Question bank updating method and device, electronic equipment and storage medium

Abstract

This disclosure discloses a method, apparatus, electronic device, and storage medium for updating a question bank, relating to the field of data processing technology. The method includes: obtaining first description information of candidate questions and corresponding first reference answers; traversing a first question bank and a second question bank based on the first description information; determining a first matching degree between the first reference answer and the second answer corresponding to the candidate question when the first question bank does not contain candidate questions but the second question bank does; updating the confidence degree of the second answer of the candidate question in the second question bank when the first matching degree is greater than a second threshold; and updating the first question bank using the candidate questions and the first reference answer when the confidence degree of the updated second answer is greater than the first threshold. Therefore, by automatically updating the first question bank based on the obtained candidate questions and corresponding first reference answers, the number of questions in the question bank is expanded, the success rate of question searching is improved, and the cost of updating the question bank is reduced.

Description

Technical Field

[0001] This disclosure relates to the field of data processing technology, and in particular to a method, apparatus, electronic device, and storage medium for updating a question bank. Background Technology

[0002] With the continuous development and improvement of computer technology, online question searching has played an important role in people's lives. Users can search for the correct answer to a question from a question bank by uploading a photo of the question or by automatically entering the question. However, if the question bank does not contain the question the user is searching for, the user will be unable to obtain the corresponding question and answer. Therefore, how to improve the success rate of users' question searching has become an urgent problem to be solved. Summary of the Invention

[0003] This disclosure aims to at least partially address one of the technical problems in the related art.

[0004] The first aspect of this disclosure proposes a method for updating a question bank, comprising:

[0005] Obtain the first description information of the candidate questions and the corresponding first reference answer;

[0006] Based on the first description information, the first question bank and the second question bank are traversed respectively, wherein the confidence level of the first answer corresponding to each first question in the first question bank is greater than or equal to the first threshold, and the confidence level of the second answer corresponding to each second question in the second question bank is less than the first threshold.

[0007] If the first question bank does not contain the candidate question, but the second question bank does contain the candidate question, determine the first matching degree between the first reference answer and the second answer corresponding to the candidate question;

[0008] If the first matching degree is greater than the second threshold, update the confidence degree of the second answer corresponding to the candidate question in the second question bank;

[0009] If the confidence level of the updated second answer is greater than the first threshold, the first question bank is updated using the candidate questions and the first reference answer.

[0010] Optionally, before obtaining the first description information of the candidate questions and the corresponding reference answers, the process includes:

[0011] Receive an answer upload request, wherein the answer upload request includes a tag for the third question;

[0012] If the label of the third question is a preset label, the third question is determined to be the candidate question.

[0013] Optional, also includes:

[0014] Obtain a question search request, wherein the search request includes second description information of the fourth question to be searched;

[0015] Based on the second description information, the first question bank and the second question bank are traversed respectively;

[0016] If the fourth question is not found in either the first question bank or the second question bank, the tag for the fourth question is determined to be the preset tag.

[0017] The fourth question is now being published.

[0018] Optionally, the step of traversing the first question bank and the second question bank based on the first description information includes:

[0019] Determine the second degree of matching between the first description information and the reference description information corresponding to each subject;

[0020] Based on each second matching degree, the subject to which the candidate question belongs is determined;

[0021] Based on the subject to which the candidate questions belong, determine the first target question bank and the second target question bank to be traversed;

[0022] Based on the first description information, the first target question bank and the second target question bank are traversed respectively.

[0023] Optional, also includes:

[0024] Obtain the labeled dataset, wherein the labeled dataset includes multiple reference questions, third description information of each reference question, and the subject to which each reference question belongs;

[0025] Based on the subject to which each reference question belongs, the multiple reference questions are clustered to determine the reference question set corresponding to each subject;

[0026] The third descriptive information in each of the reference question sets is merged to determine the type and / or content of the reference descriptive information corresponding to each subject.

[0027] Optionally, updating the first question bank using the candidate questions and the first reference answer includes:

[0028] Determine the search parameters corresponding to the candidate question;

[0029] If the search parameters meet the preset conditions, the first question bank is updated using the candidate questions and the first reference answer.

[0030] Optionally, determining the search parameters corresponding to the candidate question includes:

[0031] The initial search parameters corresponding to the candidate question are determined based on the number of times and the search frequency of the candidate question in the preset time period before the current time.

[0032] Obtain the question search log within the preset time period, wherein the question search log includes a fourth description of each search question;

[0033] Determine each of the first word segments contained in each of the fourth description information;

[0034] A reference word set is determined based on the frequency of each of the first word segments in the title search log;

[0035] The correction coefficient is determined based on the third matching degree between each second word in the first description information and each reference word in the reference word set;

[0036] The initial search parameters are corrected according to the correction coefficient to obtain the current search parameters corresponding to the candidate question.

[0037] Optionally, the first description information is text information, and the step of traversing the first question bank and the second question bank based on the first description information includes:

[0038] Determine the first hash value corresponding to the text information;

[0039] The first hash value is split to obtain multiple first sub-hash values;

[0040] Calculate the fourth matching degree of each first sub-hash value with each second sub-hash value corresponding to each first question in the first question bank, and the fifth matching degree with each third sub-hash value corresponding to each second question in the second question bank;

[0041] Based on the multiple fourth matching degrees corresponding to each first question, determine whether each first question is a candidate question;

[0042] Based on the multiple fifth matching degrees corresponding to each second question, determine whether each second question is a candidate question.

[0043] A second aspect of this disclosure provides an apparatus for updating a question bank, comprising:

[0044] The first acquisition module is used to acquire the first description information of the candidate questions and the corresponding first reference answer;

[0045] The traversal module is used to traverse the first question bank and the second question bank respectively based on the first description information, wherein the confidence level of the first answer corresponding to each first question in the first question bank is greater than or equal to the first threshold, and the confidence level of the second answer corresponding to each second question in the second question bank is less than the first threshold.

[0046] The determining module is used to determine a first matching degree between the first reference answer and the second answer corresponding to the candidate question when the first question bank does not contain the candidate question and the second question bank contains the candidate question;

[0047] The first update module is used to update the confidence level of the second answer corresponding to the candidate question in the second question bank when the first matching degree is greater than the second threshold.

[0048] The second update module is used to update the first question bank using the candidate questions and the first reference answer when the confidence level of the updated second answer is greater than the first threshold.

[0049] Optionally, the first acquisition module is specifically used for:

[0050] Receive an answer upload request, wherein the answer upload request includes a tag for the third question;

[0051] If the label of the third question is a preset label, the third question is determined to be the candidate question.

[0052] Optional, also includes:

[0053] The second acquisition module is used to acquire a question search request, wherein the search request includes the second description information of the fourth question to be searched;

[0054] The second acquisition module is further configured to traverse the first question bank and the second question bank respectively based on the second description information;

[0055] The second acquisition module is further configured to determine the tag of the fourth question as the preset tag when neither the first question bank nor the second question bank contains the fourth question.

[0056] The second acquisition module is also used to publish the fourth question.

[0057] Optionally, the traversal module is specifically used for:

[0058] Determine the second degree of matching between the first description information and the reference description information corresponding to each subject;

[0059] Based on each second matching degree, the subject to which the candidate question belongs is determined;

[0060] Based on the subject to which the candidate questions belong, determine the first target question bank and the second target question bank to be traversed;

[0061] Based on the first description information, the first target question bank and the second target question bank are traversed respectively.

[0062] Optional, also includes:

[0063] The third acquisition module is used to acquire the labeled dataset, wherein the labeled dataset includes multiple reference questions, third description information of each reference question, and the subject to which each reference question belongs;

[0064] The third acquisition module is further configured to cluster the multiple reference questions according to the subject to which each reference question belongs, so as to determine the reference question set corresponding to each subject;

[0065] The third acquisition module is further configured to integrate the third description information in each of the reference question sets to determine the type and / or content of the reference description information corresponding to each subject.

[0066] Optionally, the second update module includes:

[0067] A determining unit is used to determine the search parameters currently corresponding to the candidate question;

[0068] An update unit is used to update the first question bank using the candidate questions and the first reference answer when the search parameters meet preset conditions.

[0069] Optionally, the determining unit is specifically used for:

[0070] The initial search parameters corresponding to the candidate question are determined based on the number of times and the search frequency of the candidate question in the preset time period before the current time.

[0071] Obtain the question search log within the preset time period, wherein the question search log includes a fourth description of each search question;

[0072] Determine each of the first word segments contained in each of the fourth description information;

[0073] A reference word set is determined based on the frequency of each of the first word segments in the title search log;

[0074] The correction coefficient is determined based on the third matching degree between each second word in the first description information and each reference word in the reference word set;

[0075] The initial search parameters are corrected according to the correction coefficient to obtain the current search parameters corresponding to the candidate question.

[0076] Optionally, the first description information is text information, and the traversal module is specifically used for:

[0077] Determine the first hash value corresponding to the text information;

[0078] The first hash value is split to obtain multiple first sub-hash values;

[0079] Calculate the fourth matching degree of each first sub-hash value with each second sub-hash value corresponding to each first question in the first question bank, and the fifth matching degree with each third sub-hash value corresponding to each second question in the second question bank;

[0080] Based on the multiple fourth matching degrees corresponding to each first question, determine whether each first question is a candidate question;

[0081] Based on the multiple fifth matching degrees corresponding to each second question, determine whether each second question is a candidate question.

[0082] A third aspect of this disclosure provides an electronic device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, it implements a method for updating a question bank as proposed in a first aspect of this disclosure.

[0083] A fourth aspect of this disclosure provides a non-transitory computer-readable storage medium storing a computer program that, when executed by a processor, implements a method for updating a question bank as described in a first aspect of this disclosure.

[0084] A fifth aspect of this disclosure provides a computer program product that, when executed by an instruction processor, performs a question bank update method according to a first aspect of this disclosure.

[0085] The method, apparatus, electronic device, and storage medium for updating the question bank disclosed herein have the following beneficial effects:

[0086] In this embodiment, firstly, first description information of candidate questions and corresponding first reference answers are obtained; then, based on the first description information, a first question bank and a second question bank are traversed respectively; if the first question bank does not contain candidate questions but the second question bank does, a first matching degree between the first reference answer and the second answer corresponding to the candidate question is determined; then, if the first matching degree is greater than a second threshold, the confidence degree of the second answer corresponding to the candidate question in the second question bank is updated; finally, if the confidence degree of the updated second answer is greater than the first threshold, the first question bank is updated using the candidate questions and the first reference answers. Therefore, by automatically updating the first question bank based on the obtained candidate questions and corresponding first reference answers, not only are the number of questions in the question bank expanded and the success rate of question searching improved, but the cost of updating the question bank is also reduced.

[0087] Additional aspects and advantages of this disclosure will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this disclosure. Attached Figure Description

[0088] The above and / or additional aspects and advantages of this disclosure will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, in which:

[0089] Figure 1 This is a schematic flowchart illustrating a method for updating a question bank according to an embodiment of the present disclosure.

[0090] Figure 2 A flowchart illustrating a question bank updating method provided in another embodiment of this disclosure;

[0091] Figure 3 A flowchart illustrating a question bank updating method provided in another embodiment of this disclosure;

[0092] Figure 4 This is a schematic diagram of a process for obtaining a reference title set according to an embodiment of the present disclosure;

[0093] Figure 5 This is a schematic diagram illustrating another process for obtaining a reference title set according to an embodiment of the present disclosure;

[0094] Figure 6 A flowchart illustrating a question bank updating method provided in another embodiment of this disclosure;

[0095] Figure 7 A schematic diagram of the structure of a question bank updating device provided in an embodiment of this disclosure;

[0096] Figure 8 A schematic diagram of the structure of a question bank updating device provided in another embodiment of this disclosure;

[0097] Figure 9 A block diagram of an exemplary electronic device suitable for implementing embodiments of the present disclosure is shown. Detailed Implementation

[0098] Embodiments of this disclosure are described in detail below, examples of which are illustrated in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this disclosure, and should not be construed as limiting this disclosure.

[0099] In order to solve the problem of low success rate of user search due to the limited number of questions in the question bank, this disclosure automatically updates the question bank by combining the answers provided by users with the corresponding questions. This not only expands the number of questions in the question bank and improves the success rate of search, but also reduces the cost of updating the question bank.

[0100] The following description, with reference to the accompanying drawings, outlines a method, apparatus, electronic device, and storage medium for updating a question bank according to embodiments of the present disclosure.

[0101] Figure 1 This is a flowchart illustrating a method for updating a question bank provided in an embodiment of this disclosure.

[0102] like Figure 1 As shown, the method for updating this question bank may include the following steps:

[0103] Step 101: Obtain the first description information of the candidate questions and the corresponding first reference answer.

[0104] The candidate questions can be those searched by the user through images or entered manually; this disclosure does not limit this.

[0105] The first reference answer can be the answer corresponding to the candidate question uploaded by the user.

[0106] The first descriptive information can be of any type and can be used to uniquely represent the candidate topic. For example, it can be text information, graphic information, etc., and this disclosure does not limit it.

[0107] As an example, if the candidate question is a math question, the corresponding first descriptive information may include text information and graphic information. Alternatively, if the candidate question is a history question, the corresponding first descriptive information may be text information, etc. This disclosure does not limit this.

[0108] Step 102: Based on the first description information, traverse the first question bank and the second question bank respectively.

[0109] In the first question bank, the confidence level of the first answer corresponding to each first question is greater than or equal to the first threshold, and the confidence level of the second answer corresponding to each second question in the second question bank is less than the first threshold.

[0110] Optionally, if the first description information is text information, the cosine distance between the first description information and the description information corresponding to each first question and each second question can be calculated respectively. Then, based on the cosine distance corresponding to each first question, it can be determined whether each first question is a candidate question, and based on the cosine distance corresponding to each second question, it can be determined whether each second question is a candidate question.

[0111] In one possible implementation, since the same question may have multiple corresponding text descriptions, meaning the same question may correspond to multiple descriptions, directly matching questions based on the text information might lead to matching failures. Therefore, in this disclosure, to avoid matching failures for the same question, the description information corresponding to the question can be split first, and then matched. That is, in one possible implementation, step 102 above may include:

[0112] Determine the first hash value corresponding to the text information;

[0113] The first hash value is split to obtain multiple first sub-hash values;

[0114] Calculate the fourth matching degree of each first sub-hash value with each second sub-hash value corresponding to each first question in the first question bank, and the fifth matching degree with each third sub-hash value corresponding to each second question in the second question bank;

[0115] Based on the multiple fourth matching degrees corresponding to each first question, determine whether each first question is a candidate question;

[0116] Based on the multiple fifth-degree matching values ​​corresponding to each second question, determine whether each second question is a candidate question.

[0117] In other words, in this disclosure, the first hash value corresponding to the text information can be determined first, and then the hash value can be split to obtain multiple first sub-hash values. After that, the matching degree between the first sub-hash value and each second or third sub-hash corresponding to the question in the question bank can be calculated. Finally, based on the matching degree between each sub-hash value, it can be determined whether the question bank contains a question that matches the candidate question.

[0118] Furthermore, in this disclosure, in order to avoid mismatches due to different word order of identical content in text information, after dividing the first hash value into multiple first sub-hash values, the multiple first sub-hash values ​​can be randomly combined, and then the combined sub-hash values ​​can be matched with each question in the question bank.

[0119] For example, if the first description information is hashed, the resulting hash value is a 64-bit hash value. This 64-bit hash value is then split into four substrings of length 16, and these substrings are combined in pairs to obtain six hash keys. These six hash keys are then matched against the hash keys corresponding to each question in both the first and second question banks. Since each part of the description information appears multiple times in various combinations across the six hash keys, this greatly avoids the possibility of matching the same question's description information failing due to word order issues.

[0120] Step 103: If the first question bank does not contain candidate questions and the second question bank does contain candidate questions, determine the first matching degree between the first reference answer and the second answer corresponding to the candidate question.

[0121] Optionally, the first matching degree between the first reference answer and the second answer corresponding to the candidate question can be calculated using Euclidean distance formula, Manhattan distance, or other methods. Alternatively, the cosine similarity between the first reference answer and the second answer corresponding to the candidate question can be calculated and used as the first matching degree between the first reference answer and the second answer corresponding to the candidate question. This disclosure does not limit the scope of the method.

[0122] Step 104: If the first matching degree is greater than the second threshold, update the confidence degree of the second answer corresponding to the candidate question in the second question bank.

[0123] It should be noted that if the first matching degree is greater than the second threshold, it means that the first reference answer and the second answer corresponding to the candidate question have a high degree of similarity. This further indicates that the confidence of the second answer corresponding to the candidate question in the second question bank is higher than the previous confidence. Therefore, when the first matching degree is greater than the second threshold, the confidence of the second answer corresponding to the candidate question in the second question bank can be increased.

[0124] In this disclosure, since the confidence level of the answer corresponding to each second question in the second question bank is low, that is, the answer may not be the correct answer to the second question, in this disclosure, when a new reference answer to the second question is obtained, the newly obtained reference answer can be matched with the known second answer to update the confidence level of the second answer or update the answer to the question.

[0125] It should be noted that if the matching degree between the first reference answer and the second answer is less than the threshold, the first reference answer can also be stored in the second question bank. Thus, in the updated second question bank, the candidate question contains at least two answers, and each answer corresponds to a confidence level.

[0126] Step 105: If the confidence level of the updated second answer is greater than the first threshold, update the first question bank using candidate questions and the first reference answer.

[0127] It should be noted that if the confidence level of the second answer corresponding to the candidate question in the updated second question bank is greater than the first threshold, it means that the second answer is the correct answer to the candidate question, that is, the first reference answer is the correct answer to the candidate question. Therefore, the candidate question and its corresponding first reference answer can be added to the first question bank to expand the first question bank and improve the success rate of question searching.

[0128] Optionally, after updating the first question bank with candidate questions and the first reference answer, the candidate questions and their corresponding second answers in the second question bank can be deleted from the second question bank to free up storage space in the second question bank.

[0129] Understandably, if the confidence level of the updated second answer is still less than or equal to the first threshold, the second answer and its confidence level for that candidate question still need to be monitored and updated until it is finally determined that the confidence level of a certain answer is greater than the first threshold. At this point, the candidate question and its corresponding second answer can be added to the first question bank. Thus, not only is the first question bank automatically updated, but the questions updated in this question bank are also based on user data, making the questions in the question bank more aligned with user needs.

[0130] In this embodiment, firstly, first description information of candidate questions and corresponding first reference answers are obtained; then, based on the first description information, a first question bank and a second question bank are traversed respectively; if the first question bank does not contain candidate questions but the second question bank does, a first matching degree between the first reference answer and the second answer corresponding to the candidate question is determined; then, if the first matching degree is greater than a second threshold, the confidence degree of the second answer corresponding to the candidate question in the second question bank is updated; finally, if the confidence degree of the updated second answer is greater than the first threshold, the first question bank is updated using the candidate questions and the first reference answers. Therefore, by automatically updating the first question bank based on the obtained candidate questions and corresponding first reference answers, not only are the number of questions in the question bank expanded and the success rate of question searching improved, but the cost of updating the question bank is also reduced.

[0131] Figure 2 This is a flowchart illustrating a method for updating a question bank according to an embodiment of the present disclosure, as shown below. Figure 2As shown, the method for updating this question bank may include the following steps:

[0132] Step 201: Receive the answer upload request, wherein the answer upload request includes the tag for the third question.

[0133] The third question can be the question corresponding to the answer uploaded by the user.

[0134] The tag for the third question can be any tag used to represent the question bank to which the question belongs. For example, if the third question belongs to the first question bank, its corresponding tag can be A; if the third question belongs to the second question bank, its corresponding tag can be B, and so on. This disclosure does not impose any restrictions on this.

[0135] In this disclosure, each time a new topic is obtained, the corresponding tag for that topic can be determined based on the search results for that topic.

[0136] Optionally, if the new question is triggered by a user's answer search request, the disclosure may further include, prior to step 201 above:

[0137] Obtain the question search request, wherein the search request includes the second description information of the fourth question to be searched;

[0138] Based on the second description information, traverse the first question bank and the second question bank respectively;

[0139] If the fourth question is not found in either the first or second question bank, the fourth question is assigned the default tag.

[0140] Finally, the fourth question will be released.

[0141] The second descriptive information can be of any type and can be used to uniquely represent the fourth topic. For example, it can be text information, graphic information, etc., and this disclosure does not limit it.

[0142] The preset tags can be any tags that can be used to represent that there is no question or no answer for the question in the question bank. The specific implementation can be set as needed, and this disclosure does not limit it.

[0143] In this disclosure, the method of publishing the fourth question can be set as needed. For example, it can be provided to users as a practice question, and then the reference answer for the fourth question can be obtained based on the results submitted by the users. Alternatively, the question can be directly published in a function module such as "Answer Collection" to collect answers, etc. This disclosure does not limit this approach.

[0144] It should be noted that since each question in both the first and second question banks has a corresponding answer and confidence level, if, after traversing the first and second question banks based on the second descriptive information, it is determined that the fourth question is not included in the first question bank but is included in the second question bank, then the answer and confidence level corresponding to the fourth question can be returned to the user. Furthermore, since the confidence level of the answer in the second question bank is relatively low, this disclosure can simultaneously return the answer to the user and the confidence level of that answer, allowing the user to evaluate the answer.

[0145] Step 202: If the label of the third question is a preset label, then the third question is determined as a candidate question.

[0146] Step 203: Based on the first description information corresponding to the candidate questions, traverse the first question bank and the second question bank respectively.

[0147] Step 204: If the first question bank does not contain candidate questions but the second question bank does contain candidate questions, determine the first matching degree between the first reference answer corresponding to the candidate question and the second answer corresponding to the candidate question in the second question bank.

[0148] Step 205: If the first matching degree is greater than the second threshold, update the confidence degree of the second answer corresponding to the candidate question in the second question bank.

[0149] The specific implementation of steps 203-206 can be found in the detailed descriptions of other embodiments in this disclosure, and will not be repeated here.

[0150] Step 206: If the confidence level of the updated second answer is greater than the first threshold, determine the search parameters corresponding to the current candidate question.

[0151] The search parameters indicate the search activity of the candidate topic within a certain time period. For example, they can be the number of searches and search frequency within a preset time period. For instance, the search parameters could be the number of searches and search frequency of the candidate topic within a month; or the number of searches and search frequency of the candidate topic within a week. This disclosure does not limit this.

[0152] In this disclosure, in order to further ensure that the questions in the first question bank better reflect user needs, only questions that meet certain conditions in terms of search count and / or search frequency can be added to the first question bank.

[0153] That is, the second question bank can also store the search parameters corresponding to each second question. So when a question search request is received each time, the search parameters of the corresponding question can be updated according to the search situation and the known search parameters.

[0154] For example, if a search request for question A is received at the current moment, it can be determined through matching that question A is in the second question bank, and the search count for question A in the second question bank is 'a'. Therefore, the search count for question A can be updated to 'a+1'.

[0155] In one possible implementation, the search parameters can also be used to indicate the degree of match between the question and the user's frequently used search terms. That is, the search parameters corresponding to the candidate question can be determined using the following process:

[0156] (1) Determine the initial search parameters corresponding to the candidate topic at present based on the number of times and the search frequency of the candidate topic in the preset time period before the current time.

[0157] (2) Obtain the question search log within a preset time period, wherein the question search log includes the fourth description information of each search question.

[0158] The fourth descriptive information can be of any type and can be used to uniquely represent each search question. For example, it can be text information, graphic information, etc., and this disclosure does not limit it.

[0159] Optionally, the question search log may include not only the fourth description information of each search question, but also: the user's request number when searching for the question, the user's device number when searching for the question, the time information when searching for the question, etc. This disclosure does not limit this.

[0160] (3) Determine each first word contained in each fourth description information.

[0161] The first segment consists of words in the fourth descriptive information that reflect key information about each search topic.

[0162] For example, if the fourth descriptive information is "What is the cosine value of 30°?", then the corresponding first segmentation can be "30°" and "cosine value".

[0163] Alternatively, if the fourth descriptive information is "What is the antonym of warmth?", then the corresponding first participle can be "warm" or "antonym".

[0164] It should be noted that the above examples are merely illustrative and should not be taken as limitations on the fourth descriptive information and the first word segmentation in this disclosure.

[0165] (4) Determine the reference word set based on the frequency of each first word in the title search log.

[0166] Optionally, a frequency threshold can be set to include each first word segmented with a frequency greater than the frequency threshold in the reference word segmentation set.

[0167] (5) Optionally, the N most frequent first-order words can also be included in the reference word set.

[0168] The correction coefficient is determined based on the third matching degree between each second word in the first description information and each reference word in the reference word set.

[0169] (6) Adjust the initial search parameters according to the correction coefficient to obtain the current search parameters corresponding to the candidate questions.

[0170] Optionally, the third matching degree between each second word and each reference word can be calculated using the Euclidean distance formula or the Manhattan distance formula. Alternatively, the cosine similarity between each second word and each reference word can be calculated and used as the third matching degree between each second word and each reference word. This disclosure does not limit the scope of the disclosure.

[0171] Optionally, we can first obtain the third matching degree corresponding to each second word to determine whether the second word is a word in the reference word set. If the second word is a word in the reference word set, the correction coefficient corresponding to the question can be considered to be a positive coefficient. Thus, the initial search parameters can be updated in the direction of increasing the value to increase the probability of it being used to update the first question bank. Conversely, if each second word is not in the reference word set, we can determine to update the search parameters corresponding to the question in the direction of decreasing the value to reduce the probability of it being used to update the first question bank.

[0172] The specific method for determining the correction coefficient can be determined as needed. For example, the corresponding third matching degree can be weighted and summed based on the weight of each second word in the first description information, or the value of the correction coefficient can be determined based on the range of the third matching degree corresponding to the second word, etc. This disclosure does not limit this.

[0173] Furthermore, since the same user may initiate multiple search requests for the same question, in order to avoid affecting the accuracy of parameters such as the number of searches and frequency of each question, this disclosure may also refer to the device number and / or user number corresponding to each search request when updating the search parameters of the questions in the second question bank. That is, for the same question, a search request with the same user number or the same device number can be regarded as a single search request.

[0174] Step 207: If the search parameters meet the preset conditions, update the first question bank using candidate questions and the first reference answer.

[0175] If the search parameters meet the preset conditions, it means that the search frequency and number of searches for the candidate questions are relatively high within the preset time period, and that there is a strong correlation between them and the searches within the preset time period. Therefore, the first question bank can be updated based on the candidate questions and the first reference answer, so that users who search for the candidate questions later can obtain the corresponding correct answers.

[0176] In this embodiment, firstly, first description information of candidate questions and corresponding first reference answers are obtained. Then, based on the first description information, a first question bank and a second question bank are traversed. If the first question bank does not contain candidate questions, but the second question bank does, a first matching degree is determined between the first reference answer and the second answer corresponding to the candidate question. Next, if the first matching degree is greater than a second threshold, the confidence degree of the second answer corresponding to the candidate question in the second question bank is updated. Finally, if the updated confidence degree of the second answer is greater than the first threshold, the search parameters corresponding to the candidate question are determined. If the search parameters meet preset conditions, the first question bank is updated using the candidate questions and the first reference answer. Therefore, by automatically updating the first question bank using candidate questions and the first reference answer whose search parameters meet preset conditions within the current time period, not only are the number of questions in the question bank expanded, increasing the success rate of question searching, but the cost of updating the question bank is also reduced, and the effectiveness of the questions in the question bank is improved.

[0177] The above analysis shows that this disclosure can determine the question bank containing the candidate questions based on the first descriptive information of the candidate questions, and further update the question bank based on the candidate questions and the first reference answer. In one possible implementation, the subject to which the candidate questions belong can be determined first based on the first descriptive information of the candidate questions, and then the target question bank containing the candidate questions can be determined based on the determined subject, followed by question matching. The following section will combine... Figure 3 The above process will be further explained.

[0178] Figure 3 A flowchart illustrating a question bank updating method provided in another embodiment of this disclosure is shown below. Figure 3 As shown, the method for updating this question bank may include the following steps:

[0179] Step 301: Obtain the first description information of the candidate questions and the corresponding first reference answer.

[0180] The specific implementation of step 301 can be found in the detailed descriptions of other embodiments in this disclosure, and will not be repeated here.

[0181] Step 302: Determine the second degree of matching between the first description information and the reference description information corresponding to each subject.

[0182] It should be noted that the solution provided in any embodiment of this disclosure can be used to determine the second matching degree between the first description information and each reference description information, which will not be elaborated here.

[0183] The reference description information for each subject can be pre-set or automatically generated based on the subject's type and description, and this disclosure does not limit this.

[0184] Alternatively, you can obtain reference descriptions for each subject through the following methods:

[0185] (1) Obtain the labeled dataset, which includes multiple reference questions, the third description information of each reference question, and the subject to which each reference question belongs.

[0186] The third descriptive information can be of any type and can be used to uniquely represent each reference topic. For example, it can be text information, graphic information, etc., and this disclosure does not limit it.

[0187] It is understandable that the reference questions in the labeled dataset can be any questions whose corresponding descriptions and subjects are known.

[0188] Understandably, annotated datasets can contain reference questions for multiple subjects such as Chinese, mathematics, and English, and each subject may have multiple reference questions. Therefore, the annotated dataset can be analyzed to determine the reference description information for each subject.

[0189] (2) Cluster multiple reference questions according to the subject to which each reference question belongs, so as to determine the reference question set corresponding to each subject.

[0190] Optionally, the third description information of each reference question in the same subject can be filtered first to obtain reference questions that contain graphic information and reference questions that contain text information in the third description information.

[0191] Figure 4 This is a schematic diagram of a process for obtaining a reference topic set according to an embodiment of the present disclosure, as shown below. Figure 4As shown, for multiple reference questions containing graphic information in the third description information, K-means clustering can be used to cluster the graphic information in the multiple reference questions to obtain n1 first question clusters. Then, the reference questions in each first question cluster are filtered (the figure only shows the filtering process for first question cluster 1) to determine the first reference question set corresponding to each subject. For example, identical reference questions in first question cluster 1 can be merged into one reference question, or, based on the frequency of each reference question in first question cluster 1, the m1 reference questions that appear most frequently in first question cluster 1 can be obtained, and then the m1 reference questions can be added to the first reference question set.

[0192] Figure 5 This is a schematic diagram illustrating another process for obtaining a reference title set according to an embodiment of the present disclosure, such as... Figure 5 As shown, for multiple reference questions containing text information in the third description information, we can first calculate the hash value corresponding to the text information in each reference question, then divide and combine the hash values ​​corresponding to each text information in pairs to obtain multiple hash key values. Then, based on the multiple hash key values ​​corresponding to each text information, we can cluster the reference questions to obtain n² second question clusters.

[0193] Since each reference question corresponds to multiple hash keys, the same question may appear in multiple clusters in the clustering results. This indicates that the content of the questions within these clusters is the same and can be merged. Therefore, the second question clusters containing the multiple hash keys corresponding to the text information of each reference question are merged into one question cluster. Then, multiple reference questions within the merged question cluster are filtered within the cluster, such as deduplication and merging (the figure only shows the filtering process for question cluster 1). Optionally, the above clustering operation can be performed multiple times to ensure that there are no duplicates in the final first reference question set corresponding to each subject.

[0194] It should be noted that, finally, the sets of first and second reference question sets corresponding to the same subject can be deduplicated to obtain the reference question set for each subject.

[0195] (3) Integrate the third description information in each set of reference questions to determine the type and / or content of the reference description information for each subject.

[0196] The reference description information can be of any type and can be used to uniquely represent each subject. For example, it can be text information, graphic information, etc., and this disclosure does not limit it.

[0197] Step 303: Determine the subject to which the candidate question belongs based on each second matching degree.

[0198] Optionally, the subject corresponding to the highest value in each second matching degree is the subject to which the candidate question belongs.

[0199] Step 304: Determine the first target question bank and the second target question bank to be traversed based on the subject to which the candidate questions belong.

[0200] It should be noted that the target question bank varies depending on the subject of the candidate questions. For example, if the candidate questions belong to mathematics, then both the first and second target question banks will contain mathematics questions. Conversely, if the candidate questions belong to Chinese language and literature, then both the first and second target question banks will contain Chinese language and literature questions.

[0201] In the first target question bank, the confidence level of the first answer corresponding to each first question is greater than or equal to the first threshold, and the confidence level of the second answer corresponding to each second question in the second target question bank is less than the first threshold.

[0202] Step 305: Based on the first description information, traverse the first target question bank and the second target question bank respectively.

[0203] In this embodiment of the disclosure, the subject to which the candidate questions belong is first obtained. Then, based on the subject to which the candidate questions belong, the first target question bank and the second target question bank to be traversed are determined. Finally, based on the first description information, the first target question bank and the second target question bank are traversed respectively. Thus, the scope of the question bank to be traversed is narrowed according to the subject to which the candidate questions belong, the workload of traversing the question bank is reduced, and the efficiency of updating the question bank is improved.

[0204] Step 306: If the first target question bank does not contain candidate questions, but the second target question bank does contain candidate questions, determine the first matching degree between the first reference answer and the second answer corresponding to the candidate question.

[0205] Step 307: If the first matching degree is greater than the second threshold, update the confidence degree of the second answer corresponding to the candidate question in the second target question bank.

[0206] Step 308: If the confidence level of the updated second answer is greater than the first threshold, update the first target question bank using candidate questions and the first reference answer.

[0207] The specific implementation of steps 305-308 can be found in the detailed descriptions of other embodiments in this disclosure, and will not be repeated here.

[0208] In this embodiment, the subject to which the candidate questions belong is first obtained. Then, based on the subject to which the candidate questions belong, a first target question bank and a second target question bank to be traversed are determined. If the first matching degree between the first reference answer and the second answer corresponding to the candidate question is greater than a threshold, the confidence degree of the second answer corresponding to the candidate question in the second target question bank is updated. Finally, if the confidence degree of the updated second answer is greater than the first threshold, the first target question bank is updated using the candidate questions and the first reference answer. Thus, the scope of the question bank to be traversed is narrowed based on the subject to which the candidate questions belong. By combining the candidate questions and the corresponding first reference answer, the question bank is automatically updated, which not only reduces the workload of traversing the question bank and improves the efficiency of updating the question bank, but also expands the number of questions in the question bank, improves the success rate of searching for questions, and reduces the cost of updating the question bank.

[0209] As can be seen from the above analysis, this disclosure updates the question bank in real time based on the first description information of the candidate questions and the corresponding first reference answer. In one possible implementation, the question bank can also be updated periodically. The following section will discuss this further. Figure 6 The above process will be further explained.

[0210] Figure 6 A flowchart illustrating a question bank updating method provided in another embodiment of this disclosure is shown below. Figure 6 As shown, the method for updating this question bank may include the following steps:

[0211] Step 601: Obtain the question search log within the preset time period.

[0212] The search log includes a description of each search question and the source of the corresponding search results.

[0213] The description information of the search terms can be of any type and can be used to uniquely identify each search term. For example, it can be text information, graphic information, etc., and this disclosure does not limit it.

[0214] Step 602: Determine the search terms to be processed based on the source of each search result.

[0215] Optionally, in response to any search result coming from the first question bank, the search question corresponding to any search result is determined to be an invalid search question.

[0216] Optionally, in response to any search result coming from the second question bank, the search question corresponding to any search result is determined as the search question to be processed. The questions in the second question bank can be questions uploaded by users, and the questions in the second question bank may have no answer, or the confidence level of the corresponding answer may be lower than the first threshold.

[0217] Step 603: Based on the description information of each search question to be processed, cluster the search questions to be processed to obtain the target question.

[0218] The specific implementation of step 603 can be found in the detailed descriptions of other embodiments in this disclosure, and will not be repeated here.

[0219] Step 604: Update the question bank using the target questions.

[0220] In this embodiment of the disclosure, the question search log within a preset time period is first obtained. Then, based on the source of each search result, the search questions to be processed are determined. Next, based on the description information of each search question to be processed, the search questions to be processed are clustered to obtain the target questions. Finally, the question bank is updated using the target questions. Thus, the question bank can be updated periodically, expanding the question bank and improving the success rate of question searching.

[0221] It should be noted that you can choose to update the question bank in real time or periodically, depending on your needs.

[0222] To implement the above embodiments, this disclosure also proposes a question bank updating device.

[0223] Figure 7 This is a schematic diagram of the structure of the question bank updating device provided in the embodiments of this disclosure.

[0224] like Figure 7 As shown, the question bank update device 700 may include: a first acquisition module 710, a traversal module 720, a determination module 730, a first update module 740, and a second update module 750.

[0225] The first acquisition module 710 is used to acquire the first description information of the candidate questions and the corresponding first reference answer;

[0226] The traversal module 720 is used to traverse the first question bank and the second question bank respectively based on the first description information. In the first question bank, the confidence of the first answer corresponding to each first question is greater than or equal to the first threshold, and the confidence of the second answer corresponding to each second question in the second question bank is less than the first threshold.

[0227] The module 730 is used to determine the first matching degree between the first reference answer and the second answer corresponding to the candidate question when the first question bank does not contain candidate questions and the second question bank contains candidate questions;

[0228] The first update module 740 is used to update the confidence of the second answer corresponding to the candidate question in the second question bank when the first matching degree is greater than the second threshold.

[0229] The second update module 750 is used to update the first question bank using candidate questions and the first reference answer when the confidence level of the updated second answer is greater than the first threshold.

[0230] The functions and specific implementation principles of the modules described in this embodiment can be found in the above method embodiments, and will not be repeated here.

[0231] In this embodiment, the question bank updating device 700 first obtains first description information of candidate questions and corresponding first reference answers; then, based on the first description information, it traverses the first question bank and the second question bank respectively; if the first question bank does not contain candidate questions but the second question bank does, it determines a first matching degree between the first reference answer and the second answer corresponding to the candidate question; then, if the first matching degree is greater than a second threshold, it updates the confidence degree of the second answer corresponding to the candidate question in the second question bank; finally, if the confidence degree of the updated second answer is greater than the first threshold, it updates the first question bank using the candidate questions and the first reference answers. Therefore, by automatically updating the first question bank based on the obtained candidate questions and corresponding first reference answers, it not only expands the question bank and improves the success rate of question searching, but also reduces the cost of question bank updating.

[0232] like Figure 8 As shown, Figure 8 This is a schematic diagram of the structure of a question bank updating device provided in another embodiment of the present disclosure. The question bank updating device 800 may include: a first acquisition module 810, a traversal module 820, a determination module 830, a first update module 840, a second update module 850, a second acquisition module 860, and a third acquisition module 870.

[0233] In one possible implementation, the first acquisition module 810 is specifically used for:

[0234] Receive answer upload requests, where the answer upload request includes the tag for the third question;

[0235] If the third question is labeled with the preset label, then the third question is selected as a candidate question.

[0236] In one possible implementation, the question bank updating device 800 further includes a second acquisition module 860 and a third acquisition module 860, specifically used for:

[0237] Obtain the question search request, wherein the search request includes the second description information of the fourth question to be searched;

[0238] Based on the second description information, traverse the first question bank and the second question bank respectively;

[0239] If the fourth question is not found in either the first or second question bank, the fourth question is assigned the default tag.

[0240] The fourth question is now being released.

[0241] In one possible implementation, module 820 is traversed, specifically for:

[0242] Determine the second degree of matching between the first descriptive information and the reference descriptive information corresponding to each subject;

[0243] Based on each second matching degree, determine the subject to which the candidate question belongs;

[0244] Based on the subject to which the candidate questions belong, determine the first target question bank and the second target question bank to be traversed;

[0245] Based on the first description information, the first target question bank and the second target question bank are traversed respectively.

[0246] In one possible implementation, the question bank updating device 800 further includes a third acquisition module 870, which is specifically used for:

[0247] Obtain the labeled dataset, which includes multiple reference topics, a third description of each reference topic, and the subject to which each reference topic belongs;

[0248] Based on the subject to which each reference question belongs, multiple reference questions are clustered to determine the reference question set corresponding to each subject;

[0249] The third descriptive information from each set of reference questions is merged to determine the type and / or content of the reference descriptive information for each subject.

[0250] In one possible implementation, the second update module 850 includes:

[0251] Unit 8501 is used to determine the search parameters corresponding to the current candidate question.

[0252] The update unit 8502 is used to update the first question bank using candidate questions and the first reference answer when the search parameters meet the preset conditions.

[0253] In one possible implementation, the determining unit 8501 is specifically used for:

[0254] The initial search parameters for the current candidate topic are determined based on the number of times and frequency of searches of the candidate topic within a preset time period prior to the current time.

[0255] Obtain the question search log within a preset time period, wherein the question search log includes the fourth description information for each searched question;

[0256] Identify each first word contained in each fourth description piece of information;

[0257] The reference word set is determined based on the frequency of each first word in the question search log;

[0258] The correction coefficient is determined based on the third matching degree between each second word segment in the first description information and each reference word segment in the reference word segmentation set;

[0259] The initial search parameters are adjusted based on the correction coefficient to obtain the current search parameters corresponding to the candidate questions.

[0260] In one possible implementation, the first descriptive information is text information, and the traversal module 820 is specifically used for:

[0261] Determine the first hash value corresponding to the text information;

[0262] The first hash value is split to obtain multiple first sub-hash values;

[0263] Calculate the fourth matching degree of each first sub-hash value with each second sub-hash value corresponding to each first question in the first question bank, and the fifth matching degree with each third sub-hash value corresponding to each second question in the second question bank;

[0264] Based on the multiple fourth matching degrees corresponding to each first question, determine whether each first question is a candidate question;

[0265] Based on the multiple fifth-degree matching values ​​corresponding to each second question, determine whether each second question is a candidate question.

[0266] It is understood that this embodiment is accompanied by Figure 8 The question bank updating device 800 in the above embodiment is the same as the question bank updating device 700 in the above embodiment; the first acquisition module 810 is the same as the first acquisition module 710 in the above embodiment; the traversal module 820 is the same as the traversal module 720 in the above embodiment; the determination module 830 is the same as the determination module 730 in the above embodiment; the first update module 840 is the same as the first update module 740 in the above embodiment; and the second update module 850 is the same as the second update module 750 in the above embodiment. These devices can have the same functions and structures.

[0267] The functions and specific implementation principles of the modules described in this embodiment can be found in the above method embodiments, and will not be repeated here.

[0268] In this embodiment, the question bank updating device 800 first obtains the subject to which the candidate questions belong. Then, based on the subject to which the candidate questions belong, it determines the first target question bank and the second target question bank to be traversed. If the first matching degree between the first reference answer and the second answer corresponding to the candidate question is greater than a threshold, it updates the confidence degree of the second answer corresponding to the candidate question in the second target question bank. Finally, if the confidence degree of the updated second answer is greater than the first threshold, it updates the first target question bank using the candidate questions and the first reference answer. Thus, the scope of the question bank to be traversed is narrowed based on the subject to which the candidate questions belong. By combining the candidate questions and the corresponding first reference answer, the question bank is automatically updated, which not only reduces the workload of traversing the question bank and improves the efficiency of updating the question bank, but also expands the number of questions in the question bank, increases the success rate of searching for questions, and reduces the cost of updating the question bank.

[0269] To implement the above embodiments, this disclosure also proposes an electronic device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it implements the question bank update method proposed in the foregoing embodiments of this disclosure.

[0270] To implement the above embodiments, this disclosure also proposes a non-transitory computer-readable storage medium storing a computer program, which, when executed by a processor, implements the question bank update method as proposed in the foregoing embodiments of this disclosure.

[0271] To implement the above embodiments, this disclosure also proposes a computer program product that, when the instruction processor in the computer program product is executed, performs the question bank update method as proposed in the foregoing embodiments of this disclosure.

[0272] Figure 9 A block diagram of an exemplary electronic device suitable for implementing embodiments of the present disclosure is shown. Figure 9 The electronic device 12 shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments disclosed herein.

[0273] like Figure 9 As shown, the electronic device 12 is represented in the form of a general-purpose computing device. The components of the computer device 12 may include, but are not limited to: one or more processors or processing units 16, system memory 28, and a bus 18 connecting different system components (including system memory 28 and processing unit 16).

[0274] Bus 18 represents one or more of several bus architectures, including a memory bus or memory controller, a peripheral bus, a graphics acceleration port, a processor, or a local bus using any of the various bus architectures. Examples of these architectures include, but are not limited to, the Industry Standard Architecture (ISA) bus, the Micro Channel Architecture (MAC) bus, the Enhanced ISA bus, the Video Electronics Standards Association (VESA) local bus, and the Peripheral Component Interconnect (PCI) bus.

[0275] Computer device 12 typically includes a variety of computer system readable media. These media can be any available media that can be accessed by computer device 12, including volatile and non-volatile media, removable and non-removable media.

[0276] Memory 28 may include computer system readable media in the form of volatile memory, such as Random Access Memory (RAM) 30 and / or cache memory 32. Computer device 12 may further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, storage system 34 may be used to read and write non-removable, non-volatile magnetic media (…). Figure 9 Not shown; usually referred to as a "hard drive"). Although Figure 9 Not shown, a disk drive for reading and writing to a removable non-volatile disk (e.g., a "floppy disk") and an optical disc drive for reading and writing to a removable non-volatile optical disc (e.g., a compact disc read-only memory (CD-ROM), a digital video disc read-only memory (DVD-ROM), or other optical media) may be provided. In these cases, each drive may be connected to bus 18 via one or more data media interfaces. Memory 28 may include at least one program product having a set (e.g., at least one) of program modules configured to perform the functions of the embodiments of this disclosure.

[0277] A program / utility 40 having a set (at least one) of program modules 42 may be stored, for example, in memory 28. Such program modules 42 include, but are not limited to, an operating system, one or more application programs, other program modules, and program data. Each or some combination of these examples may include an implementation of a network environment. Program modules 42 typically perform the functions and / or methods described in the embodiments of this disclosure.

[0278] Computer device 12 can also communicate with one or more external devices 14 (e.g., keyboard, pointing device, display 24, etc.), and with one or more devices that enable a user to interact with computer device 12, and / or with any device that enables computer device 12 to communicate with one or more other computing devices (e.g., network card, modem, etc.). This communication can be performed via input / output (I / O) interface 22. Furthermore, computer device 12 can also communicate with one or more networks (e.g., local area network (LAN), wide area network (WAN), and / or public networks, such as the Internet) via network adapter 20. As shown, network adapter 20 communicates with other modules of computer device 12 via bus 18. It should be understood that, although not shown in the figures, other hardware and / or software modules can be used in conjunction with computer device 12, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0279] The processing unit 16 executes various functional applications and data processing by running programs stored in the system memory 28, such as implementing the methods mentioned in the foregoing embodiments.

[0280] The technical solution disclosed herein first obtains the first description information of candidate questions and their corresponding first reference answers; then, based on the first description information, iterates through a first question bank and a second question bank respectively; if the first question bank does not contain candidate questions but the second question bank does, determine the first matching degree between the first reference answer and the second answer corresponding to the candidate question; then, if the first matching degree is greater than a second threshold, update the confidence degree of the second answer corresponding to the candidate question in the second question bank; finally, if the confidence degree of the updated second answer is greater than the first threshold, update the first question bank using the candidate questions and the first reference answers. Therefore, by automatically updating the first question bank based on the obtained candidate questions and their corresponding first reference answers, not only is the number of questions in the question bank expanded and the success rate of question searching improved, but the cost of updating the question bank is also reduced.

[0281] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this disclosure. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0282] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this disclosure, "a plurality of" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0283] Any process or method description in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or more executable instructions for implementing custom logic functions or processes, and the scope of preferred embodiments of this disclosure includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as will be understood by those skilled in the art to which embodiments of this disclosure pertain.

[0284] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable media include: an electrical connection having one or more wires (electronic device), a portable computer disk drive (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Alternatively, the computer-readable medium may be paper or other suitable media on which the program can be printed, since the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in a computer memory.

[0285] It should be understood that various parts of this disclosure can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented using software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0286] Those skilled in the art will understand that all or part of the steps of the methods in the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, the program includes one or a combination of the steps of the method embodiments.

[0287] Furthermore, the functional units in the various embodiments of this disclosure can be integrated into a processing module, or each unit can exist physically separately, or two or more units can be integrated into a module. The integrated module can be implemented in hardware or as a software functional module. If the integrated module is implemented as a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium.

[0288] The storage medium mentioned above can be a read-only memory, a disk, or an optical disk, etc. Although embodiments of the present disclosure have been shown and described above, it is to be understood that the above embodiments are exemplary and should not be construed as limiting the present disclosure. Those skilled in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of the present disclosure.

Claims

1. A method for updating a question bank, characterized in that, include: Obtain the first description information of the candidate questions and the corresponding first reference answer; Based on the first description information, the first question bank and the second question bank are traversed respectively, wherein the confidence level of the first answer corresponding to each first question in the first question bank is greater than or equal to the first threshold, and the confidence level of the second answer corresponding to each second question in the second question bank is less than the first threshold. If the first question bank does not contain the candidate question, but the second question bank does contain the candidate question, determine the first matching degree between the first reference answer and the second answer corresponding to the candidate question; If the first matching degree is greater than the second threshold, update the confidence degree of the second answer corresponding to the candidate question in the second question bank; If the confidence level of the updated second answer is greater than the first threshold, the first question bank is updated using the candidate questions and the first reference answer. The step of traversing the first question bank and the second question bank based on the first description information includes: Determine the second degree of matching between the first description information and the reference description information corresponding to each subject; Based on each second matching degree, the subject to which the candidate question belongs is determined; Based on the subject to which the candidate questions belong, determine the first target question bank and the second target question bank to be traversed; Based on the first description information, the first target question bank and the second target question bank are traversed respectively; The method further includes: Obtain the labeled dataset, wherein the labeled dataset includes multiple reference questions, third description information of each reference question, and the subject to which each reference question belongs; Based on the subject to which each reference question belongs, the multiple reference questions are clustered to determine the reference question set corresponding to each subject; The third descriptive information in each of the reference question sets is merged to determine the type and / or content of the reference descriptive information corresponding to each subject.

2. The method as described in claim 1, characterized in that, Before obtaining the first description information of the candidate questions and the corresponding reference answers, the process includes: Receive an answer upload request, wherein the answer upload request includes a tag for the third question; If the label of the third question is a preset label, the third question is determined to be the candidate question.

3. The method as described in claim 2, characterized in that, Also includes: Obtain a question search request, wherein the search request includes second description information of the fourth question to be searched; Based on the second description information, the first question bank and the second question bank are traversed respectively; If the fourth question is not found in either the first question bank or the second question bank, the tag for the fourth question is determined to be the preset tag. The fourth question is now being published.

4. The method according to any one of claims 1-3, characterized in that, The step of updating the first question bank using the candidate questions and the first reference answer includes: Determine the search parameters corresponding to the candidate question; If the search parameters meet the preset conditions, the first question bank is updated using the candidate questions and the first reference answer.

5. The method as described in claim 4, characterized in that, Determining the search parameters corresponding to the candidate question includes: The initial search parameters corresponding to the candidate question are determined based on the number of times and the search frequency of the candidate question in the preset time period before the current time. Obtain the question search log within the preset time period, wherein the question search log includes a fourth description of each search question; Determine each of the first word segments contained in each of the fourth description information; A reference word set is determined based on the frequency of each of the first word segments in the title search log; The correction coefficient is determined based on the third matching degree between each second word in the first description information and each reference word in the reference word set; The initial search parameters are corrected according to the correction coefficient to obtain the current search parameters corresponding to the candidate question.

6. The method according to any one of claims 1-3, characterized in that, The first description information is text information. The step of traversing the first question bank and the second question bank based on the first description information includes: Determine the first hash value corresponding to the text information; The first hash value is split to obtain multiple first sub-hash values; Calculate the fourth matching degree of each first sub-hash value with each second sub-hash value corresponding to each first question in the first question bank, and the fifth matching degree with each third sub-hash value corresponding to each second question in the second question bank; Based on the multiple fourth matching degrees corresponding to each first question, determine whether each first question is a candidate question; Based on the multiple fifth matching degrees corresponding to each second question, determine whether each second question is a candidate question.

7. A question bank updating device, characterized in that, include: The first acquisition module is used to acquire the first description information of the candidate questions and the corresponding first reference answer; The traversal module is used to traverse the first question bank and the second question bank respectively based on the first description information, wherein the confidence level of the first answer corresponding to each first question in the first question bank is greater than or equal to the first threshold, and the confidence level of the second answer corresponding to each second question in the second question bank is less than the first threshold. The determining module is used to determine a first matching degree between the first reference answer and the second answer corresponding to the candidate question when the first question bank does not contain the candidate question and the second question bank contains the candidate question; The first update module is used to update the confidence level of the second answer corresponding to the candidate question in the second question bank when the first matching degree is greater than the second threshold. The second update module is used to update the first question bank using the candidate questions and the first reference answer when the confidence level of the updated second answer is greater than the first threshold. The traversal module is specifically used for: Determine the second degree of matching between the first description information and the reference description information corresponding to each subject; Based on each second matching degree, the subject to which the candidate question belongs is determined; Based on the subject to which the candidate questions belong, determine the first target question bank and the second target question bank to be traversed; Based on the first description information, the first target question bank and the second target question bank are traversed respectively; The device further includes a third acquisition module, which is specifically used for: Obtain the labeled dataset, wherein the labeled dataset includes multiple reference questions, third description information of each reference question, and the subject to which each reference question belongs; Based on the subject to which each reference question belongs, the multiple reference questions are clustered to determine the reference question set corresponding to each subject; The third descriptive information in each of the reference question sets is merged to determine the type and / or content of the reference descriptive information corresponding to each subject.

8. The apparatus as claimed in claim 7, characterized in that, The first acquisition module is specifically used for: Receive an answer upload request, wherein the answer upload request includes a tag for the third question; If the label of the third question is a preset label, the third question is determined to be the candidate question.

9. The apparatus as claimed in claim 8, characterized in that, It also includes a second acquisition module, specifically used for: Obtain a question search request, wherein the search request includes second description information of the fourth question to be searched; Based on the second description information, the first question bank and the second question bank are traversed respectively; If the fourth question is not found in either the first question bank or the second question bank, the tag for the fourth question is determined to be the preset tag. The fourth question is now being published.

10. The apparatus according to any one of claims 7-9, characterized in that, The second update module includes: A determining unit is used to determine the search parameters currently corresponding to the candidate question; An update unit is used to update the first question bank using the candidate questions and the first reference answer when the search parameters meet preset conditions.

11. The apparatus as claimed in claim 10, characterized in that, The determining unit is specifically used for: The initial search parameters corresponding to the candidate question are determined based on the number of times and the search frequency of the candidate question in the preset time period before the current time. Obtain the question search log within the preset time period, wherein the question search log includes a fourth description of each search question; Determine each of the first word segments contained in each of the fourth description information; A reference word set is determined based on the frequency of each of the first word segments in the title search log; The correction coefficient is determined based on the third matching degree between each second word in the first description information and each reference word in the reference word set; The initial search parameters are corrected according to the correction coefficient to obtain the current search parameters corresponding to the candidate question.

12. The apparatus according to any one of claims 7-9, characterized in that, The first description information is text information, and the traversal module is specifically used for: Determine the first hash value corresponding to the text information; The first hash value is split to obtain multiple first sub-hash values; Calculate the fourth matching degree of each first sub-hash value with each second sub-hash value corresponding to each first question in the first question bank, and the fifth matching degree with each third sub-hash value corresponding to each second question in the second question bank; Based on the multiple fourth matching degrees corresponding to each first question, determine whether each first question is a candidate question; Based on the multiple fifth matching degrees corresponding to each second question, determine whether each second question is a candidate question.

13. An electronic device, characterized in that, It includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, it implements the method for updating the question bank as described in any one of claims 1-6.

14. A non-transitory computer-readable storage medium is proposed, which stores a computer program that, when executed by a processor, implements the method for updating a question bank as described in any one of claims 1-6.

15. A computer program product, characterized in that, It includes a computer program that, when executed by a processor, implements the method for updating the question bank as described in any one of claims 1-6.

Images