Test structuring method and apparatus, electronic device, and storage medium

Abstract

The application discloses a test question structuring method and device, electronic equipment and a storage medium, relates to the technical field of computers, and in particular to the fields of artificial intelligence such as natural language processing and computer vision. The specific scheme is: obtaining a similar test question group, wherein the similar test question group comprises similar first test questions; extracting first structured content of each first test question in the first test questions; determining second structured content corresponding to the similar test question group according to the first structured content of the first test questions; and updating the first structured content of each first test question according to the second structured content to obtain target structured content of each first test question. Thus, based on the structured content of the test questions in the similar test question group, the structured content corresponding to the similar test question group is determined, the structured content of each test question in the similar test question group is updated using the structured content corresponding to the similar test question group, the structured content of the similar test questions is expanded, and the test question quality is improved.

Description

Technical Field

[0001] This application relates to the field of computer technology, and in particular to the fields of artificial intelligence such as natural language processing and computer vision, specifically to a test question structuring method, apparatus, electronic device, and storage medium. Background Technology

[0002] With the development of science and technology and the popularization of internet information, artificial intelligence and big data, the demand for online education is increasing, and online education cannot be separated from high-quality test resources.

[0003] In practical applications, some test questions lack structured content, such as subject, grade level, knowledge point, and chapter, which can affect the quality of the test questions. Summary of the Invention

[0004] This application provides a method, apparatus, electronic device, and storage medium for structuring test questions. The specific solutions are as follows:

[0005] According to one aspect of this application, a method for structuring test questions is provided, comprising:

[0006] Get similar question groups, where a similar question group includes multiple similar first questions;

[0007] Extract the first structured content of each first question from multiple first-question sets;

[0008] Based on the first structured content of multiple first test questions, determine the second structured content corresponding to similar test question groups;

[0009] Based on the second structured content, the first structured content of each first question is updated to obtain the target structured content of each first question.

[0010] According to another aspect of this application, a test item structuring device is provided, comprising:

[0011] The first acquisition module is used to acquire similar question groups, wherein the similar question groups include multiple similar first questions;

[0012] The extraction module is used to extract the first structured content of each first question from multiple first questions;

[0013] The first determination module is used to determine the second structured content corresponding to similar question groups based on the first structured content of multiple first question groups.

[0014] The update module is used to update the first structured content of each first question based on the second structured content, so as to obtain the target structured content of each first question.

[0015] According to another aspect of this application, an electronic device is provided, comprising:

[0016] At least one processor; and

[0017] A memory communicatively connected to the at least one processor; wherein,

[0018] The memory stores instructions that can be executed by the at least one processor, which, when executed by the at least one processor, enables the at least one processor to perform the method described in the above embodiments.

[0019] According to another aspect of this application, a non-transitory computer-readable storage medium storing computer instructions is provided, wherein the computer instructions are used to cause the computer to perform the method described in the above embodiments.

[0020] According to another aspect of this application, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps of the method described in the above embodiments.

[0021] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of this application, nor is it intended to limit the scope of this application. Other features of this application will become readily apparent from the following description. Attached Figure Description

[0022] The accompanying drawings are provided for a better understanding of this solution and do not constitute a limitation of this application. Wherein:

[0023] Figure 1 A flowchart illustrating a test question structuring method provided in an embodiment of this application;

[0024] Figure 2 A flowchart illustrating a test question structuring method provided in another embodiment of this application;

[0025] Figure 3 A flowchart illustrating a test question structuring method provided in another embodiment of this application;

[0026] Figure 4 A flowchart illustrating a test question structuring method provided in another embodiment of this application;

[0027] Figure 5 A flowchart illustrating a test question structuring method provided in another embodiment of this application;

[0028] Figure 6 A schematic diagram of a test question structuring process provided in an embodiment of this application;

[0029] Figure 7This is a schematic diagram of the structure of a test question structuring device provided in an embodiment of this application;

[0030] Figure 8 This is a block diagram of an electronic device used to implement the test question structuring method of the embodiments of this application. Detailed Implementation

[0031] The following description, in conjunction with the accompanying drawings, illustrates exemplary embodiments of this application, including various details to aid understanding. These should be considered merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of this application. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description.

[0032] The following description, with reference to the accompanying drawings, outlines a test question structuring method, apparatus, electronic device, and storage medium according to embodiments of this application.

[0033] Figure 1 This is a flowchart illustrating a test question structuring method provided in an embodiment of this application.

[0034] The test question structuring method of this application embodiment can be executed by the test question structuring device of this application embodiment. The device can be configured in an electronic device to determine the structured content corresponding to the similar test question group based on the structured content of multiple test questions in the similar test question group, and update the structured content of each test question in the similar test question group using the structured content corresponding to the similar test question group, thereby expanding the structured content of the similar test questions and improving the quality of the test questions.

[0035] Among them, electronic devices can be any device with computing capabilities, such as personal computers, mobile terminals, servers, etc. Mobile terminals can be hardware devices with various operating systems, touch screens and / or displays, such as in-vehicle devices, mobile phones, tablets, personal digital assistants, wearable devices, etc.

[0036] like Figure 1 As shown, the structured approach for this test question includes:

[0037] Step 101: Obtain similar question groups, where each similar question group includes multiple similar first questions.

[0038] In this application, similar question groups can be obtained directly from a file storing similar question groups, or they can be obtained based on the user's upload of similar questions, or a similar question search can be performed using a certain question, and similar questions can be determined based on the search results, thereby obtaining a similar question group. Other methods can also be used to obtain similar question groups, and this application does not limit this.

[0039] Step 102: Extract the first structured content of each first test question from multiple first test questions.

[0040] In this application, for each first test question in a similar test question group, extract the structured content of the first test question. Among them, the first structured content may include, but is not limited to, one or more of the stem, answer, analysis, evaluation, thinking guidance, test point, subject, school stage, grade, chapter, knowledge point, problem-solving video, etc.

[0041] In this application, the stem and answer can be regarded as a complete test question. If the stem and / or answer of the first test question is a test question with pure pictures, the pictures can be subjected to OCR (Optical Character Recognition) to obtain text, and then the structured content can be extracted. If there are serial numbers in the stem of the first test question, such as "1, 2, 3", "one, two, three", "A, B, C, D", etc., the serial numbers can be filtered out before extracting the structured content.

[0042] In this application, the structured content such as the analysis, evaluation, problem-solving test questions, subject, school stage, grade, etc. of the test question can be extracted from the answer of the first test question by means of keyword matching.

[0043] Optionally, if the first test question is a multiple-choice question, the options can be processed into a structured form according to the keyword "answer". For example, if a certain option is the correct answer, it is marked as the selected state, otherwise it is the non-selected state.

[0044] Optionally, if the answer of the first test question is not empty, but the answer content is non-substantive content such as "略", the entire test question can be discarded.

[0045] Step 103: Determine the second structured content corresponding to the similar test question group according to the first structured content of multiple first test questions.

[0046] In this application, the attribute items corresponding to the similar test question group can be determined according to the attribute items in the first structured content of multiple first test questions. For each attribute item corresponding to the similar test question group, the attribute value of each attribute item in the first structured content of multiple first test questions is used to determine the attribute value of each attribute item corresponding to the similar test question group. Then, the second structured content is determined according to the attribute items corresponding to the similar test question group and the attribute values corresponding to the attribute items.

[0047] Step 104: Update the first structured content of each first test question according to the second structured content to obtain the target structured content of each first test question.

[0048] In this application, the first structured content of each first question in a similar question group can be updated to the second structured content corresponding to the similar question group, thereby obtaining the target structured content of each first question in the similar question group, that is, the final structured content of each first question.

[0049] In this embodiment, a similar question group is obtained, and the first structured content of each first question in the multiple first questions within the similar question group is extracted. Then, based on the first structured content of the multiple first questions, the second structured content corresponding to the similar question group is determined. Finally, based on the second structured content, the first structured content of each first question is updated to obtain the target structured content of each first question. Thus, based on the structured content of multiple first questions in the similar question group, the structured content corresponding to the similar question group is determined. Using the structured content corresponding to the similar question group, the structured content of each first question in the similar question group is updated, thereby expanding the structured content of the similar questions and improving the quality of the questions.

[0050] Figure 2 This is a flowchart illustrating a test question structuring method provided in another embodiment of this application.

[0051] like Figure 2 As shown, the structured approach for this test question includes:

[0052] Step 201: Obtain similar question groups, wherein the similar question groups include multiple similar first questions.

[0053] Step 202: Extract the first structured content of each first question from multiple first questions.

[0054] In this application, steps 201-202 can be implemented in any of the embodiments of this application, and this application does not limit them.

[0055] Step 203: The union of attribute items in the first structured content of multiple first test questions is determined as the attribute items corresponding to the similar test question group.

[0056] In this application, the attribute items contained in the first structured content of multiple test questions belonging to the same similar test question group may be the same or different. Based on this, the union of the attribute items in the first structured content of multiple first test questions can be determined as the attribute items corresponding to the similar test question group.

[0057] For example, a similar question set includes questions A1, A2, and A3. The attribute items in the first structured content of question A1 include the question stem, answer, thought process tips, and subject. The attribute items in the first structured content of question A2 include the question stem, answer, thought process tips, and test point. The attribute items in the first structured content of question A3 include the question stem, answer, thought process tips, grade level, and grade. Therefore, it can be determined that the attribute items corresponding to this similar question set include the question stem, answer, thought process tips, subject, test point, grade level, and grade.

[0058] Step 204: Determine the second attribute value corresponding to the same attribute item based on the first attribute value corresponding to the same attribute item in the first structured content of multiple first test questions.

[0059] In this application, for attribute items corresponding to similar question groups, if only one question among multiple first questions has an attribute value corresponding to a certain attribute item, then the attribute value corresponding to that attribute item in the first structured content of that question can be used as the attribute value corresponding to that attribute item in the similar question group.

[0060] If two or more questions in the first structured content of a certain attribute item have a certain attribute item, the attribute value corresponding to that attribute item in the first structured content of any question can be selected as the attribute value corresponding to that attribute item in the similar question group. Alternatively, the union of the attribute values ​​corresponding to that attribute item in the first structured content of two or more questions can be used as the attribute value corresponding to that attribute item in the similar question group.

[0061] To further improve the quality of test questions, this application also allows for setting corresponding processing rules for each attribute item based on the question types of similar question groups. When determining the second attribute value corresponding to the same attribute item, the processing rules for the same attribute item can be used to process the first attribute values ​​corresponding to the same attribute item in the first structured content of multiple first test questions to determine the second attribute value corresponding to the same attribute item. Therefore, corresponding processing rules can be applied to different attribute items to process the first attribute value of the same attribute item and obtain the second attribute value, thus improving the accuracy of the second attribute value.

[0062] For example, for multiple-choice questions, the rule for processing the answer can be to take the answer of any question in the similar question group as the answer of the similar question group; the rule for processing the subject can be to take the subject of any question in the similar question group as the subject of the similar question group; and the rule for processing the answer explanation can be to take the answer explanation with the most characters in the similar question group as the answer explanation of the similar question group.

[0063] For example, for problem-solving questions, the rule for matching answers can be to take the answer with the most characters in the similar question group as the answer corresponding to the similar question group; the rule for matching subjects can be to take the subject of any question in the similar question group as the subject corresponding to the similar question group; and the rule for matching knowledge points can be to take the union of the knowledge points of each question in the similar question group as the knowledge point corresponding to the similar question group.

[0064] It should be noted that the processing rules corresponding to the above attribute items are only examples and can be set according to actual needs. This application does not limit them.

[0065] Step 205: Determine the second structured content based on the attribute items corresponding to similar question groups and the second attribute value corresponding to each attribute item.

[0066] In this application, the attribute items corresponding to similar question groups and the second attribute value corresponding to each attribute item can be used as the second structured content corresponding to similar question groups.

[0067] For example, if the attribute items corresponding to a similar question group include the question stem, answer, thought process tips, subject, test point, grade level, and grade, these attribute items and their corresponding second attribute values ​​can be used as the structured content corresponding to the similar question group.

[0068] Step 206: Update the first structured content of each first question based on the second structured content to obtain the target structured content of each first question.

[0069] In this application, step 206 can be implemented in any of the embodiments of this application, and this application does not limit it.

[0070] Optionally, in this application, if the first structured content of the first test question and the second structured content have the same attribute items, the first attribute value corresponding to the same attribute in the first structured content of the first test question can be updated to the union of the second attribute value and the first attribute value.

[0071] In this embodiment, when determining the second structured content corresponding to a similar question group based on the first structured content of multiple first questions, the union of attribute items in the first structured content of multiple first questions can be determined as the attribute item corresponding to the similar question group. Furthermore, based on the first attribute value corresponding to the same attribute item in the first structured content of multiple first questions, the second attribute value corresponding to the same attribute item can be determined. Finally, based on the attribute item corresponding to the similar question group and the second attribute value corresponding to each attribute item, the second structured content is determined. Therefore, by using the union of attribute items of each question in the similar question group as the attribute item corresponding to the similar question group, the attribute items of the similar questions are expanded. Thus, based on the structured content corresponding to the similar question group, the structured content of each question in the similar question group can be updated, thereby expanding the structured content of the similar questions.

[0072] Figure 3 This is a flowchart illustrating a test question structuring method provided in another embodiment of this application.

[0073] like Figure 3 As shown, the structured approach for this test question includes:

[0074] Step 301: Obtain similar question groups, wherein the similar question groups include multiple similar first questions.

[0075] Step 302: Extract the first structured content of each first question from multiple first questions.

[0076] Step 303: The union of attribute items in the first structured content of multiple first test questions is determined as the attribute items corresponding to the similar test question group.

[0077] Step 304: Determine the second attribute value corresponding to the same attribute item based on the first attribute value corresponding to the same attribute item in the first structured content of multiple first test questions.

[0078] Step 305: Determine the second structured content based on the attribute items corresponding to similar question groups and the second attribute value corresponding to each attribute item.

[0079] In this application, steps 301-305 can be implemented in any of the embodiments of this application, and this application does not limit them.

[0080] Step 306: If the first structured content and the second structured content of any first test question have the same attribute item, update the first attribute value corresponding to the same attribute in the first structured content of any first test question to the second attribute value corresponding to the same attribute in the second structured content.

[0081] In this application, if the first structured content of any first test question has the same attribute item as the second structured content, the first attribute value corresponding to the same attribute in the first structured content of the first test question can be updated to the second attribute value corresponding to the same attribute in the second structured content.

[0082] For example, if the second structured content of a similar question group contains the attribute item "subject", and the first structured content of a question in the same similar question group also contains the attribute item "subject", then the first attribute value corresponding to the subject in the first structured content of that question can be updated to the second attribute value corresponding to the subject in the second structured content.

[0083] Step 307: If any attribute item in the second structured content is missing in the first structured content of any first question, supplement any attribute item and the corresponding second attribute value in the first structured content of any first question.

[0084] In practical applications, the structured content of test questions varies. Therefore, in this application, if any attribute item in the second structured content is missing from the first structured content of any first test question in a similar test question group, then any attribute item and the corresponding second attribute value can be added to the first structured content of any first test question.

[0085] For example, a similar question group includes questions A1, A2, and A3. The corresponding attribute items for this similar question group are question stem, answer, thought process tips, subject, test point, grade level, and grade. Compared to the second structured content, the first structured content of question A1 lacks the attribute items test point, grade level, and grade. These three attribute items and their corresponding second attribute values ​​can be added to the first structured content of question A1 to obtain the target structured content of question A1. The first structured content of question A2 lacks the attribute items subject, grade level, and grade. These three attribute items and their corresponding second attribute values ​​can be added to the first structured content of question A2 to obtain the target structured content of question A2. The first structured content of question A3 lacks the attribute items subject and test point. These two attribute items and their corresponding second attribute values ​​can be added to the first structured content of question A3 to obtain the target structured content of question A3.

[0086] In this embodiment, if the first structured content of any first question in a similar question group has the same attribute item as the second structured content, the first attribute value corresponding to the same attribute in the first structured content of any first question can be updated to the second attribute value corresponding to the same attribute in the second structured content. If any attribute item in the second structured content is missing from the first structured content of any first question, the missing attribute item and its corresponding second attribute value can be added to the first structured content of any first question. Thus, compared to the second structured content, updating the attribute values ​​of the attribute items present in the first structured content of a question to the attribute values ​​of those attribute items in the second structured content supplements the missing attribute items and attribute values ​​in the first structured content, thereby expanding the structured content of the question.

[0087] Figure 4 This is a flowchart illustrating a test question structuring method provided in another embodiment of this application.

[0088] like Figure 4 As shown, the structured approach for this test question includes:

[0089] Step 401: Obtain the set of question identifiers corresponding to similar question groups from the question queue.

[0090] In this application, the question queue can store multiple sets of question identifiers corresponding to similar question groups, that is, the question queue can store multiple sets of question identifiers. Each set of question identifiers can be a set of question identifiers for the first question in multiple similar question groups.

[0091] To improve processing efficiency, this application can obtain a set of question identifiers for similar question groups from the question queue in parallel. For example, multiple consumer threads can be used to concurrently obtain a set of question identifiers corresponding to similar question groups from the question queue, so as to perform structured processing on multiple similar question groups at the same time.

[0092] Step 402: Obtain multiple first questions based on multiple question identifiers in the question identifier set.

[0093] In this application, the question identifier set includes multiple question identifiers. Based on the multiple question identifiers in the question identifier set, the question corresponding to each question identifier in the question identifier set, i.e., the first question, can be obtained.

[0094] Step 403: Extract the first structured content of each first question from multiple first questions.

[0095] Step 404: Based on the first structured content of multiple first test questions, determine the second structured content corresponding to similar test question groups.

[0096] Step 405: Update the first structured content of each first question based on the second structured content to obtain the target structured content of each first question.

[0097] In this application, steps 403-405 can be implemented in any of the embodiments of this application, and this application does not limit them.

[0098] In this embodiment, when obtaining similar question groups, a set of question identifiers corresponding to the similar question groups can be obtained from the question queue, and multiple first questions can be obtained based on multiple question identifiers in the question identifier set. Therefore, the question queue enables parallel acquisition of similar question groups, thereby achieving parallel processing of multiple similar question groups and improving processing efficiency.

[0099] Figure 5 This is a flowchart illustrating a test question structuring method provided in another embodiment of this application.

[0100] like Figure 5 As shown, the structured approach for this test question includes:

[0101] Step 501: Obtain the test question identification file and divide the test question identification file into multiple sub-files.

[0102] In this application, the test question identification file may contain multiple test question identifications. The test question identifications in the test question identification file may be the test question identifications of all test questions in a certain test question resource library, or they may be the test question identifications of all test questions in multiple similar test question groups, or they may be the test question identifications of any test question in each of the multiple similar test question groups, etc. This application does not limit this.

[0103] For example, a test question resource library contains 600 million test questions, of which there are 80 million similar test question groups, totaling 320 million test questions. The test question identification file can contain the test question identifications of the 600 million test questions, or it can contain the test question identifications of the 320 million test questions, or it can contain the 80 million test question identifications. The 80 million test question identifications can be obtained by extracting the test question identification of a single question from each of the 80 million similar test question groups.

[0104] Since the number of question identifiers in the question identifier file may be large, reading each question identifier one by one from the file would be slow. Therefore, to improve reading efficiency, this application proposes dividing the question identifier file into multiple sub-files. The number of question identifiers contained in each sub-file can be the same or different; this application does not impose any limitation on this.

[0105] Step 502: Based on concurrent threads, read the question identifiers from each sub-file.

[0106] In this application, multiple threads can be started, and the test question identifiers can be read from each sub-file using concurrent threads, thereby enabling parallel reading of test question identifiers from multiple sub-files and improving reading efficiency.

[0107] In this application, a corresponding log file can be generated during the processing of each sub-file. Each time a question identifier is read, it is written to the log file. Then, the number of question identifiers in the log file can be counted to determine the number of processed files. After each sub-file is processed, it is moved to the completed directory. This allows unprocessed sub-files to be read again after service interruption, thereby achieving overall progress statistics and breakpoint resumption.

[0108] Step 503: Obtain the target question identifier of the question that is similar to the question corresponding to the question identifier.

[0109] In this application, for each question identifier read, a target question identifier similar to the question corresponding to the question identifier can be obtained from a database storing similar question groups.

[0110] For example, questions B1, B2, B3, and B4 are similar questions. Among them, B1, B2, B3, and B4 are question identifiers. When B1 is read, the question identifiers B2, B3, and B4 of questions similar to B1 can be obtained.

[0111] Step 504: Push the test question identifiers and target test question identifiers belonging to the same similar test question group to the test question queue.

[0112] In this application, the test question identifiers and target test question identifiers belonging to the same similar test question group can be pushed to the test question queue, that is, the set of test question identifiers of each test question in the similar test question group can be pushed to the test question queue.

[0113] For example, questions B1, B2, B3, and B4 are similar questions. Among them, B1, B2, B3, and B4 are question identifiers. When B1 is read, the question identifiers B2, B3, and B4 of questions similar to B1 can be obtained, and B1, B2, B3, and B4 can be pushed into the question queue.

[0114] If the question identifier file also includes B2, B3, and B4, since question identifiers for similar questions have already been obtained based on B1 and B1, B2, B3, and B4 have been pushed into the question queue, B2, B3, and B4 do not need to be pushed into the question queue when they are read later, in order to avoid duplicate processing.

[0115] Step 505: Obtain the set of question identifiers corresponding to similar question groups from the question queue.

[0116] Step 506: Obtain multiple first questions based on multiple question identifiers in the question identifier set.

[0117] Step 507: Extract the first structured content of each first question from multiple first questions.

[0118] Step 508: Based on the first structured content of multiple first test questions, determine the second structured content corresponding to similar test question groups.

[0119] Step 509: Update the first structured content of each first question based on the second structured content to obtain the target structured content of each first question.

[0120] In this application, steps 505-509 can be implemented in any of the embodiments of this application, and this application does not limit them.

[0121] It should be noted that the above-mentioned process of reading sub-files and pushing question identifiers belonging to the same similar question group to the question queue based on concurrent threads, and obtaining the set of question identifiers corresponding to similar question groups from the question queue, can be done asynchronously. For example, if the question queue reaches its maximum length, it can be paused. After obtaining the set of question identifiers corresponding to similar question groups from the question queue, if the question queue is not full, question identifiers of the same similar question group can continue to be pushed to the question queue.

[0122] In this embodiment, by dividing the question identifier file into multiple sub-files, and using concurrent threads to read question identifiers from each sub-file, and pushing question identifiers belonging to the same similar question group to the question queue, parallel processing can be achieved based on the question queue, thus improving processing efficiency. Furthermore, dividing the question identifier file into multiple sub-files and reading question identifiers from the sub-files using concurrent threads enables parallel reading of question identifiers from multiple sub-files, further improving reading efficiency.

[0123] In practical applications, the initial construction of test question resource databases typically does not restrict the structured content of the questions. However, as the amount of test question resources increases, the structured content of the questions in the resource database becomes inconsistent. Related technologies mainly rely on teaching and research teachers to supplement the structured content of the questions, which is a cumbersome, labor-intensive, and inefficient process.

[0124] In one optional embodiment, before obtaining the question identifier file, similar questions can be searched using the question stem of each second question in the question resource library to obtain the search results for each second question. Based on the search results for each second question, similar questions for each second question are determined. If the similar questions for any second question are not empty, the question identifier of any second question and the question identifier of the corresponding similar question are associated and stored in the database. Then, based on the question identifiers in the database, a question identifier file is generated.

[0125] In practical applications, some test questions may not have similar test questions. Therefore, in this application, when there are no similar test questions for any second test question, the test question identifier of any second test question and the test question identifier of the corresponding similar test question are associated and stored in the database.

[0126] In this application, a search engine with similar question search functionality can be used to search for similar questions. In practical applications, if the stem of the second question is an image, the image can be OCR-recognized to obtain text, which can then be used to search for the second question in a search engine.

[0127] The search results may include, but are not limited to, the stems of candidate questions, the question identifiers of candidate questions, and the matching degree of the candidate questions. Here, the matching degree of the candidate questions refers to the degree of matching between the stems of the candidate questions and the stems used in the search.

[0128] In this application, when determining similar questions for each second question, if the question identifier of the first candidate question in the search results for the second question is inconsistent with the question identifier of the second question, and the matching degree corresponding to the first candidate question is greater than a preset threshold, then the first candidate question can be determined as a similar question to the second question. In other words, if the question identifier of the first candidate question in the search results is not the second question itself, and the matching degree corresponding to the first candidate question is greater than a preset threshold, then the first candidate question can be considered a similar question to the second question.

[0129] If the question identifier of the first candidate question is not the second question itself, but the matching degree corresponding to the first candidate question is less than the preset threshold, it can be considered that the second question has no similar questions.

[0130] If the question identifier of the first candidate question is the same as that of the second question, and the matching degree of the second candidate question is greater than a preset threshold, the second candidate question can be determined as a similar question to the second question. In other words, if the first candidate question in the search results is the second question itself, excluding the question itself, if the matching degree of the second candidate question is greater than a preset threshold, the second candidate question can be considered a similar question to the second question.

[0131] If the first candidate question in the search results is the second question itself and the matching degree of the second candidate question is less than or equal to the preset threshold, it can be considered that the second question has no similar questions.

[0132] Therefore, based on the identifiers of candidate questions in the search results and the matching degree of the candidate questions, similar questions to the second question can be determined, which is convenient and efficient.

[0133] Optionally, if the search results do not contain the matching degree corresponding to the candidate questions, then the matching degree between the second question and the candidate questions can be calculated, and based on the calculated matching degree, similar questions to the second question can be determined.

[0134] In this application, when generating a test item identifier file based on the test item identifiers in the database, it is possible to write all the test item identifiers in the database into the test item identifier file, or it is possible to write the test item identifier of any first test item in each similar test item group into the test item identifier file.

[0135] For example, a test question resource library contains 600 million test questions, of which there are 80 million similar test question groups, totaling 320 million test questions. The test question identifiers of the 320 million test questions can be written into a test question identifier file, or the test question identifier of any test question in each similar test question group can be written into the test question identifier file, so that the test question identifier file contains 80 million test question identifiers.

[0136] In this embodiment, similar questions are searched for using the stem of each second question in the question resource library to obtain search results for each second question. Based on the search results for each second question, similar questions are determined. If a second question has similar questions, the question identifier of the second question and the question identifier of the similar questions can be associated and stored in the database. Based on the question identifiers in the database, a question identifier file is generated. Therefore, similar question groups in the question resource library can be determined using the search results for each second question, and the question identifier file can be used to structure the similar question groups, improving processing efficiency, saving manpower costs, and improving the quality of questions in the question resource library.

[0137] To facilitate understanding of the above embodiments, taking a preset threshold of 90% as an example, the following is combined with... Figure 6 To explain, Figure 6 This is a schematic diagram of a test question structuring process provided in an embodiment of this application.

[0138] like Figure 6 As shown, the structuring process for this test question includes:

[0139] Step 601: Obtain the question stem.

[0140] In this application, the question stems can be obtained from the question resource database.

[0141] Step 602: Does the question stem contain an image? If it does, proceed to step 603; otherwise, proceed to step 604.

[0142] Step 603: Obtain the OCR-recognized text.

[0143] In this application, if the question stem contains an image, the image can be OCR-recognized to obtain OCR-recognized text.

[0144] Step 604: Search the middle platform to obtain search results.

[0145] In this application, the search platform is a search engine with the function of searching for similar test questions. The search platform contains test questions from the test question resource library, and can be used to search for test questions and obtain search results.

[0146] Step 605: Is the matching degree greater than 90%? If it is greater than 90%, proceed to step 606; otherwise, it is considered that there are no similar questions for this question.

[0147] In this application, it is possible to determine whether the highest matching degree among the search results that are not the test questions themselves is greater than 90%.

[0148] Step 606: Store similarity relationships.

[0149] In this application, if a test question has similar test questions, the test question identifier and the test question identifiers of the similar test questions can be stored in the database. Therefore, the stem of each test question in the test question resource database can be used for searching, and similar test questions can be determined based on the search results.

[0150] Step 607: Obtain the individual question identifier of the similar question group, write it into the question identifier file, and divide it into multiple sub-files.

[0151] In this application, the single question identifier of a similar question group can refer to the question identifier of any question in the similar question group.

[0152] Step 608: Start concurrent threads.

[0153] Step 609: Obtain the list of sub-files and read and process them in a loop.

[0154] In this application, a list of sub-files can be obtained and read in a loop using concurrent threads.

[0155] Step 610: After the sub-files are processed, move them to the completed directory.

[0156] In this application, after the question identifier reading and processing of each sub-file is completed, the sub-file can be moved to the completed directory.

[0157] In this application, a corresponding log file can be generated during the processing of each sub-file. Each time a question identifier is read, it is written to the log file. Then, the number of question identifiers in the log file can be counted to determine the number of processed files. After each sub-file is processed, it is moved to the completed directory. This allows unprocessed sub-files to be read again after service interruption, thereby achieving overall progress statistics and breakpoint resumption.

[0158] Step 611: Is the target file empty? If it is not empty, proceed to step 612; otherwise, end the process.

[0159] Step 612: Obtain similar question groups and push them into the question queue.

[0160] In this application, the read question identifier can be used to obtain the target question identifier of the question that is similar to the question corresponding to the question identifier. Thus, the question identifiers and target question identifiers belonging to the same similar question group can be pushed into the question queue, that is, the similar question group can be pushed into the question queue.

[0161] In this application, if the question queue reaches its maximum length, that is, the question queue is full, the reading of question identifiers in the sub-file can be paused, and the reading of question identifiers in the sub-file can continue when the question queue is not full.

[0162] Step 613: Obtain the set of question identifiers corresponding to similar question groups from the question queue.

[0163] Step 614: Obtain the questions from the similar question group.

[0164] Step 615: Extract structured content.

[0165] In this application, the structured content of each question in a similar question group can be extracted.

[0166] Step 616: Supplement the structured content.

[0167] In this application, the structured content corresponding to the similar question group can be determined based on the structured content of each question in the similar question group, and then the structured content of each question in the similar question group can be supplemented based on the structured content corresponding to the similar question group. For example, the process of updating the first structured content of the first question based on the second structured content in the above embodiment can be referred to.

[0168] Step 617: Update the test question resource library.

[0169] After supplementing the structured content of each question in a similar question group, the final structured content of the questions can be updated to the question resource library.

[0170] Therefore, structuring the test questions in the test question resource database using the above methods can improve the overall quality of the test question resources.

[0171] To achieve the above embodiments, this application also proposes a test question structuring device. Figure 7 This is a schematic diagram of the structure of a test question structuring device provided in an embodiment of this application.

[0172] like Figure 7 As shown, the test question structuring device 700 includes:

[0173] The first acquisition module 710 is used to acquire similar question groups, wherein the similar question groups include multiple similar first questions;

[0174] Extraction module 720 is used to extract the first structured content of each first question from multiple first questions;

[0175] The first determining module 730 is used to determine the second structured content corresponding to similar question groups based on the first structured content of multiple first question groups.

[0176] The update module 740 is used to update the first structured content of each first test question based on the second structured content, so as to obtain the target structured content of each first test question.

[0177] In one possible implementation of this application embodiment, the first determining module 730 is used for:

[0178] The union of attribute items in the first structured content of multiple first test questions is determined as the attribute items corresponding to similar test question groups;

[0179] Based on the first attribute value corresponding to the same attribute item in the first structured content of multiple first test questions, determine the second attribute value corresponding to the same attribute item;

[0180] The second structured content is determined based on the attribute items corresponding to similar question groups and the second attribute value corresponding to each attribute item.

[0181] In one possible implementation of this application embodiment, the first determining module 730 is used for:

[0182] Based on the processing rules corresponding to the same attribute item, the first attribute value corresponding to the same attribute item in the first structured content of multiple first test questions is processed to determine the second attribute value corresponding to the same attribute item.

[0183] In one possible implementation of this application embodiment, the first determining module 730 is used for:

[0184] If the first structured content and the second structured content of any first test question have the same attribute item, update the first attribute value corresponding to the same attribute in the first structured content of any first test question to the second attribute value corresponding to the same attribute in the second structured content.

[0185] If any attribute item in the second structured content is missing in the first structured content of any first question, supplement the first structured content of any first question with the missing attribute item and the corresponding second attribute value.

[0186] In one possible implementation of this application embodiment, the first acquisition module 710 is used for:

[0187] Retrieve the set of question identifiers corresponding to similar question groups from the question queue;

[0188] Based on multiple question identifiers in the question identifier set, obtain multiple first questions.

[0189] In one possible implementation of this application embodiment, the device may further include:

[0190] The segmentation module is used to obtain the test question identifier file and split the test question identifier file into multiple sub-files;

[0191] The reading module is used to read question identifiers from each sub-file based on concurrent threads;

[0192] The second acquisition module is used to acquire target question identifiers for questions similar to those corresponding to question identifiers.

[0193] The push module is used to push the test question identifiers and target test question identifiers belonging to the same similar test question group to the test question queue.

[0194] In one possible implementation of this application embodiment, the device may further include:

[0195] The third acquisition module is used to search for similar questions using the question stem of each second question in the question resource library, so as to obtain the search results for each second question;

[0196] The second determining module is used to determine similar questions for each second question based on the search results for each second question;

[0197] The storage module is used to associate and store the question identifier of any second question and the question identifier of the corresponding similar question in the database when the similar questions of any second question are not empty;

[0198] The generation module is used to generate test question identifier files based on the test question identifiers in the database.

[0199] In one possible implementation of this application embodiment, the search results include the question identifiers of candidate questions and the matching degree corresponding to the candidate questions. The second determining module is used for:

[0200] If the question identifier of the first candidate question is inconsistent with the question identifier of the second question, and the matching degree of the first candidate question is greater than the preset threshold, the first candidate question is determined to be a similar question to the second question.

[0201] If the question identifier of the first candidate question is the same as that of the second question, and the matching degree of the second candidate question is greater than a preset threshold, the second candidate question is determined to be a similar question to the second question.

[0202] It should be noted that the explanation of the aforementioned test question structuring method embodiment also applies to the test question structuring device of this embodiment, so it will not be repeated here.

[0203] In this embodiment, a similar question group is obtained, and the first structured content of each first question in the multiple first questions within the similar question group is extracted. Then, based on the first structured content of the multiple first questions, the second structured content corresponding to the similar question group is determined. Finally, based on the second structured content, the first structured content of each first question is updated to obtain the target structured content of each first question. Thus, based on the structured content of multiple first questions in the similar question group, the structured content corresponding to the similar question group is determined. Using the structured content corresponding to the similar question group, the structured content of each first question in the similar question group is updated, thereby expanding the structured content of the similar questions and improving the quality of the questions.

[0204] According to embodiments of this application, this application also provides an electronic device, a readable storage medium, and a computer program product.

[0205] Figure 8 A schematic block diagram of an example electronic device 800 that can be used to implement embodiments of this application is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the application described and / or claimed herein.

[0206] like Figure 8As shown, device 800 includes a computing unit 801, which can perform various appropriate actions and processes based on a computer program stored in ROM (Read-Only Memory) 802 or a computer program loaded from storage unit 808 into RAM (Random Access Memory) 803. RAM 803 can also store various programs and data required for the operation of device 800. The computing unit 801, ROM 802, and RAM 803 are interconnected via bus 804. I / O (Input / Output) interface 805 is also connected to bus 804.

[0207] Multiple components in device 800 are connected to I / O interface 805, including: input unit 806, such as keyboard, mouse, etc.; output unit 807, such as various types of monitors, speakers, etc.; storage unit 808, such as disk, optical disk, etc.; and communication unit 809, such as network card, modem, wireless transceiver, etc. Communication unit 809 allows device 800 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0208] The computing unit 801 can be various general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 801 include, but are not limited to, CPUs (Central Processing Units), GPUs (Graphics Processing Units), various special-purpose AI (Artificial Intelligence) computing chips, various computing units running machine learning model algorithms, DSPs (Digital Signal Processors), and any suitable processor, controller, microcontroller, etc. The computing unit 801 performs the various methods and processes described above, such as the test item structuring method. For example, in some embodiments, the test item structuring method can be implemented as a computer software program tangibly contained in a machine-readable medium, such as storage unit 808. In some embodiments, part or all of the computer program can be loaded and / or installed on device 800 via ROM 802 and / or communication unit 809. When the computer program is loaded into RAM 803 and executed by the computing unit 801, one or more steps of the test item structuring method described above can be performed. Alternatively, in other embodiments, the computing unit 801 may be configured to perform a question structuring method by any other suitable means (e.g., by means of firmware).

[0209] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, FPGAs (Field Programmable Gate Arrays), ASICs (Application-Specific Integrated Circuits), ASSPs (Application-Specific Standard Products), SOCs (System-on-Chips), CPLDs (Complex Programmable Logic Devices), computer hardware, firmware, software, and / or combinations thereof. These various implementations may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0210] The program code used to implement the methods of this application may be written in any combination of one or more programming languages. This program code may be provided to a processor or controller of a general-purpose computer, special-purpose computer, or other programmable data processing device, such that when executed by the processor or controller, the functions / operations specified in the flowcharts and / or block diagrams are implemented. The program code may be executed entirely on a machine, partially on a machine, as a standalone software package partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0211] In the context of this application, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, RAM, ROM, EPROM (Electrically Programmable Read-Only Memory) or flash memory, optical fiber, CD-ROM (Compact Disc Read-Only Memory), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0212] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device for displaying information to the user (e.g., a CRT (Cathode-Ray Tube) or LCD (Liquid Crystal Display) monitor); and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the computer. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0213] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or middleware components (e.g., application servers), or frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication (e.g., communication networks) of any form or medium. Examples of communication networks include LANs (Local Area Networks), WANs (Wide Area Networks), the Internet, and blockchain networks.

[0214] Computer systems can include clients and servers. Clients and servers are generally geographically separated and typically interact via communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. A server can be a cloud server, also known as a cloud computing server or cloud host, a hosting product within the cloud computing service system that addresses the shortcomings of traditional physical hosts and VPS (Virtual Private Server) services, such as high management difficulty and weak business scalability. Servers can also be servers for distributed systems or servers integrated with blockchain technology.

[0215] According to embodiments of this application, this application also provides a computer program product that, when an instruction processor in the computer program product is executed, performs the test question structuring method proposed in the above embodiments of this application.

[0216] It should be understood that the various forms of processes shown above can be used to rearrange, add, or delete steps. For example, the steps described in this application can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution disclosed in this application can be achieved, and this is not limited herein.

[0217] The specific embodiments described above do not constitute a limitation on the scope of protection of this application. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A method for structuring test questions, comprising: Obtain similar question groups, wherein the similar question groups include multiple similar first questions; Extract the first structured content of each of the multiple first test questions; The union of the attribute items in the first structured content of the multiple first test questions is determined as the attribute item corresponding to the similar test question group; Based on the first attribute value corresponding to the same attribute item in the first structured content of the multiple first test questions, determine the second attribute value corresponding to the same attribute item; Based on the attribute items corresponding to the similar question groups and the second attribute value corresponding to each attribute item, determine the second structured content corresponding to the similar question groups; Based on the second structured content, the first structured content of each first question is updated to obtain the target structured content of each first question.

2. The method as described in claim 1, wherein, The step of determining the second attribute value corresponding to the same attribute item based on the first attribute value corresponding to the same attribute item in the first structured content of the plurality of first test questions includes: According to the processing rules corresponding to the same attribute item, the first attribute value corresponding to the same attribute item in the first structured content of the multiple first test questions is processed to determine the second attribute value corresponding to the same attribute item.

3. The method as described in claim 1, wherein, The step of updating the first structured content of each first question based on the second structured content to obtain the target structured content of each first question includes: If the first structured content of any first test question has the same attribute item as the second structured content, the first attribute value corresponding to the same attribute in the first structured content of any first test question is updated to the second attribute value corresponding to the same attribute in the second structured content. If any attribute item in the second structured content is missing in the first structured content of any first question, then the attribute item and the corresponding second attribute value shall be added to the first structured content of any first question.

4. The method according to any one of claims 1-3, wherein, The acquisition of similar question sets includes: Obtain the set of question identifiers corresponding to the similar question groups from the question queue; Based on multiple question identifiers in the question identifier set, obtain the multiple first questions.

5. The method of claim 4, further comprising: Obtain the test question identification file and divide the test question identification file into multiple sub-files; Based on concurrent threads, the question identifiers are read from each sub-file; Obtain target question identifiers for questions similar to the question identifiers corresponding to the question identifiers; The test question identifiers belonging to the same similar test question group and the target test question identifier are pushed to the test question queue.

6. The method of claim 5, wherein, Before obtaining the test question identification file, the following is also included: Search for similar questions using the question stem of each second question in the question resource database to obtain the search results for each second question; Based on the search results for each second question, identify similar questions for each second question; If any similar question to a second question is not empty, the question identifier of the second question and the question identifier of the similar question corresponding to the second question are associated and stored in the database. Based on the question identifiers in the database, the question identifier file is generated.

7. The method of claim 6, wherein, The search results include the question identifiers of the candidate questions and the matching degree corresponding to the candidate questions. The step of determining similar questions for each second question based on the search results for each second question includes: If the question identifier of the first candidate question is inconsistent with the question identifier of the second question and the matching degree corresponding to the first candidate question is greater than a preset threshold, the first candidate question is determined to be a similar question to the second question. If the question identifier of the first candidate question is the same as that of the second question and the matching degree of the second candidate question is greater than the preset threshold, then the second candidate question is determined to be a similar question to the second question.

8. A test item structuring device, comprising: The first acquisition module is used to acquire similar question groups, wherein the similar question groups include multiple similar first questions; The extraction module is used to extract the first structured content of each of the multiple first test questions; The first determining module is used to determine the attribute item corresponding to the similar question group by taking the union of the attribute items in the first structured content of the plurality of first test questions; to determine the second attribute value corresponding to the same attribute item according to the first attribute value corresponding to the same attribute item in the first structured content of the plurality of first test questions; and to determine the second structured content corresponding to the similar question group according to the attribute item corresponding to the similar question group and the second attribute value corresponding to each attribute item. The update module is used to update the first structured content of each first question based on the second structured content, so as to obtain the target structured content of each first question.

9. The apparatus of claim 8, wherein, The first determining module is used for: According to the processing rules corresponding to the same attribute item, the first attribute value corresponding to the same attribute item in the first structured content of the multiple first test questions is processed to determine the second attribute value corresponding to the same attribute item.

10. The apparatus of claim 8, wherein, The first determining module is used for: If the first structured content of any first test question has the same attribute item as the second structured content, the first attribute value corresponding to the same attribute in the first structured content of any first test question is updated to the second attribute value corresponding to the same attribute in the second structured content. If any attribute item in the second structured content is missing in the first structured content of any first question, then the attribute item and the corresponding second attribute value shall be added to the first structured content of any first question.

11. The apparatus according to any one of claims 8-10, wherein, The first acquisition module is used for: Obtain the set of question identifiers corresponding to the similar question groups from the question queue; Based on multiple question identifiers in the question identifier set, obtain the multiple first questions.

12. The apparatus of claim 11, further comprising: The segmentation module is used to obtain the test question identification file and segment the test question identification file into multiple sub-files; The reading module is used to read question identifiers from each sub-file based on concurrent threads; The second acquisition module is used to acquire target question identifiers of questions similar to the question identifiers corresponding to the question identifiers; The push module is used to push the question identifier and the target question identifier belonging to the same similar question group to the question queue.

13. The apparatus of claim 12, further comprising: The third acquisition module is used to search for similar questions using the question stem of each second question in the question resource library, so as to obtain the search results for each second question; The second determining module is used to determine similar questions for each second question based on the search results for each second question; The storage module is used to associate and store the question identifier of any second question and the question identifier of the corresponding similar question in the database when the similar questions of any second question are not empty; The generation module is used to generate the test question identifier file based on the test question identifiers in the database.

14. The apparatus of claim 13, wherein, The search results include the question identifiers of the candidate questions and the matching degree corresponding to the candidate questions. The second determining module is used for: If the question identifier of the first candidate question is inconsistent with the question identifier of the second question and the matching degree corresponding to the first candidate question is greater than a preset threshold, the first candidate question is determined to be a similar question to the second question. If the question identifier of the first candidate question is the same as that of the second question and the matching degree of the second candidate question is greater than the preset threshold, then the second candidate question is determined to be a similar question to the second question.

15. An electronic device comprising: At least one processor; as well as A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-7.

16. A non-transitory computer-readable storage medium storing computer instructions, wherein, The computer instructions are used to cause the computer to perform the method according to any one of claims 1-7.

17. A computer program product comprising a computer program that, when executed by a processor, implements the steps of the method according to any one of claims 1-7.

Images