tag and the above question and answer information group, the question and answer information index tree corresponding to the virtual question and answer information library is indexed and updated to obtain an updated question and answer information index tree; in response to determining that the number of question and answer information included in the above question and answer information group is equal to the above target number, based on the above question and answer information group, the question and answer information index tree is indexed and updated to obtain an updated question and answer information index tree.

[0007] In a second aspect, some embodiments of the present disclosure provide a question and answer information processing device based on permission labels, the device comprising: an acquisition unit, configured to acquire a question and answer information set, wherein the question and answer information in the above question and answer information set corresponds to different question and answer information bases that are data-isolated from each other, and the question and answer information comprises: a question and answer information source, question and answer content information and a question and answer information label, wherein the question and answer information label represents the access rights of the question and answer information; an information feature extraction unit, configured to extract question and answer information features from the question and answer content information included in each question and answer information in the above question and answer information set to generate question and answer information features; a heterogeneous question and answer information merging unit, configured to merge heterogeneous question and answer information in the above question and answer information set according to the question and answer information features corresponding to the question and answer information, to generate question and answer information features; Generate a set of question and answer information groups; an execution unit is configured to perform the following second processing step for each question and answer information group in the above question and answer information group set: in response to determining that the number of question and answer information included in the above question and answer information group is greater than the target number, based on the question and answer information source and question and answer information label included in the question and answer information in the above question and answer information group, generate an updated question and answer information label for the above question and answer information group; based on the above updated question and answer information label and the above question and answer information group, perform index update on the question and answer information index tree corresponding to the virtual question and answer information library to obtain an updated question and answer information index tree; in response to determining that the number of question and answer information included in the above question and answer information group is equal to the above target number, perform index update on the above question and answer information index tree based on the above question and answer information group to obtain an updated question and answer information index tree.

[0008] 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.

[0009] 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.

[0010] The above-mentioned various embodiments of the present disclosure have the following beneficial effects: through the question and answer information processing method based on permission labels of some embodiments of the present disclosure, the waste of storage resources is reduced. Specifically, the reason for the waste of storage resources is that for multiple question and answer information bases, when there are repeated files for question and answer information extraction, it often leads to repeated extraction and storage of question and answer information, which may cause waste of computing resources and storage resources. Based on this, the question and answer information processing method based on permission labels of some embodiments of the present disclosure first obtains a question and answer information set, wherein the question and answer information in the above-mentioned question and answer information set corresponds to different question and answer information bases with data isolated from each other, and the question and answer information includes: question and answer information source, question and answer content information and question and answer information label, wherein the question and answer information label represents the access rights of the question and answer information. In this way, the question and answer information to be merged and stored is obtained. Then, the question and answer information feature extraction is performed on the question and answer content information included in each question and answer information in the above-mentioned question and answer information set to generate a question and answer information feature. Through the feature extraction of question and answer information, firstly, the question and answer information can be refined at the feature granularity, and secondly, the question and answer information of different question and answer information repositories can be mapped to the same feature space. Further, according to the question and answer information features corresponding to the question and answer information, the question and answer information in the above question and answer information set is merged with heterogeneous question and answer information to generate a question and answer information group set. In this way, the question and answer information with different sources but the same content is merged. In addition, for each question and answer information group in the above question and answer information group set, the following first processing step is performed: the first step, in response to determining that the number of question and answer information included in the above question and answer information group is greater than the target number, according to the question and answer information source and question and answer information label included in the question and answer information in the above question and answer information group, an updated question and answer information label for the above question and answer information group is generated. That is, when the question and answer information group contains at least one question and answer information from a different question and answer information repositories, it is necessary to update the overall question and answer information label of the question and answer information group according to the question and answer information label in the question and answer information group. The second step is to update the index of the question and answer information index tree corresponding to the virtual question and answer information library according to the above-mentioned updated question and answer information label and the above-mentioned question and answer information group, and obtain the updated question and answer information index tree. By updating the question and answer information index tree, the question and answer information from different question and answer information libraries can be stored as a whole, avoiding the problem of waste of storage resources caused by independent storage when there is information redundancy. The third step is in response to determining that the number of question and answer information included in the above-mentioned question and answer information group is equal to the above-mentioned target number, and according to the above-mentioned question and answer information group, the index of the above-mentioned question and answer information index tree is updated to obtain the updated question and answer information index tree. In practice, when the number of question and answer information included in the question and answer information group is equal to the above-mentioned target number, that is, the question and answer information included in the characterization question and answer information group is all from the same question and answer information library, the question and answer information index tree can be directly updated at this time. By merging and storing the question and answer information in multiple question and answer information libraries in this way, the utilization rate of storage resources is greatly improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] 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.

[0012] Figure 1 is a flowchart of some embodiments of the question and answer information processing method based on authority tags according to the present disclosure;

[0013] Figure 2 It is a schematic diagram of the process of obtaining a set of question and answer information;

[0014] Figure 3 It is a tree structure diagram of the question and answer information source relationship tree;

[0015] Figure 4 It is a structural schematic diagram of some embodiments of the question-answer information processing device based on permission labels according to the present disclosure;

[0016] Figure 5 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] The present disclosure will be described in detail below with reference to the accompanying drawings and in conjunction with embodiments.

[0023] Referring to FIG. 1 , a process 100 of some embodiments of the question-answer information processing method based on permission tags according to the present disclosure is shown. The question-answer information processing method based on permission tags comprises the following steps:

[0024] Step 101, obtaining a question and answer information set.

[0025] In some embodiments, the execution subject (for example, a computing device) of the question-and-answer information processing method based on the permission label can obtain the question-and-answer information set through a wired connection or a wireless connection. Among them, the question-and-answer information in the question-and-answer information set corresponds to different question-and-answer information bases that are data-isolated from each other. The question-and-answer information includes: a question-and-answer information source, question-and-answer content information, and a question-and-answer information label, wherein the question-and-answer information label represents the access rights of the question-and-answer information. The question-and-answer information source represents the source of the question-and-answer information. In practice, the question-and-answer information source may include: a question-and-answer information base address and a file address. Among them, the question-and-answer information base address represents the address of the question-and-answer information base that temporarily stores the question-and-answer information. The file address represents the storage address of the file used to refine the question-and-answer information. The question-and-answer content information is stored in the form of a key-value pair. In practice, since the question-and-answer information bases corresponding to the question-and-answer information are data-isolated from each other, if the question-and-answer information of different question-and-answer information bases is directly merged into a question-and-answer information set, information leakage may occur. Therefore, when obtaining the question-and-answer information of different question-and-answer information bases, the question-and-answer content included in the question-and-answer information will be encrypted.

[0026] It should be noted that the above-mentioned wireless connection methods may include but are not limited to 3G / 4G / 5G connection, WiFi connection, Bluetooth connection, WiMAX connection, Zigbee connection, UWB (ultra wideband) connection, and other wireless connection methods currently known or to be developed in the future.

[0027] It should be noted that the above-mentioned computing device can be hardware or software. When the computing device is hardware, it can be implemented as a distributed cluster consisting of multiple servers or terminal devices, or it can be implemented as a single server or a single terminal device. When the computing device is embodied as software, it can be installed in the hardware devices listed above. It can be implemented as multiple software or software modules for providing distributed services, for example, or it can be implemented as a single software or software module. No specific limitation is made here. It should be understood that the number of computing devices can be any number according to the implementation needs.

[0028] As an example, see Figure 2 The schematic diagram of the process of obtaining the question and answer information set is shown in FIG. Figure 2 It includes a question and answer information library 201 and a question and answer information library 202, wherein the data between the question and answer information library 201 and the question and answer information library 202 are isolated. For example, the question and answer information library 201 may include: question and answer information A1, question and answer information A2. The question and answer information library 202 may include: question and answer information A3, question and answer information A4. In order to avoid directly sending the question and answer information A1, question and answer information A2, question and answer information A3, and question and answer information A4 as a question and answer information set to the above-mentioned execution subject, thereby causing the problem of information leakage. Therefore, the question and answer content information included in the question and answer information A1 and the question and answer content information included in the question and answer information A2 both need to be encrypted through the encryption model M1. And the question and answer content information included in the question and answer information A3 and the question and answer content information included in the question and answer information A2 both need to be encrypted through the encryption model M2, so as to obtain question and answer information B1, question and answer information B2, question and answer information B3, and question and answer information B4 as the question and answer information set 203. Among them, the model structures of the encryption model M1 and the encryption model M2 are the same, and both use the Word2Vec model. However, in the model training stage, different training samples are used for training. For example, the encryption model M1 is trained with the question and answer information corresponding to the question and answer information library 201 and annotated with information. The encryption model M2 is trained with the question and answer information corresponding to the question and answer information library 202 and annotated with information. Since the model parameters of the encryption model M1 and the encryption model M2 are different, it is impossible to directly infer the unencrypted question and answer content information, so as to achieve the purpose of information protection. At the same time, the encryption is performed by feature extraction, which further facilitates the subsequent further feature extraction (question and answer information feature extraction). Compared with the encryption and decryption method based on the key, there is no need to perform decryption operation before information processing (question and answer information feature extraction), which avoids the waste of computing resources corresponding to the decryption process, and at the same time, avoids the risk of information leakage from the stage after decryption to the stage before data processing.

[0029] Step 102: Extract question and answer information features from the question and answer content information included in each question and answer information in the question and answer information set to generate question and answer information features.

[0030] In some embodiments, the above-mentioned execution entity may extract question and answer information features from the question and answer content information included in each question and answer information in the question and answer information set to generate question and answer information features. Optionally, the question and answer information in the question and answer information set includes: inquiry information and reply information. Among them, the inquiry information and reply information are stored in the form of key-value pairs. Specifically, since the question and answer information of different question and answer information libraries are encrypted by the corresponding encryption model before being sent to the above-mentioned execution entity, the reply information included in the question and answer information is stored in the form of a feature vector.

[0031] In some optional implementations of some embodiments, the execution subject extracts question and answer information features from the question and answer content information included in each question and answer information in the question and answer information set to generate question and answer information features, which may include the following steps:

[0032] The first step is to activate a query information feature extraction model that matches the query information included in the question and answer information.

[0033] Among them, the information types of the above-mentioned inquiry information include: text information type and audio information type. In practice, the inquiry information feature extraction model includes: an inquiry information feature extraction model for extracting features from inquiry information of text information type, and an inquiry information feature extraction model for extracting features from inquiry information of audio information type. Specifically, the inquiry information feature extraction model for extracting features from inquiry information of text information type adopts the TextCNN (Text Convolutional Neural Network) model. The inquiry information feature extraction model for extracting features from inquiry information of audio information type adopts the Bi-LSTM (Bidirectional Long Short-Term Memory) model.

[0034] In the second step, in response to successful activation, the query information included in the question and answer information is extracted through the query information feature extraction model to obtain the query information feature.

[0035] The third step is to activate the answer information feature extraction model that matches the answer information included in the above question and answer information.

[0036] Among them, the information types of the above reply information include: file information type, text information type, image information type and audio information type. In practice, the reply information feature extraction model includes: a reply information feature extraction model for feature extraction of reply information of file information type, a reply information feature extraction model for feature extraction of reply information of text information type, a reply information feature extraction model for feature extraction of reply information of image information type, and a model for feature extraction of reply information of audio information type. The reply information feature extraction model for feature extraction of reply information of text information type can adopt the TextCNN model, and the reply information feature extraction model for feature extraction of reply information of image information type can adopt the ResNet (Residual Network) model. The reply information feature extraction model for feature extraction of reply information of audio information type can adopt the Bi-LSTM model. The reply information feature extraction model for feature extraction of reply information of file information type can adopt a multimodal model composed of the TextCNN model, the ResNet model and the Bi-LSTM model. When the TextCNN model, the ResNet model, and the Bi-LSTM model constitute a multimodal model, a common cross-attention mechanism layer can be connected after the TextCNN model and the ResNet model, and a common cross-attention mechanism layer can be connected after the TextCNN model and the Bi-LSTM model. A concatenation layer (Concat layer) is connected after the two cross-attention mechanism layers. In addition, in order to further improve the reuse of the model, for reply information of text information type, reply information of image information type, and reply information of audio information type, the TextCNN model, ResNet model, and Bi-LSTM model in the reply information feature extraction model for feature extraction of reply information of file information type can be used for feature extraction.

[0037] The fourth step, in response to successful activation, extracts reply information features from the reply information included in the question and answer information through the reply information feature extraction model to obtain reply information features.

[0038] In the fifth step, the inquiry information features and the reply information features are respectively downsampled to obtain downsampled inquiry information features and downsampled reply information features.

[0039] In practice, two downsampling networks are set after the concatenation layer, including: downsampling network A and downsampling network B. Among them, downsampling network A is used to downsample the query information features to obtain downsampled query information features. Downsampling network B is used to downsample the reply information features to obtain downsampled reply information features.

[0040] The sixth step is to perform feature splicing on the query information features after downsampling and the answer information features after downsampling as the question and answer information features corresponding to the question and answer content information.

[0041] Step 103: Based on the question and answer information features corresponding to the question and answer information, the question and answer information in the question and answer information set is merged with the heterogeneous question and answer information to generate a question and answer information group set.

[0042] In some embodiments, the above execution state can merge heterogeneous question and answer information in the question and answer information set in various ways according to the question and answer information features corresponding to the question and answer information to generate a question and answer information group set.

[0043] In some optional implementations of some embodiments, the execution subject merges heterogeneous question and answer information in the question and answer information set according to question and answer information features corresponding to the question and answer information to generate a question and answer information group set, which may include the following steps:

[0044] The first step is to determine at least one initial cluster center based on the question and answer information set.

[0045] The number of initial cluster centers in at least one initial cluster center is less than the number of question-answer information in the question-answer information set. In practice, the number of initial cluster centers can be determined by cross-validation.

[0046] In the second step, based on at least one initial cluster center, the following merging steps are performed:

[0047] The first sub-step is to select a question and answer information from the question and answer information set as candidate question and answer information.

[0048] As an example, the question and answer information set may include: question and answer information B1, question and answer information B2, question and answer information B3, question and answer information B4, and question and answer information B5, wherein the candidate question and answer information may be B5.

[0049] The second sub-step is to determine the feature distance between the question and answer information feature corresponding to the candidate question and answer information and each of the initial cluster centers in the at least one initial cluster center.

[0050] In practice, the initial cluster center may be a question and answer information in a pre-selected question and answer information set. Therefore, when calculating the feature distance, the feature distance between the question and answer information features corresponding to the question and answer information corresponding to the initial cluster center and the question and answer information features corresponding to the candidate question and answer information may be calculated.

[0051] In the third sub-step, based on the characteristic distance, a cluster center that meets the characteristic distance condition is selected from at least one initial cluster center as the target cluster center.

[0052] Among them, the feature distance condition is: the corresponding feature distance is the shortest.

[0053] In a fourth sub-step, in response to determining that the question and answer information source included in the candidate question and answer information is different from the question and answer information source included in the cluster corresponding to the target cluster center, the candidate question and answer information is divided into the cluster corresponding to the target cluster center.

[0054] In practice, in response to the question and answer information source included in the candidate question and answer information being different from the question and answer information source included in the cluster corresponding to the target cluster center, the above-mentioned execution entity can divide the candidate question and answer information into the cluster corresponding to the target cluster center.

[0055] In a fifth sub-step, in response to determining that the question and answer information set after removing the candidate question and answer information is empty, for each initial cluster center in at least one initial cluster center, the initial cluster center is updated according to the question and answer information in the cluster corresponding to the initial cluster center to obtain an updated initial cluster center.

[0056] In practice, the above-mentioned execution entity can recalculate the centroid as the updated initial cluster center based on the question and answer information features corresponding to the question and answer information in the cluster corresponding to the initial cluster center.

[0057] In the sixth sub-step, in response to determining that at least one updated initial cluster center is the same as at least one initial cluster center, the question and answer information in the cluster corresponding to each cluster center in the at least one cluster center is determined as a question and answer information group to obtain the above-mentioned question and answer information group set.

[0058] In the third step, in response to determining that the at least one updated initial cluster center is different from the at least one initial cluster center, the at least one updated initial cluster center is used as the at least one initial cluster center, and the merging step is performed again.

[0059] Step 104: for each question and answer information group in the question and answer information group set, perform the following first processing step:

[0060] Step 1041, in response to determining that the number of question and answer information included in the question and answer information group is greater than the target number, generating an updated question and answer information tag for the question and answer information group according to the question and answer information source and question and answer information tag included in the question and answer information group.

[0061] In some embodiments, the execution subject may generate an updated question and answer information label for the question and answer information group in response to determining that the number of question and answer information included in the question and answer information group is greater than the target number, based on the question and answer information source and question and answer information label included in the question and answer information in the question and answer information group. The updated question and answer information representation is the access permission label for the entire question and answer information group. The target number is 1. That is, when the number is greater than 1, it represents that the question and answer information group contains at least 2 question and answer information from different sources.

[0062] As an example, the question and answer information group includes: question and answer information B1, question and answer information B2, and question and answer information B3. Among them, question and answer information B1 corresponds to question and answer information source S1, question and answer information B2 corresponds to question and answer information source S2, and question and answer information B3 corresponds to question and answer information source S3. Among them, question and answer information source S1, question and answer information source S2, and question and answer information source S3 are of different sources. The question and answer information label of question and answer information B1 is "queryable", the question and answer information label of question and answer information B2 is "queryable and modifiable", and the question and answer information label of question and answer information B1 is "queryable and downloadable". Therefore, the above-mentioned execution entity can generate a list of question and answer information sources and question and answer information labels as access permission labels.

[0063] In some optional implementations of some embodiments, the execution entity generates an updated question and answer information tag for the question and answer information group according to the question and answer information source and the question and answer information tag included in the question and answer information group, which may include the following steps:

[0064] The first step is to generate a question and answer information source relationship tree for the question and answer information group according to the source dependency relationship between the question and answer information sources included in the question and answer information in the question and answer information group.

[0065] As an example, Figure 3 The tree structure diagram of the question and answer information source relationship tree shown in the figure, wherein the question and answer information sources often do not exist in isolation. For example, question and answer information source S2, question and answer information source S3 and question and answer information source S4 have the same parent level, which is question and answer information source S1, that is, question and answer information source S2, question and answer information source S3 and question and answer information source S4 are at the same level. For another example, the parent level of question and answer information source S5 is question and answer information source S2. For another example, the parent level of question and answer information source S6 is question and answer information source S3. For another example, the parent level of question and answer information source S7 is question and answer information S6. Therefore, according to the dependency relationship between question and answer information, the following can be constructed: Figure 3 The question and answer information source relationship tree shown.

[0066] In the second step, based on the question and answer information source relationship tree and the question and answer information tags included in the question and answer information in the question and answer information group, the question and answer information tags are transferred and updated to obtain updated question and answer information tags for the question and answer information group.

[0067] In practice, for the same question and answer content information, the access rights represented by the question and answer information tag corresponding to the previous level question and answer information source should be greater than or equal to the access rights represented by the question and answer information tag corresponding to the next level question and answer information source.

[0068] As an example, the question-and-answer information group includes: question-and-answer information B1, question-and-answer information B2, and question-and-answer information B3. Among them, question-and-answer information B1 corresponds to question-and-answer information source S1, question-and-answer information B2 corresponds to question-and-answer information source S2, and question-and-answer information B3 corresponds to question-and-answer information source S3. Among them, question-and-answer information source S1, question-and-answer information source S2, and question-and-answer information source S3 are of different sources. The question-and-answer information label of question-and-answer information B1 is "queryable", the question-and-answer information label of question-and-answer information B2 is "queryable and modifiable", and the question-and-answer information label of question-and-answer information B1 is "queryable and downloadable". Since question and answer information source S1 is the upper level of question and answer information source S2 and question and answer information source S3, for the same question and answer content information, the permissions of question and answer information source S2 and question and answer information source S3 should be less than or equal to question and answer information source S1. Therefore, the above-mentioned execution entity can update the question and answer information tags of question and answer information source S1, question and answer information source S2 and question and answer information source S2 from top to bottom according to the question and answer information source relationship tree, that is, synchronize the question and answer information tags of question and answer information source S2 and question and answer information source S2 with the question and answer information tag of question and answer information source S1 as the updated question and answer information tag.

[0069] Step 1042: update the question and answer information index tree corresponding to the virtual question and answer information database according to the updated question and answer information tag and question and answer information group to obtain an updated question and answer information index tree.

[0070] In some embodiments, the execution subject may update the index of the question and answer information index tree corresponding to the virtual question and answer information library according to the updated question and answer information tag and the question and answer information group to obtain an updated question and answer information index tree. The virtual question and answer information library may be a distributed library for storing question and answer information in question and answer information libraries with data isolation from each other. By constructing a virtual question and answer information library, the problem of repeated storage of repeated question and answer information in different question and answer information libraries may be effectively avoided. At the same time, a unified interface may be provided to different users through the virtual question and answer information library to access and update question and answer information, which reduces operation and maintenance costs compared to setting up multiple virtual question and answer information libraries.

[0071] In some optional implementations of some embodiments, the execution subject updates the index of the question and answer information index tree corresponding to the virtual question and answer information library according to the updated question and answer information tag and the question and answer information group to obtain the updated question and answer information index tree, which may include the following steps:

[0072] The first step is to determine the update position of the question and answer information group in the question and answer information index tree according to the question and answer relationship dependency of the question and answer information group in the question and answer information index tree.

[0073] In practice, the above-mentioned execution subject can locate the position of the file corresponding to the above-mentioned question and answer information in the question and answer information index tree. Among them, the question and answer information index tree is an index tree used to index the question and answer information extracted from different question files. For example, the question and answer information index tree corresponds to file A, file B, and file C. Among them, file A and file B are regulatory files. File C is a non-regulatory file. Question and answer information B1 comes from file A. Question and answer information B2 comes from file B, and question and answer information B3 comes from file 3. Therefore, file A and file B correspond to the tree node of "regulatory files". File C corresponds to the tree node of "non-regulatory files". Question and answer information B1 is the child node of the tree node corresponding to file C. Question and answer information B2 is the child node of the tree node corresponding to file B. Question and answer information B3 is the child node of the tree node corresponding to file C. Therefore, the above-mentioned execution subject can locate the update position of the above-mentioned question and answer information group in the above-mentioned question and answer information index tree by deep traversal.

[0074] In the second step, at the update position, the question and answer information index tree is updated according to the question and answer information group to obtain a candidate question and answer information index tree.

[0075] In practice, the above execution entity can insert a tree node at the update position and associate it with the question and answer information group.

[0076] The third step is to take the above-mentioned updated position as the starting point and, based on the above-mentioned updated question and answer information label, update the question and answer information in the above-mentioned candidate question and answer information index tree with the above-mentioned question and answer information group as the parent node to obtain the above-mentioned updated question and answer information index tree.

[0077] In practice, since there is a dependency relationship between question and answer information, it is necessary to update the question and answer information labels of the question and answer information corresponding to the tree nodes on the subtree starting with the parent node according to the updated question and answer information labels of the question and answer information corresponding to the parent node. For example, the updated question and answer information label of question and answer information A is "queryable", and question and answer information B and question and answer information C are new question and answer information derived from question and answer information A. Therefore, it is necessary to update the question and answer information labels of question and answer information B and question and answer information C according to the updated question and answer information label "queryable" of question and answer information A.

[0078] Step 1043, in response to determining that the number of question and answer information included in the question and answer information group is equal to the target number, the question and answer information index tree is updated according to the question and answer information group to obtain an updated question and answer information index tree.

[0079] In some embodiments, in response to determining that the number of question and answer information included in the question and answer information group is equal to the target number, the question and answer information index tree is indexed and updated according to the question and answer information group to obtain an updated question and answer information index tree. Among them, the target number is 1, that is, when the number of question and answer information in the question and answer information group is 1, the question and answer information group only contains one question and answer information for one question and answer information source. Therefore, the above-mentioned execution entity can update the question and answer information index tree according to the question and answer information tag of the only question and answer information included in the question and answer information group to obtain an updated question and answer information index tree. Specifically, refer to the step "in some optional implementation methods" in step 1042 to update the index of the question and answer information index tree.

[0080] Optionally, the above method further includes:

[0081] The first step is to respond to receiving a question-answering request initiated by a target client.

[0082] The question-and-answer request includes: target query information and a target user identifier, wherein the target user identifier is an identity identifier of a user bound to the target client.

[0083] In practice, the above-mentioned execution entity can intercept the request initiated by the client through the NGINX server, so as to achieve the purpose of receiving the question and answer request initiated by the target client.

[0084] The second step is to determine the access rights that match the target user ID.

[0085] In practice, the execution subject can determine the access rights that match the target user ID through the access rights decision tree, where the access rights decision tree records the access rights of different users.

[0086] The third step is to create a virtual access object that matches the above access permissions.

[0087] In practice, a virtual access object is an object dedicated to accessing a virtual question and answer information database. For each physical user, when accessing a virtual question and answer information database, a corresponding virtual access object needs to be allocated. This ensures the unified call of the access interface and weakens the differences between different physical users.

[0088] The fourth step is to send a dynamic password to the above target client for identity authentication.

[0089] In practice, the above-mentioned execution entity may send a verification code to the target client or use biometric verification to perform identity verification.

[0090] Step 5: In response to the identity authentication being passed, the virtual access object is bound to the target client.

[0091] In practice, the above-mentioned execution entity can assign the virtual access object to the target client, that is, when the target client has not stopped accessing the virtual question and answer information library, or the virtual access object has not been cancelled, only the target client is run to access the virtual question and answer information library through the virtual access object.

[0092] Step 6: In response to successful binding, the question and answer information is filtered on the updated question and answer information index tree according to the access rights through the virtual access object to obtain a filtered question and answer information set.

[0093] In practice, the execution subject may use a breadth traversal method to filter the question and answer information on the updated question and answer information index tree according to the access rights to obtain a filtered question and answer information set. Since the access rights of a child tree node are not greater than the access rights of the parent node, using a breadth traversal method can increase the traversal speed.

[0094] The seventh step is to determine the information matching degree between each filtered question and answer information in the filtered question and answer information set and the target query information.

[0095] In practice, the above-mentioned execution entity can determine the information matching degree between the filtered question and answer information and the above-mentioned target inquiry information by calculating the feature matching degree.

[0096] In the eighth step, according to the information matching degree corresponding to the filtered question and answer information, the filtered question and answer information that meets the filtering conditions is filtered out from the above filtered question and answer information set as the target question and answer information.

[0097] Among them, the screening condition is: the corresponding information matching degree is the first K information matching degrees.

[0098] The ninth step is to send the question and answer content information included in the target question and answer information to the target client.

[0099] Optionally, first, the question and answer information can be stored in the virtual question and answer information library in an encrypted manner to ensure data security. Secondly, when the user accesses, the key is distributed to the virtual access object through the key center. Next, considering the situation where the corresponding question and answer information source is a question and answer information group of heterogeneous sources, that is, one question and answer information corresponds to multiple question and answer information sources, that is, corresponds to multiple possible access users from different sources, in addition to the key management and distribution through the key center, when the corresponding question and answer information source is a question and answer information group of heterogeneous sources and is modified, a voting mechanism is used to modify it. To ensure the consistency of the modified question and answer information.

[0100] The above-mentioned various embodiments of the present disclosure have the following beneficial effects: through the question and answer information processing method based on permission labels of some embodiments of the present disclosure, the waste of storage resources is reduced. Specifically, the reason for the waste of storage resources is that for multiple question and answer information bases, when there are repeated files for question and answer information extraction, it often leads to repeated extraction and storage of question and answer information, which may cause waste of computing resources and storage resources. Based on this, the question and answer information processing method based on permission labels of some embodiments of the present disclosure first obtains a question and answer information set, wherein the question and answer information in the above-mentioned question and answer information set corresponds to different question and answer information bases with data isolated from each other, and the question and answer information includes: question and answer information source, question and answer content information and question and answer information label, wherein the question and answer information label represents the access rights of the question and answer information. In this way, the question and answer information to be merged and stored is obtained. Then, the question and answer information feature extraction is performed on the question and answer content information included in each question and answer information in the above-mentioned question and answer information set to generate a question and answer information feature. Through the feature extraction of question and answer information, firstly, the question and answer information can be refined at the feature granularity, and secondly, the question and answer information of different question and answer information repositories can be mapped to the same feature space. Further, according to the question and answer information features corresponding to the question and answer information, the question and answer information in the above question and answer information set is merged with heterogeneous question and answer information to generate a question and answer information group set. In this way, the question and answer information with different sources but the same content is merged. In addition, for each question and answer information group in the above question and answer information group set, the following first processing step is performed: the first step, in response to determining that the number of question and answer information included in the above question and answer information group is greater than the target number, according to the question and answer information source and question and answer information label included in the question and answer information in the above question and answer information group, an updated question and answer information label for the above question and answer information group is generated. That is, when the question and answer information group contains at least one question and answer information from a different question and answer information repositories, it is necessary to update the overall question and answer information label of the question and answer information group according to the question and answer information label in the question and answer information group. The second step is to update the index of the question and answer information index tree corresponding to the virtual question and answer information library according to the updated question and answer information label and the question and answer information group, and obtain the updated question and answer information index tree. By updating the question and answer information index tree, the question and answer information from different question and answer information libraries can be stored as a whole, avoiding the problem of waste of storage resources caused by independent storage when there is information redundancy. The third step is to update the index of the question and answer information index tree according to the question and answer information group in response to determining that the number of question and answer information included in the question and answer information group is equal to the target number, and obtain the updated question and answer information index tree. In practice, when the number of question and answer information included in the question and answer information group is equal to the target number, that is, the question and answer information included in the question and answer information group is derived from the same question and answer information library, the question and answer information index tree can be directly updated at this time. By merging and storing the question and answer information in multiple question and answer information libraries in this way, the utilization rate of storage resources is greatly improved.

[0101] Further references Figure 4 As an implementation of the methods shown in the above figures, the present disclosure provides some embodiments of a question-answer information processing device based on permission labels. These device embodiments are similar to Figure 1 Corresponding to the method embodiments shown, the question-and-answer information processing device based on permission labels can be specifically applied to various electronic devices.

[0102] like Figure 4 As shown, some embodiments of the question and answer information processing device 400 based on permission labels include: an acquisition unit 401, an information feature extraction unit 402, a heterogeneous question and answer information merging unit 403 and an execution unit 404. The acquisition unit 401 is configured to acquire a question and answer information set, wherein the question and answer information in the question and answer information set corresponds to different question and answer information bases that are data-isolated from each other, and the question and answer information includes: a question and answer information source, question and answer content information and a question and answer information label, wherein the question and answer information label represents the access rights of the question and answer information; the information feature extraction unit 402 is configured to extract question and answer information features from the question and answer content information included in each question and answer information in the question and answer information set to generate question and answer information features; the heterogeneous question and answer information merging unit 403 is configured to merge heterogeneous question and answer information from the question and answer information in the question and answer information set according to the question and answer information features corresponding to the question and answer information to generate a question and answer information group set; the execution unit 404 is configured to perform heterogeneous question and answer information merging on the question and answer information in the question and answer information set according to the question and answer information features corresponding to the question and answer information to generate a question and answer information group set; 4. It is configured to execute the following second processing step for each question and answer information group in the above-mentioned question and answer information group set: in response to determining that the number of question and answer information included in the above-mentioned question and answer information group is greater than the target number, generating an updated question and answer information tag for the above-mentioned question and answer information group according to the question and answer information source and the question and answer information tag included in the question and answer information in the above-mentioned question and answer information group; based on the above-mentioned updated question and answer information tag and the above-mentioned question and answer information group, index updating the question and answer information index tree corresponding to the virtual question and answer information library to obtain an updated question and answer information index tree; in response to determining that the number of question and answer information included in the above-mentioned question and answer information group is equal to the above-mentioned target number, index updating the above-mentioned question and answer information index tree according to the above-mentioned question and answer information group to obtain an updated question and answer information index tree.

[0103] It can be understood that the units recorded in the question-answer information processing device 400 based on the authority label are similar to the reference Figure 1 Therefore, the operations, features and beneficial effects described above for the method are also applicable to the question-answer information processing device 400 based on the permission label and the units contained therein, and will not be described in detail here.

[0104] Reference below Figure 5 , which shows a structural schematic diagram of an electronic device (eg, a computing device) 500 suitable for implementing some embodiments of the present disclosure. Figure 5 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.

[0105] like Figure 5 As shown, the electronic device 500 may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 501, which can perform various appropriate actions and processes according to a program stored in a read-only memory 502 or a program loaded from a storage device 508 to a random access memory 503. Various programs and data required for the operation of the electronic device 500 are also stored in the random access memory 503. The processing device 501, the read-only memory 502, and the random access memory 503 are connected to each other via a bus 504. An input / output interface 505 is also connected to the bus 504.

[0106] Typically, the following devices may be connected to the input / output interface 505: an input device 506 including, for example, a touch screen, a touch pad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; an output device 507 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 508 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 509. The communication device 509 may allow the electronic device 500 to communicate with other devices wirelessly or by wire to exchange data. Although Figure 5 The electronic device 500 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 5 Each block shown in the figure may represent one device, or may represent multiple devices as required.

[0107] 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 509, or installed from the storage device 508, or installed from the read-only memory 502. When the computer program is executed by the processing device 501, the above-mentioned functions defined in the method of some embodiments of the present disclosure are executed.

[0108] 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, which may 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. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer readable signal media may also be any computer readable medium other than computer readable storage media, 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 appropriate medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0109] In some embodiments, the client and the server may communicate using any currently known or future developed network protocol such as HTTP (Hyper Text 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.

[0110] The above-mentioned computer-readable medium may be included in the above-mentioned electronic device; or it may exist independently without being installed in the electronic device. The above-mentioned computer-readable medium carries one or more programs. When the above-mentioned one or more programs are executed by the electronic device, the electronic device: obtains a question and answer information set, wherein the question and answer information in the above-mentioned question and answer information set corresponds to different question and answer information libraries that are data-isolated from each other, and the question and answer information includes: question and answer information source, question and answer content information and question and answer information label, wherein the question and answer information label represents the access rights of the question and answer information; extracts question and answer information features from the question and answer content information included in each question and answer information in the above-mentioned question and answer information set to generate question and answer information features; merges heterogeneous question and answer information in the above-mentioned question and answer information set according to the question and answer information features corresponding to the question and answer information to generate a question and answer information group set. ; For each question and answer information group in the above-mentioned question and answer information group set, perform the following first processing step: in response to determining that the number of question and answer information included in the above-mentioned question and answer information group is greater than the target number, based on the question and answer information source and question and answer information label included in the question and answer information in the above-mentioned question and answer information group, generate an updated question and answer information label for the above-mentioned question and answer information group; based on the above-mentioned updated question and answer information label and the above-mentioned question and answer information group, update the index of the question and answer information index tree corresponding to the virtual question and answer information library to obtain an updated question and answer information index tree; in response to determining that the number of question and answer information included in the above-mentioned question and answer information group is equal to the above-mentioned target number, based on the above-mentioned question and answer information group, update the index of the above-mentioned question and answer information index tree to obtain an updated question and answer information index tree.

[0111] 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 the case of a remote computer, the remote computer may be connected to the user's computer through 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).

[0112] 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.

[0113] 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 comprising an acquisition unit, an information feature extraction unit, a heterogeneous question and answer information merging unit, and an execution unit. Among them, the names of these units do not constitute a limitation on the unit itself under certain circumstances. For example, the heterogeneous question and answer information merging unit may also be described as "a unit that performs heterogeneous question and answer information merging on the question and answer information in the above-mentioned question and answer information set according to the question and answer information features corresponding to the question and answer information, so as to generate a question and answer information group set".

[0114] 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.

[0115] 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 technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above inventive concept. For example, the above features are replaced with (but not limited to) technical features with similar functions disclosed in the embodiments of the present disclosure.

Claims

1. A question-answer information processing method based on authority labels, comprising: Acquire a question and answer information set, wherein the question and answer information in the question and answer information set corresponds to different question and answer information bases that are data-isolated from each other, and the question and answer information includes: a question and answer information source, question and answer content information, and a question and answer information tag, wherein the question and answer information tag represents the access permission of the question and answer information; Extracting question and answer information features from question and answer content information included in each question and answer information in the question and answer information set to generate question and answer information features; According to the question and answer information features corresponding to the question and answer information, merging the heterogeneous question and answer information in the question and answer information set to generate a question and answer information group set; For each question and answer information group in the question and answer information group set, the following first processing step is performed: In response to determining that the number of question and answer information included in the question and answer information group is greater than the target number, generating an updated question and answer information tag for the question and answer information group according to the question and answer information source and the question and answer information tag included in the question and answer information group; According to the updated question and answer information tag and the question and answer information group, the question and answer information index tree corresponding to the virtual question and answer information library is updated to obtain an updated question and answer information index tree; In response to determining that the number of question and answer information included in the question and answer information group is equal to the target number, the question and answer information index tree is indexed and updated according to the question and answer information group to obtain an updated question and answer information index tree, wherein The step of updating the index of the question and answer information index tree corresponding to the virtual question and answer information database according to the updated question and answer information tag and the question and answer information group to obtain the updated question and answer information index tree includes: Determining an update position of the question and answer information group in the question and answer information index tree according to the question and answer relationship dependency of the question and answer information group in the question and answer information index tree; At the update position, according to the question and answer information group, the question and answer information index tree is updated to obtain a candidate question and answer information index tree; Taking the update position as a starting point, and according to the updated question and answer information label, the question and answer information with the question and answer information group as a parent node in the candidate question and answer information index tree is updated to obtain the updated question and answer information index tree.

2. The method according to claim 1, wherein: The method further comprises: In response to receiving a question-and-answer request initiated by a target client, wherein the question-and-answer request includes: target inquiry information and a target user identifier, wherein the target user identifier is an identity identifier of a user bound to the target client; Determining access rights that match the target user identifier; Creating a virtual access object matching the access rights; Sending a dynamic password to the target client for identity authentication; In response to the identity authentication being passed, binding the virtual access object with the target client; In response to successful binding, filtering the question and answer information on the updated question and answer information index tree according to the access authority through the virtual access object to obtain a filtered question and answer information set; Determining the information matching degree between each filtered question and answer information in the filtered question and answer information set and the target query information; According to the information matching degree corresponding to the filtered question and answer information, filtered question and answer information satisfying the filtering condition is filtered out from the filtered question and answer information set as the target question and answer information; The question and answer content information included in the target question and answer information is sent to the target client.

3. The method according to claim 2, wherein: The question and answer information in the question and answer information set includes: inquiry information and answer information; and The step of extracting question and answer information features from the question and answer content information included in each question and answer information in the question and answer information set to generate question and answer information features includes: activating a question information feature extraction model that matches the question information included in the question-answer information, wherein the information type of the question information includes: a text information type and an audio information type; In response to successful activation, query information features are extracted from the query information included in the question-and-answer information by using the query information feature extraction model to obtain query information features; activating a reply information feature extraction model that matches the reply information included in the question and answer information, wherein the information type of the reply information includes: a file information type, a text information type, an image information type, and an audio information type; In response to successful activation, extracting answer information features from the answer information included in the question and answer information through the answer information feature extraction model to obtain answer information features; Downsampling the inquiry information features and the reply information features respectively to obtain downsampled inquiry information features and downsampled reply information features; The downsampled inquiry information features and the downsampled answer information features are concatenated to serve as question and answer information features corresponding to the question and answer content information.

4. The method according to claim 3, wherein: The step of merging heterogeneous question and answer information in the question and answer information set according to the question and answer information features corresponding to the question and answer information to generate a question and answer information group set includes: Determine at least one initial cluster center according to the question-answer information set; Based on at least one initial cluster center, perform the following merging steps: Select a question and answer information from the question and answer information set as candidate question and answer information; Determine a feature distance between a question and answer information feature corresponding to the candidate question and answer information and each of the at least one initial cluster center; According to the characteristic distance, a cluster center that meets the characteristic distance condition is selected from at least one initial cluster center as the target cluster center; In response to determining that the source of the question and answer information included in the candidate question and answer information is different from the source of the question and answer information included in the cluster corresponding to the target cluster center, classifying the candidate question and answer information into the cluster corresponding to the target cluster center; In response to determining that the question and answer information set after removing the candidate question and answer information is empty, for each initial cluster center of the at least one initial cluster center, the initial cluster center is updated according to the question and answer information in the cluster corresponding to the initial cluster center to obtain an updated initial cluster center; In response to determining that the at least one updated initial cluster center is the same as the at least one initial cluster center, determining the question and answer information in the cluster corresponding to each cluster center in the at least one cluster center as a question and answer information group, and obtaining the question and answer information group set; In response to determining that the obtained at least one updated initial cluster center is different from the at least one initial cluster center, the at least one updated initial cluster center is used as the at least one initial cluster center, and the merging step is performed again.

5. The method according to claim 4, wherein: The step of generating an updated question and answer information tag for the question and answer information group according to the question and answer information source and the question and answer information tag included in the question and answer information group includes: generating a question and answer information source relationship tree for the question and answer information group according to source dependency relationships between question and answer information sources included in the question and answer information group; According to the question and answer information source relationship tree and the question and answer information tags included in the question and answer information in the question and answer information group, the question and answer information tags are transferred and updated to obtain updated question and answer information tags for the question and answer information group.

6. A question-answer information processing device based on authority labels, comprising: an acquisition unit, configured to acquire a question-and-answer information set, wherein the question-and-answer information in the question-and-answer information set corresponds to different question-and-answer information repositories that are data-isolated from each other, and the question-and-answer information includes: a question-and-answer information source, question-and-answer content information, and a question-and-answer information tag, wherein the question-and-answer information tag represents an access permission for the question-and-answer information; an information feature extraction unit, configured to extract question and answer information features from the question and answer content information included in each question and answer information in the question and answer information set to generate question and answer information features; a heterogeneous question and answer information merging unit, configured to merge heterogeneous question and answer information in the question and answer information set according to question and answer information features corresponding to the question and answer information, so as to generate a question and answer information group set; The execution unit is configured to perform the following second processing step for each question and answer information group in the question and answer information group set: in response to determining that the number of question and answer information included in the question and answer information group is greater than the target number, generating an updated question and answer information tag for the question and answer information group according to the question and answer information source and the question and answer information tag included in the question and answer information in the question and answer information group; performing an index update on the question and answer information index tree corresponding to the virtual question and answer information library according to the updated question and answer information tag and the question and answer information group to obtain an updated question and answer information index tree; in response to determining that the number of question and answer information included in the question and answer information group is equal to the target number, performing an index update on the question and answer information index tree according to the question and answer information group to obtain an updated question and answer information index tree, wherein The step of updating the index of the question and answer information index tree corresponding to the virtual question and answer information database according to the updated question and answer information tag and the question and answer information group to obtain the updated question and answer information index tree includes: Determining an update position of the question and answer information group in the question and answer information index tree according to the question and answer relationship dependency of the question and answer information group in the question and answer information index tree; At the update position, according to the question and answer information group, the question and answer information index tree is updated to obtain a candidate question and answer information index tree; Taking the update position as a starting point, and according to the updated question and answer information label, the question and answer information with the question and answer information group as a parent node in the candidate question and answer information index tree is updated to obtain the updated question and answer information index tree.

7. 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 5.

8. 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 5 is implemented.

Images