A test question matching method and device

Abstract

The present disclosure provides a test question matching method and device, relates to the field of data processing, and particularly relates to the technical field of text data processing. The specific implementation scheme is as follows: content conversion processing is performed on test question content of each test question to obtain content parameters representing the test question content of each test question; the content parameters corresponding to each test question are divided into a plurality of sub-parameters; test questions corresponding to the same sub-parameters are matched to obtain similar test questions. The test question matching scheme provided in the present disclosure can realize test question matching.

Description

A test item matching method and apparatus Technical Field

[0001] This disclosure relates to the field of data processing technology, particularly to the field of text data processing technology, and especially to a test question matching method and apparatus. Background Technology

[0002] Educational support platforms can acquire a vast number of test questions and provide them to users so that they can test and practice the knowledge points covered in the questions.

[0003] The educational support platform aims to provide users with other similar questions while they are answering the questions provided on the platform, in order to help them consolidate the knowledge points involved in the questions, or to provide users with the solution process of other questions when they encounter difficulties in solving the questions.

[0004] Therefore, a test question matching scheme is needed so that educational support platforms can match the test questions they acquire to obtain similar test questions. Summary of the Invention

[0005] This disclosure provides a test item matching method and apparatus.

[0006] According to one aspect of this disclosure, a test item matching method is provided, comprising:

[0007] The content of each test question is transformed to obtain content parameters that represent the content of each test question;

[0008] The content parameters corresponding to each test question are divided into multiple sub-parameters;

[0009] Match questions with the same sub-parameters to obtain similar questions.

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

[0011] The content conversion module is used to convert the content of each test question to obtain content parameters that represent the content of each test question.

[0012] The parameter segmentation module is used to divide the content parameters corresponding to each question into multiple sub-parameters;

[0013] The question matching module is used to match questions with the same sub-parameters to obtain similar questions.

[0014] According to another aspect of this disclosure, an electronic device is provided, comprising:

[0015] At least one processor; and

[0016] A memory communicatively connected to the at least one processor; wherein,

[0017] The memory stores instructions that can be executed by the at least one processor, which enables the at least one processor to perform the above-described test question matching method.

[0018] According to another aspect of this disclosure, a non-transitory computer-readable storage medium is provided storing computer instructions, wherein the computer instructions are used to cause the computer to perform the above-described test question matching method.

[0019] According to another aspect of this disclosure, a computer program product is provided, including a computer program that, when executed by a processor, implements the above-described test question matching method.

[0020] As can be seen from the above, when using the solution provided in this embodiment of the present disclosure for question matching, the content parameter represents the content of the question. After dividing the content parameter corresponding to the question into multiple sub-parameters, each sub-parameter also has a representational nature for the question. In this way, questions corresponding to the same sub-parameter can be understood as relatively similar questions. Thus, by matching questions corresponding to the same sub-parameter, similar questions can be accurately matched among all questions.

[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 disclosure, nor is it intended to limit the scope of this disclosure. Other features of this disclosure will become readily apparent from the following description. Attached Figure Description

[0022] The accompanying drawings are provided to better understand this solution and do not constitute a limitation of this disclosure. Wherein:

[0023] Figure 1 is a flowchart illustrating the first test question matching method provided in this embodiment of the present disclosure;

[0024] Figure 2 is a flowchart illustrating the second test question matching method provided in this embodiment of the present disclosure;

[0025] Figure 3 is a flowchart illustrating the third test question matching method provided in this embodiment of the present disclosure;

[0026] Figure 4a is a flowchart illustrating the fourth test question matching method provided in this embodiment of the present disclosure;

[0027] Figure 4b is a schematic flowchart of a method for determining sub-parameters according to an embodiment of this disclosure;

[0028] Figure 5 is a flowchart illustrating the fifth test question matching method provided in this embodiment of the present disclosure;

[0029] Figure 6 is a flowchart illustrating the sixth test question matching method provided in this embodiment of the present disclosure;

[0030] Figure 7 is a flowchart illustrating the seventh question matching method provided in this embodiment of the present disclosure;

[0031] Figure 8 is a flowchart illustrating the eighth question matching method provided in this embodiment of the present disclosure;

[0032] Figure 9 is a flowchart illustrating the first test question matching method provided in this embodiment of the present disclosure;

[0033] Figure 10 is a flowchart illustrating the second test question matching method provided in this embodiment of the present disclosure;

[0034] Figure 11 is a flowchart illustrating the third test question matching method provided in this embodiment of the present disclosure;

[0035] Figure 12 is a block diagram of an electronic device used to implement the test question matching method of the present disclosure. Detailed Implementation

[0036] The exemplary embodiments of this disclosure are described below with reference to the accompanying drawings, including various details of the embodiments to aid understanding, and 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 disclosure. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description.

[0037] In existing technologies, when matching test questions, recall and similarity calculation techniques are usually used. That is, multiple test questions are recalled through recall techniques, the similarity between each pair of recalled test questions is calculated, and similar test questions are determined based on the similarity.

[0038] For example, the aforementioned recall techniques can be inverted indexing, Elastic Search (ES) recall, vector recall, etc.; the similarity between test items can be edit distance, cosine similarity, or other similarities between test items, and the similarity between test items can be calculated using a similarity calculation model.

[0039] However, existing technologies require calculating the similarity between pairs of test questions and determining similar test questions based on the calculated similarity. This results in a large amount of computation for matching when the number of test questions is large, leading to low matching efficiency.

[0040] To address the aforementioned problems, this disclosure provides a test question matching method and apparatus, which will be described in detail below through specific embodiments.

[0041] Referring to Figure 1, which is a flowchart of the first test question matching method provided in this embodiment, the method includes the following steps S101-S103.

[0042] Step S101: Perform content transformation processing on the content of each test question to obtain content parameters that represent the content of each test question.

[0043] The aforementioned test questions may be practice questions for users to practice, test questions for users to take tests, or questions for users to refer to, etc.

[0044] Specifically, after obtaining multiple test questions, the content of the test questions can be extracted, and then the content of each test question can be transformed to obtain the above-mentioned content parameters.

[0045] In one embodiment of this disclosure, when extracting the test question content, the test question content may include at least one of the following information:

[0046] The question stem, the answer choices, the images in the question, and the characters in the formulas in the question.

[0047] When an image is present in the test questions, the content of the image can be identified and used as the test question content.

[0048] When a formula exists in a test question, the characters within the formula can be identified and used as test question content.

[0049] The above information can accurately reflect the content of the test questions. Therefore, using at least one of the above information as test question content for content conversion processing can improve the accuracy of the obtained content parameters. Furthermore, by matching test questions based on more accurate content parameters, the accuracy of test question matching can be improved.

[0050] In one embodiment of this disclosure, each question may be preprocessed before question matching.

[0051] For example, it can remove redundant questions with repetitive content from various test questions, clean up test questions, and standardize test questions.

[0052] Since test questions involving different knowledge points may have some information that is the same and some that is different, for example, the question stem may be the same but the options may be different, multiple pieces of information from the above four types can be used as the test question content when extracting the test question content. This is to prevent the test questions from being accidentally deleted in the preprocessing stage, thereby increasing the test question matching rate.

[0053] In one embodiment of this disclosure, the test question content can be converted using either of the following two methods.

[0054] In the first implementation, the test content can be converted into a text vector in a preset space, and the vector can be multiplied by a preset vector in the preset space. The value of the content parameter corresponding to the preset vector is determined according to the sign of the calculation result.

[0055] For example, the text vector of the test question content can be represented as: There are two preset vectors in the preset space, namely vector Corresponding to the first parameter in the content parameters, the vector This corresponds to the second position in the content parameter. Thus, based on the vector... with vector The result of performing a vector dot product can determine the value of the first element in the content parameter, based on the vector. with vector The result of performing a vector dot product determines the value of the second element in the content parameter. For example, if the vector... with vector If the result of a vector dot product is negative, then the first value of the content parameter is 0. with vector If the result of the vector dot product is positive, then the first value of the content parameter can be determined to be 1, thus determining the content parameter to be 01.

[0056] In the second implementation, the content parameters of each test question can be calculated through step S201 in the embodiment shown in Figure 2, which will not be described in detail here.

[0057] Step S102: Divide the content parameters corresponding to each question into multiple sub-parameters.

[0058] Specifically, the content parameters corresponding to the test questions can be represented by multi-bit values. This allows the content parameters to be divided into multiple sub-parameters based on the position of each value within the content parameters.

[0059] For example, if the content parameter is represented by a 64-bit binary number, it can be divided into four 16-bit binary numbers, with each 16-bit binary number serving as a sub-parameter. Alternatively, the content parameter can be divided into eight 8-bit binary numbers, with each 8-bit binary number serving as a sub-parameter.

[0060] In one embodiment of the present invention, after segmenting the content parameters, the segmented parameters can be directly used as sub-parameters, or the segmented parameters can be combined and the combined parameters can be used as sub-parameters.

[0061] In the example above, after the content parameter is divided into eight 8-bit binary numbers, the eight binary numbers can be combined to obtain four 16-bit binary numbers, which can be used as four sub-parameters.

[0062] In one embodiment of this disclosure, sub-parameters can be obtained by combining the segmented parameters using either of the following two implementation methods.

[0063] In the first implementation, the segmented parameters can be combined in step S302 of the embodiment shown in Figure 3 to obtain sub-parameters, which will not be described in detail here.

[0064] In the second implementation, the segmented parameters can be combined arbitrarily to obtain the combined parameters, which can then be used as sub-parameters.

[0065] Step S103: Match the questions with the same sub-parameters to obtain similar questions.

[0066] Specifically, based on the sub-parameters corresponding to each question, questions with the same sub-parameters can be identified among the questions, and then the identified questions can be matched to obtain similar questions.

[0067] In one embodiment of this disclosure, the test questions corresponding to the same sub-parameters can be determined by either of the following two implementation methods.

[0068] In the first implementation, questions can be divided according to the sub-parameters corresponding to each question to obtain a question set, and questions with the same sub-parameters can be identified in the question set.

[0069] When dividing test questions, test questions corresponding to one type of sub-parameter can be divided into a test question set, in which case all test questions in the test question set are test questions corresponding to the same sub-parameter. Alternatively, test questions corresponding to multiple types of sub-parameters can be divided into a test question set. If the above test question set contains test questions corresponding to multiple types of sub-parameters, then test questions corresponding to the same sub-parameter can be further identified in the test question set.

[0070] In the second implementation, the questions corresponding to the same sub-parameters can be determined through step S403 in the embodiment shown in Figure 4a, which will not be described in detail here.

[0071] In one embodiment of this disclosure, after determining the test questions corresponding to the same sub-parameters, the determined test questions can be matched using either of the following two implementation methods.

[0072] In the first implementation method, existing question matching technology can be used to match the determined questions.

[0073] In the second implementation method, test question matching can be performed through step S503 in the embodiment shown in Figure 5, which will not be described in detail here.

[0074] As can be seen from the above, when using the solution provided in this embodiment of the present disclosure for question matching, the content parameter represents the content of the question. After dividing the content parameter corresponding to the question into multiple sub-parameters, each sub-parameter also has a representational nature for the question. In this way, questions corresponding to the same sub-parameter can be understood as relatively similar questions. Thus, by matching questions corresponding to the same sub-parameter, similar questions can be accurately matched among all questions.

[0075] Furthermore, by utilizing the sub-parameters corresponding to each question, questions with the same sub-parameter can be filtered out. This allows for matching only questions with the same sub-parameter, rather than matching all questions, thus reducing the amount of data involved in the matching process. Moreover, for a single sub-parameter, one computing node can be used to match questions corresponding to that sub-parameter; for multiple sub-parameters, multiple computing nodes can be used in parallel to match questions corresponding to multiple sub-parameters. This shortens the matching time and improves the efficiency of question matching.

[0076] When calculating the content parameters corresponding to each test question, in addition to the method mentioned in step S101 above, the content parameters corresponding to each test question can also be obtained through step S201 in the embodiment shown in Figure 2 below.

[0077] In one embodiment of this disclosure, referring to Figure 2, a flowchart of a second test item matching method is provided. In this embodiment, the above method includes the following steps S201-S203.

[0078] Step S201: For each question, perform local sensitive hash (simhash) calculation on the question content to obtain the simhash value of the question, which is used as the content parameter representing the question content.

[0079] Specifically, for each question, when performing Local Sensitive Hashing (simhash) calculation on the question content, the question content can be segmented into words. Using existing hash algorithms, each word is converted into a hash value, and weights are assigned to each word's hash value based on its importance within the question content. This weighted sum of hash values ​​is then applied to the corresponding positions in the weighted summed hash values. The sign of the summed values ​​determines the corresponding value in the question's simhash value. For example, if the summed value is negative, the corresponding value in the question's simhash value is 0; if the summed value is positive, the corresponding value is 1.

[0080] For example, if the content of a test question is "the side length is one centimeter", the content of the test question can be segmented into multiple words, namely "side length", "is", and "one centimeter". Using a hash algorithm, the hash value of each word can be obtained, and weights can be assigned to the hash values ​​of each word. The hash values ​​of each word are then weighted, and the values ​​at the same position in the weighted hash values ​​of each word are superimposed. The value at the same position in the simhash value of the test question is determined based on the sign of the value at each position in the superposition result.

[0081] For example, the hash value of "side length" can be 100101, with a weight of 4, so the weighted hash value of "side length" is [4, -4, -4, 4, -4, 4]; the hash value of "for" can be 011010, with a weight of 1, so the weighted hash value of "for" is [-1, 1, 1, -1, 1, -1]; the hash value of "one centimeter" can be 011000, with a weight of 5, so the weighted hash value of "one centimeter" is [-5, 5, 5, -5, -5, -5]. Adding the weighted hash values ​​of each word segment yields the result [-2, 2, 2, -2, -8, -2]. Based on this result, the simhash value of the question can be determined to be 011000.

[0082] In addition, when segmenting the test questions, you can do so by sliding a window.

[0083] In the example above, segmenting the test question content into words yields multiple word segments, namely "side length", "length as", "as one", "one centimeter", and "centimeter".

[0084] Step S202: Divide the content parameters corresponding to each question into multiple sub-parameters.

[0085] Step S203: Match the questions with the same sub-parameters to obtain similar questions.

[0086] The steps S202 and S203 described above are the same as those S102 and S203 described above, and will not be repeated here.

[0087] As can be seen from the above, when applying the scheme provided in this embodiment of the present disclosure for question matching, the simhash value obtained after performing local sensitive hash (simhash) calculation can accurately reflect the question content. Therefore, by using the simhash value of the question as a content parameter, the content parameter of the question content can be accurately obtained. Thus, by performing question matching based on the more accurate content parameter corresponding to the question, the accuracy of question matching can be improved.

[0088] When segmenting the content parameters corresponding to each test question, in addition to the method mentioned in step S102 above, the content parameters corresponding to each test question can also be segmented by step S302 in the embodiment shown in Figure 3 below.

[0089] In one embodiment of this disclosure, referring to Figure 3, a flowchart of a third test item matching method is provided. In this embodiment, the above method includes the following steps S301-S303.

[0090] Step S301: Perform content transformation processing on the content of each test question to obtain content parameters that represent the content of each test question.

[0091] This step is the same as step S101 above, and will not be repeated here.

[0092] Step S302: For each question, divide the content parameters corresponding to the question into multiple intermediate parameters with a preset number of bits, and determine multiple sub-parameters corresponding to the question based on each intermediate parameter and the order of the intermediate parameters.

[0093] The preset number of bits can be set manually or determined based on the number of bits in the content parameters and the number of sub-parameters to be obtained.

[0094] Specifically, for each question, the content parameters corresponding to that question can be divided at preset intervals to obtain multiple intermediate parameters of preset length. After obtaining multiple intermediate parameters, they can be combined according to their order to obtain multiple sub-parameters corresponding to that question.

[0095] In one embodiment of this disclosure, after obtaining multiple intermediate parameters, multiple sub-parameters corresponding to the test question can be determined by any of the following two implementation methods.

[0096] In the first implementation, the intermediate parameters can be combined according to pre-set combination rules to obtain multiple sub-parameters.

[0097] For example, multiple sub-parameters corresponding to the test question can be obtained through step S402 in the embodiment shown in Figure 4a. The intermediate parameters can also be combined according to other combination rules. This disclosure does not limit this.

[0098] In the second implementation, any number of intermediate parameters can be combined to obtain multiple sub-parameters.

[0099] Step S303: Match the questions with the same sub-parameters to obtain similar questions.

[0100] This step is the same as step S103 above, and will not be repeated here.

[0101] As can be seen from the above, when applying the solution provided in this embodiment of the present disclosure for question matching, for each question, the content parameters corresponding to the question are divided into multiple intermediate parameters, and based on each intermediate parameter and the order of the intermediate parameters, multiple sub-parameters corresponding to the question can be accurately determined. Thus, by using the more accurate sub-parameters corresponding to each question for question matching, the accuracy of question matching can be improved.

[0102] When determining the sub-parameters based on the intermediate parameters and their order, in addition to the method mentioned in step S302 above, the sub-parameters corresponding to the test questions can also be determined through step S402 in the embodiment shown in Figure 4a below.

[0103] In one embodiment of this disclosure, referring to Figure 4a, a flowchart of the fourth test item matching method is provided. In this embodiment, the above method includes the following steps S401-S403.

[0104] Step S401: Perform content transformation processing on the content of each test question to obtain content parameters that represent the content of each test question.

[0105] This step is the same as step S101 above, and will not be repeated here.

[0106] Step S402: For each question, the content parameters corresponding to the question are divided into multiple intermediate parameters with a preset number of bits. According to the order of the intermediate parameters, the sub-parameters to which each intermediate parameter belongs are determined in turn, and the intermediate parameters corresponding to the same sub-parameter are connected to obtain multiple sub-parameters corresponding to the question.

[0107] The aforementioned preset number of bits can be set manually.

[0108] Specifically, you can set the expected number of sub-parameters to be obtained. After obtaining multiple intermediate parameters, you can divide each expected number of intermediate parameters into the expected number of sub-parameters according to the order of the intermediate parameters. This will determine the sub-parameters corresponding to each intermediate parameter. By connecting the intermediate parameters corresponding to the same sub-parameter, you can obtain the expected number of sub-parameters corresponding to the question.

[0109] For example, as shown in Figure 4b, which is a flowchart illustrating the determination of sub-parameters, if 8 intermediate parameters are obtained and the expected number of sub-parameters is 4, then the intermediate parameters corresponding to sequence numbers 1 and 5 belong to the first sub-parameter, those corresponding to sequence numbers 2 and 6 belong to the second sub-parameter, those corresponding to sequence numbers 3 and 7 belong to the third sub-parameter, and those corresponding to sequence numbers 4 and 8 belong to the fourth sub-parameter. After determining the sub-parameters to which each intermediate parameter belongs, the determined intermediate parameters can be concatenated to obtain the 4 sub-parameters of the above question.

[0110] Step S403: Match the questions with the same sub-parameters to obtain similar questions.

[0111] This step is the same as step S103 above, and will not be repeated here.

[0112] As can be seen from the above, when applying the solution provided in this embodiment of the present disclosure for question matching, after dividing the content parameters corresponding to each question into multiple intermediate parameters, the sub-parameters to which each intermediate parameter belongs are determined in turn according to the arrangement order of each intermediate parameter, and the intermediate parameters corresponding to the same sub-parameter are connected. This can accurately obtain multiple sub-parameters corresponding to each question, thereby improving the accuracy of question matching by using the more accurate sub-parameters corresponding to each question.

[0113] When segmenting the content parameters corresponding to a test question, for each test question, the content parameters corresponding to that test question can be segmented multiple times, thereby increasing the number of sub-parameters corresponding to that test question.

[0114] In one embodiment of this disclosure, referring to Figure 5, a flowchart of the fifth test item matching method is provided. In this embodiment, the above method includes the following steps S501-S503.

[0115] Step S501: Perform content transformation processing on the content of each test question to obtain content parameters that represent the content of each test question.

[0116] This step is the same as step S101 above, and will not be repeated here.

[0117] Step S502: When there are multiple preset number of digits, for each preset number of digits, the content parameter corresponding to the question is divided into multiple intermediate parameters of the preset number of digits, and multiple sub-parameters corresponding to the question are determined according to each intermediate parameter and the order of the intermediate parameters.

[0118] Specifically, when there are multiple preset number of digits, for each preset number of digits, the content parameters corresponding to the question can be divided into multiple intermediate parameters, thereby determining multiple sub-parameters corresponding to the question. It can be seen that for each question, using one preset number of digits, multiple sub-parameters corresponding to the question can be obtained. Using multiple preset number of digits, the number of sub-parameters obtained is the sum of the number of sub-parameters corresponding to each preset number of digits.

[0119] For details on how to segment the content parameters and how to determine the sub-parameters based on the intermediate parameters and their order, please refer to step S302 above. These details will not be repeated here.

[0120] Step S503: Match the questions with the same sub-parameters to obtain similar questions.

[0121] This step is the same as step S103 above, and will not be repeated here.

[0122] As can be seen from the above, when applying the solution provided in this embodiment of the present disclosure for question matching, in the presence of multiple preset number of digits, multiple sub-parameters can be obtained using each preset number of digits. In this way, by using multiple preset number of digits, the number of sub-parameters corresponding to each question can be increased. Thus, based on the sub-parameters corresponding to each question, the number of questions corresponding to the same sub-parameters can be increased, thereby increasing the matching quantity for question matching, preventing similar questions from being missed, and improving the accuracy of question matching.

[0123] In the process of matching test questions with the same sub-parameter, in addition to using the method mentioned in step S103 above to determine the test questions with the same sub-parameter, the test questions with the same sub-parameter can also be determined and matched through steps S603-S604 in the embodiment shown in Figure 6 below.

[0124] In one embodiment of this disclosure, referring to Figure 6, a flowchart of the sixth test item matching method is provided. In this embodiment, the above method includes the following steps S601-S604.

[0125] Step S601: Perform content transformation processing on the content of each test question to obtain content parameters that represent the content of each test question.

[0126] Step S602: Divide the content parameters corresponding to each question into multiple sub-parameters.

[0127] The steps S601 and S602 described above are the same as steps S101 and S102 described above, and will not be repeated here.

[0128] Step S603: Based on the values ​​of the sub-parameters, group the sub-parameters corresponding to each question to obtain the parameter groups corresponding to each preset calculation node.

[0129] Among them, sub-parameters with the same value belong to the same parameter group.

[0130] Each parameter group can correspond to one or more sub-parameter values.

[0131] Specifically, when grouping the sub-parameters corresponding to each question, sub-parameters with the same value can be identified from the sub-parameters corresponding to each question, and sub-parameters with one or more values ​​can be grouped into the same parameter group. After obtaining multiple parameter groups, preset calculation nodes can be assigned to each parameter group.

[0132] For example, if question x corresponds to sub-parameter c1, question y corresponds to sub-parameter c2, question z corresponds to sub-parameter d1, and question v corresponds to sub-parameter e1, where the values ​​of sub-parameters c1 and c2 are the same.

[0133] When grouping the sub-parameters corresponding to these four questions, sub-parameters c1 and c2 can be assigned to one parameter group, sub-parameter d1 to another parameter group, and sub-parameter e1 to a third parameter group, thus obtaining three parameter groups.

[0134] Alternatively, sub-parameters c1, c2, and d1 can be grouped into one parameter group, and sub-parameter e1 can be grouped into another parameter group, thus obtaining two parameter groups.

[0135] When assigning preset computing nodes to parameter groups, in one case, one computing node can be assigned to each parameter group. In this case, when dividing the parameter groups, the number of preset computing nodes can be obtained, and the sub-parameters corresponding to each question can be divided into the above number of parameter groups. In another case, one preset computing node can be assigned to multiple parameter groups. In this case, when dividing the parameter groups, the sub-parameters corresponding to each question can be divided into any number of parameter groups.

[0136] Step S604: For each parameter group, obtain the similar questions obtained by the computing node corresponding to the parameter group through matching the question to be matched.

[0137] Among them, the questions to be matched are the questions corresponding to the sub-parameters with the same value in this parameter group.

[0138] Specifically, for each parameter group, if there is only one type of sub-parameter with the same value, all sub-parameters in the parameter group will have the same value. In this case, the computing node can match the questions corresponding to each sub-parameter in the parameter group to obtain similar questions. If there are multiple types of sub-parameters with the same value in the parameter group, the computing node can determine the sub-parameters with the same value among the sub-parameters in the parameter group, and then match the questions corresponding to the determined sub-parameters to obtain similar questions.

[0139] As can be seen from the above, when using the solution provided in this embodiment for question matching, the sub-parameters corresponding to each question are grouped according to their values, ensuring that sub-parameters with the same values ​​belong to the same parameter group. Thus, during question matching, it is only necessary to determine the common sub-parameters within each parameter group and find the corresponding question based on the determined sub-parameters, thereby performing the matching. No operation on the question itself is required when determining the question corresponding to the same sub-parameter. Since the size of a sub-parameter is usually smaller than the size of the question it corresponds to, the processing workload for determining the question corresponding to the same sub-parameter is reduced, thereby improving the efficiency of question matching. Furthermore, grouping the sub-parameters first and then determining the common sub-parameters within the parameter group can be understood as first performing a coarse classification of each question, followed by further subdivision. This also improves the efficiency of determining the question corresponding to the same sub-parameter, thereby improving the efficiency of question matching.

[0140] After matching test questions, the source of other similar test questions can be linked to the matched test questions, so that users can expand their learning to a wider range of knowledge points based on some of the similar test questions.

[0141] In one embodiment of this disclosure, referring to Figure 7, a flowchart of the seventh test question matching method is provided. In this embodiment, the above method includes the following steps S701-S705.

[0142] Step S701: Perform content transformation processing on the content of each test question to obtain content parameters that represent the content of each test question.

[0143] Step S702: Divide the content parameters corresponding to each question into multiple sub-parameters.

[0144] Step S703: Match the questions with the same sub-parameters to obtain similar questions.

[0145] The steps S701-S703 described above are the same as those S101-S103 described above, and will not be repeated here.

[0146] Step S704: If there is a first question among the similar questions that comes from a solution video, then attach the solution video from the first question to each second question.

[0147] The second question is: other similar questions besides the first question.

[0148] For example, if questions f1, f2, and f3 are similar questions, and question f1 comes from the solution video m, then question f1 is the first question, and questions f2 and f3 are the second questions. In this case, the solution video m can be attached to questions f2 and f3 respectively.

[0149] In addition, when obtaining the test questions to be matched, a video of the solution can be obtained, and the test questions in the video can be identified using OCR (Optical Character Recognition) technology.

[0150] Step S705: If there is a third question in the similar test questions that comes from the test paper document, then attach the test paper document from which the third question comes to each fourth question.

[0151] The fourth question is: other similar questions besides the third question.

[0152] This step is similar to step S704 above, and will not be repeated here.

[0153] In addition, when obtaining the test questions to be matched, the test paper document can be obtained and disassembled to obtain the individual test questions in the test paper document.

[0154] In one embodiment of this disclosure, the test questions in the aforementioned similar test questions may also originate from other resources, and this embodiment of the disclosure does not limit the scope of such other resources. In this case, the other resources can be attached to other test questions in the similar test questions.

[0155] As can be seen from the above, when applying the solution provided in this embodiment of the present disclosure for question matching, after obtaining similar questions, if there are questions among the similar questions that originate from solution videos or test paper documents, the solution videos or test paper documents can be attached to other questions. This allows users to obtain the attached solution videos or test paper documents based on the aforementioned other videos, thereby helping users to expand their learning to a wider range of knowledge points, thereby increasing user stickiness and expanding the user base.

[0156] After obtaining the test questions corresponding to the same sub-parameters, in addition to using the method mentioned in step S103 above for test question matching, test question matching can also be performed through steps S803-S804 in the embodiment shown in Figure 8 below.

[0157] In one embodiment of this disclosure, referring to Figure 8, a flowchart of the eighth test item matching method is provided. In this embodiment, the above method includes the following steps S801-S804.

[0158] Step S801: Perform content transformation processing on the content of each test question to obtain content parameters that represent the content of each test question.

[0159] Step S802: Divide the content parameters corresponding to each question into multiple sub-parameters.

[0160] The steps S801 and S802 described above are the same as steps S101 and S102 described above, and will not be repeated here.

[0161] Step S803: Obtain the similarity between test questions with the same sub-parameter.

[0162] Specifically, the similarity between questions with the same sub-parameter can be obtained through either of the following two implementation methods.

[0163] In the first implementation, keywords can be extracted from each question corresponding to the same sub-parameter, and the similarity of the keywords of each question can be calculated as the similarity between questions corresponding to the same sub-parameter.

[0164] In the second implementation, the text similarity of the entire content of each question corresponding to the same sub-parameter can be calculated. Furthermore, if the question content includes images in the question, the similarity between images in each question can also be calculated.

[0165] Step S804: Based on the obtained similarity, identify similar test questions among those with the same sub-parameters.

[0166] Specifically, there can be multiple questions corresponding to the same sub-parameter, resulting in multiple similarity scores. Each similarity score represents the similarity between any two questions within those multiple similarity scores. In this case, when determining similar questions based on similarity scores, questions with similarity scores greater than a preset threshold can be identified as similar questions. Alternatively, questions with the highest multiple similarity scores can also be identified as similar questions.

[0167] As can be seen from the above, when using the scheme provided in the embodiments of this disclosure for question matching, the similarity between questions can accurately reflect the degree of similarity between them. Thus, based on the similarity between questions corresponding to the same sub-parameters, similar questions can be accurately identified among those corresponding to the same sub-parameters. Therefore, applying the question matching scheme provided in the embodiments of this disclosure can improve the accuracy of question matching.

[0168] Corresponding to the above-described test question matching method, this disclosure also provides a test question matching device.

[0169] In one embodiment of this disclosure, referring to FIG9, a structural schematic diagram of a first type of test item matching device is provided. In this embodiment, the device includes:

[0170] The content conversion module 901 is used to perform content conversion processing on the content of each test question to obtain content parameters that represent the content of each test question;

[0171] The parameter segmentation module 902 is used to segment the content parameters corresponding to each question into multiple sub-parameters;

[0172] The question matching module 903 is used to match questions with the same sub-parameters to obtain similar questions.

[0173] As can be seen from the above, when using the solution provided in this embodiment of the present disclosure for question matching, the content parameter represents the content of the question. After dividing the content parameter corresponding to the question into multiple sub-parameters, each sub-parameter also has a representational nature for the question. In this way, questions corresponding to the same sub-parameter can be understood as relatively similar questions. Thus, by matching questions corresponding to the same sub-parameter, similar questions can be accurately matched among all questions.

[0174] Furthermore, by utilizing the sub-parameters corresponding to each question, questions with the same sub-parameter can be filtered out. This allows for matching only questions with the same sub-parameter, rather than matching all questions, thus reducing the amount of data involved in the matching process. Moreover, for a single sub-parameter, one computing node can be used to match questions corresponding to that sub-parameter; for multiple sub-parameters, multiple computing nodes can be used in parallel to match questions corresponding to multiple sub-parameters. This shortens the matching time and improves the efficiency of question matching.

[0175] In one embodiment of this disclosure, the question matching module 903 is specifically used for:

[0176] Based on the values ​​of the sub-parameters, the sub-parameters corresponding to each question are grouped to obtain the parameter groups corresponding to each preset calculation node. Sub-parameters with the same value belong to the same parameter group.

[0177] For each parameter group, obtain similar questions obtained by matching the question to be matched by the computing node corresponding to the parameter group. The question to be matched is the question corresponding to the sub-parameter with the same value in the parameter group.

[0178] As can be seen from the above, when using the solution provided in this embodiment for question matching, the sub-parameters corresponding to each question are grouped according to their values, ensuring that sub-parameters with the same values ​​belong to the same parameter group. Thus, during question matching, it is only necessary to determine the common sub-parameters within each parameter group and find the corresponding question based on the determined sub-parameters, thereby performing the matching. No operation on the question itself is required when determining the question corresponding to the same sub-parameter. Since the size of a sub-parameter is usually smaller than the size of the question it corresponds to, the processing workload for determining the question corresponding to the same sub-parameter is reduced, thereby improving the efficiency of question matching. Furthermore, grouping the sub-parameters first and then determining the common sub-parameters within the parameter group can be understood as first performing a coarse classification of each question, followed by further subdivision. This also improves the efficiency of determining the question corresponding to the same sub-parameter, thereby improving the efficiency of question matching.

[0179] In one embodiment of this disclosure, referring to Figure 10, a structural schematic diagram of a second type of test question matching device is provided. In this embodiment, the device includes:

[0180] The content conversion module 1001 is used to perform content conversion processing on the content of each test question to obtain content parameters that represent the content of each test question;

[0181] The parameter segmentation submodule 1002 is used to segment the content parameters corresponding to each question into multiple intermediate parameters with a preset number of bits, and determine multiple sub-parameters corresponding to the question based on each intermediate parameter and the order of the intermediate parameters.

[0182] The question matching module 1003 is used to match questions with the same sub-parameters to obtain similar questions.

[0183] As can be seen from the above, when applying the solution provided in this embodiment of the present disclosure for question matching, for each question, the content parameters corresponding to the question are divided into multiple intermediate parameters, and based on each intermediate parameter and the order of the intermediate parameters, multiple sub-parameters corresponding to the question can be accurately determined. Thus, by using the more accurate sub-parameters corresponding to each question for question matching, the accuracy of question matching can be improved.

[0184] In one embodiment of this disclosure, the parameter segmentation submodule 1002 is specifically used for:

[0185] For each question, the content parameters corresponding to the question are divided into multiple intermediate parameters with a preset number of bits. According to the order of the intermediate parameters, the sub-parameters to which each intermediate parameter belongs are determined in turn, and the intermediate parameters corresponding to the same sub-parameter are connected to obtain multiple sub-parameters corresponding to the question.

[0186] As can be seen from the above, when applying the solution provided in this embodiment of the present disclosure for question matching, after dividing the content parameters corresponding to each question into multiple intermediate parameters, the sub-parameters to which each intermediate parameter belongs are determined in turn according to the arrangement order of each intermediate parameter, and the intermediate parameters corresponding to the same sub-parameter are connected. This can accurately obtain multiple sub-parameters corresponding to each question, thereby improving the accuracy of question matching by using the more accurate sub-parameters corresponding to each question.

[0187] In one embodiment of this disclosure, the parameter segmentation submodule 1002 is specifically used for:

[0188] For each test question, if there are multiple preset number of digits, for each preset number of digits, the content parameter corresponding to the test question is divided into multiple intermediate parameters of the preset number of digits. Based on each intermediate parameter and the order of the intermediate parameters, multiple sub-parameters corresponding to the test question are determined.

[0189] As can be seen from the above, when applying the solution provided in this embodiment of the present disclosure for question matching, in the presence of multiple preset number of digits, multiple sub-parameters can be obtained using each preset number of digits. In this way, by using multiple preset number of digits, the number of sub-parameters corresponding to each question can be increased. Thus, based on the sub-parameters corresponding to each question, the number of questions corresponding to the same sub-parameters can be increased, thereby increasing the matching quantity for question matching, preventing similar questions from being missed, and improving the accuracy of question matching.

[0190] In one embodiment of this disclosure, the content conversion module 901 is specifically used for:

[0191] For each question, a locality-sensitive hash (simhash) is calculated on the question content to obtain the simhash value of the question, which serves as the content parameter representing the question content.

[0192] As can be seen from the above, when applying the scheme provided in this embodiment of the present disclosure for question matching, the simhash value obtained after performing local sensitive hash (simhash) calculation can accurately reflect the question content. Therefore, by using the simhash value of the question as a content parameter, the content parameter of the question content can be accurately obtained. Thus, by performing question matching based on the more accurate content parameter corresponding to the question, the accuracy of question matching can be improved.

[0193] In one embodiment of this disclosure, referring to Figure 11, a structural schematic diagram of a third type of test question matching device is provided. In this embodiment, the device includes:

[0194] The content conversion module 1101 is used to perform content conversion processing on the content of each test question to obtain content parameters that represent the content of each test question.

[0195] The parameter segmentation module 1102 is used to segment the content parameters corresponding to each question into multiple sub-parameters;

[0196] The question matching module 1103 is used to match questions with the same sub-parameters to obtain similar questions;

[0197] The source mounting module 1104 is used to, after obtaining similar test questions, if there is a first test question in the similar test questions that comes from a solution video, then mount the solution video from the source of the first test question to each second test question; if there is a third test question in the similar test questions that comes from a test paper document, then mount the test paper document from the source of the third test question to each fourth test question. The second test question is: other test questions in the similar test questions besides the first test question, and the fourth test question is: other test questions in the similar test questions besides the third test question.

[0198] As can be seen from the above, when applying the solution provided in this embodiment of the present disclosure for question matching, after obtaining similar questions, if there are questions among the similar questions that originate from solution videos or test paper documents, the solution videos or test paper documents can be attached to other questions. This allows users to obtain the attached solution videos or test paper documents based on the aforementioned other videos, thereby helping users to expand their learning to a wider range of knowledge points, thereby increasing user stickiness and expanding the user base.

[0199] In one embodiment of this disclosure, the question matching module 903 is specifically used for:

[0200] Obtain the similarity between test questions that have the same sub-parameters;

[0201] Based on the obtained similarity scores, similar test questions are identified among those with the same sub-parameters.

[0202] As can be seen from the above, when using the scheme provided in the embodiments of this disclosure for question matching, the similarity between questions can accurately reflect the degree of similarity between them. Thus, based on the similarity between questions corresponding to the same sub-parameters, similar questions can be accurately identified among those corresponding to the same sub-parameters. Therefore, applying the question matching scheme provided in the embodiments of this disclosure can improve the accuracy of question matching.

[0203] In one embodiment of this disclosure, the test questions include at least one of the following information:

[0204] The question stem, the answer choices, the images in the question, and the characters in the formulas in the question.

[0205] In this solution, the above information can accurately reflect the content of the test questions. By using at least one of the above information as test question content for content conversion processing, the accuracy of the obtained content parameters can be improved. Furthermore, by matching test questions based on the more accurate content parameters, the accuracy of test question matching can be improved.

[0206] According to embodiments of this disclosure, this disclosure also provides an electronic device, a readable storage medium, and a computer program product.

[0207] In one embodiment of this disclosure, an electronic device is provided, comprising:

[0208] At least one processor; and

[0209] A memory communicatively connected to the at least one processor; wherein,

[0210] 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 any of the question matching methods in the foregoing method embodiments.

[0211] In one embodiment of this disclosure, a non-transitory computer-readable storage medium storing computer instructions is provided, wherein the computer instructions are used to cause the computer to execute any of the question matching methods in the foregoing method embodiments.

[0212] In one embodiment of this disclosure, a computer program product is provided, including a computer program that, when executed by a processor, implements any of the question matching methods in the foregoing method embodiments.

[0213] Figure 12 illustrates a schematic block diagram of an example electronic device 800 that can be used to implement embodiments of the present disclosure. 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 present disclosure described and / or claimed herein.

[0214] As shown in Figure 12, device 1200 includes a computing unit 1201, which can perform various appropriate actions and processes according to a computer program stored in read-only memory (ROM) 1202 or a computer program loaded from storage unit 1208 into random access memory (RAM) 1203. The RAM 1203 may also store various programs and data required for the operation of device 1200. The computing unit 1201, ROM 1202, and RAM 1203 are interconnected via bus 1204. Input / output (I / O) interface 1205 is also connected to bus 1204.

[0215] Multiple components in device 1200 are connected to I / O interface 1205, including: input unit 1206, such as keyboard, mouse, etc.; output unit 1207, such as various types of monitors, speakers, etc.; storage unit 1208, such as disk, optical disk, etc.; and communication unit 1209, such as network card, modem, wireless transceiver, etc. Communication unit 1209 allows device 1200 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0216] The computing unit 1201 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 1201 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 1201 performs the various methods and processes described above, such as the question matching method. For example, in some embodiments, the question matching method may be implemented as a computer software program tangibly contained in a machine-readable medium, such as storage unit 1208. In some embodiments, part or all of the computer program may be loaded and / or installed on device 1200 via ROM 1202 and / or communication unit 1209. When the computer program is loaded into RAM 1203 and executed by the computing unit 1201, one or more steps of the question matching method described above may be performed. Alternatively, in other embodiments, the computing unit 1201 may be configured to perform the question matching method by any other suitable means (e.g., by means of firmware).

[0217] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments 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.

[0218] The program code used to implement the methods of this disclosure 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 apparatus, such that when executed by the processor or controller, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be 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.

[0219] In the context of this disclosure, 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. Machine-readable media 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, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0220] 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).

[0221] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as a data server), or computing systems that include middleware components (e.g., an application server), or computing systems that include frontend components (e.g., a user computer with a graphical user interface or web browser through which a user can interact with embodiments 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 of any form or medium (e.g., a communication network). Examples of communication networks include local area networks (LANs), wide area networks (WANs), and the Internet.

[0222] Computer systems can include clients and servers. Clients and servers are generally located far apart and typically interact via communication networks. Client-server relationships are created by computer programs running on the respective computers and having a client-server relationship with each other. Servers can be cloud servers, servers in distributed systems, or servers incorporating blockchain technology.

[0223] 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 disclosure can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution disclosed in this disclosure can be achieved, and this is not limited herein.

[0224] The specific embodiments described above do not constitute a limitation on the scope of protection of this disclosure. 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 disclosure should be included within the scope of protection of this disclosure.

Claims

1. A test item matching method, comprising: The content of each test question is transformed to obtain content parameters that represent the content of each test question; The content parameters corresponding to each test question are divided into multiple sub-parameters. Based on the sub-parameters corresponding to each test question, test questions with the same sub-parameters are identified among the test questions. The identified test questions are then matched to obtain similar test questions. Sub-parameters are representative of test questions. The process of dividing the content parameters corresponding to each test question into multiple sub-parameters includes: for each test question, dividing the content parameters corresponding to the test question into multiple intermediate parameters of a preset number of positions, and determining multiple sub-parameters corresponding to the test question based on the intermediate parameters and their arrangement order. The process of determining multiple sub-parameters corresponding to the test question based on the intermediate parameters and their arrangement order includes: setting the expected number of sub-parameters to be obtained; after dividing and obtaining multiple intermediate parameters, allocating each expected number of intermediate parameters to the expected number of sub-parameters according to their arrangement order, thereby determining the sub-parameters corresponding to each intermediate parameter; and connecting the intermediate parameters corresponding to the same sub-parameter to obtain the expected number of sub-parameters corresponding to the test question.

2. The method according to claim 1, wherein, Matching questions with the same sub-parameters to obtain similar questions includes: grouping the sub-parameters corresponding to each question according to their values ​​to obtain parameter groups corresponding to each preset calculation node, wherein sub-parameters with the same values ​​belong to the same parameter group; for each parameter group, obtaining similar questions obtained by matching the questions to be matched by the calculation node corresponding to that parameter group, wherein the questions to be matched are the questions corresponding to the sub-parameters with the same values ​​in that parameter group.

3. The method according to claim 1, wherein, The step of dividing the content parameter corresponding to the test question into multiple intermediate parameters of a preset number of digits and determining multiple sub-parameters corresponding to the test question according to each intermediate parameter and its arrangement order includes: when there are multiple preset number of digits, for each preset number of digits, dividing the content parameter corresponding to the test question into multiple intermediate parameters of the preset number of digits, and determining multiple sub-parameters corresponding to the test question according to each intermediate parameter and its arrangement order.

4. The method according to claim 1 or 2, wherein, The process of converting the content of each test question to obtain content parameters representing the content of each test question includes: for each test question, performing local sensitive hash (simhash) calculation on the content of the test question to obtain the simhash value of the test question, which is used as the content parameter representing the content of the test question.

5. The method according to claim 1 or 2, wherein, After obtaining similar test questions, the process further includes: if there is a first test question among the similar test questions that originates from a solution video, then each second test question is attached with the solution video from which the first test question originated, wherein the second test question is: other test questions among the similar test questions besides the first test question; if there is a third test question among the similar test questions that originates from a test paper document, then each fourth test question is attached with the test paper document from which the third test question originated, wherein the fourth test question is: other test questions among the similar test questions besides the third test question.

6. The method according to claim 1, wherein, Matching test questions with the same sub-parameters to obtain similar test questions includes: obtaining the similarity between test questions with the same sub-parameters; and determining similar test questions among those with the same sub-parameters based on the obtained similarity.

7. The method according to claim 1 or 2, wherein, The test questions include at least one of the following: the question stem, the question options, the image content in the question, and the characters in the question formula.

8. A test item matching device, comprising: The content conversion module is used to convert the content of each test question to obtain content parameters that represent the content of each test question. The parameter segmentation module is used to segment the content parameters corresponding to each test question into multiple sub-parameters. The test question matching module is used to determine the test questions with the same sub-parameters in each test question based on the sub-parameters corresponding to each test question, and then match the determined test questions to obtain similar test questions. The sub-parameters are representative of the test questions. The parameter segmentation module includes: a parameter segmentation submodule, used to segment the content parameters corresponding to each test question into multiple intermediate parameters of a preset number of positions, and determine multiple sub-parameters corresponding to the test question based on each intermediate parameter and its arrangement order. The parameter segmentation submodule is specifically used to: segment the content parameters corresponding to each test question into multiple intermediate parameters of a preset number of positions, set the expected number of sub-parameters to be obtained, and after segmenting into multiple intermediate parameters, divide each expected number of intermediate parameters into the expected number of sub-parameters according to the arrangement order of each intermediate parameter, thereby determining the sub-parameters corresponding to each intermediate parameter, and connecting the intermediate parameters corresponding to the same sub-parameter to obtain the expected number of sub-parameters corresponding to the test question.

9. The apparatus according to claim 8, wherein, The question matching module is specifically used for: grouping the sub-parameters corresponding to each question according to the value of the sub-parameters to obtain each parameter group corresponding to each preset calculation node, wherein sub-parameters with the same value belong to the same parameter group; for each parameter group, obtaining similar questions obtained by matching the question to be matched by the calculation node corresponding to the parameter group, wherein the question to be matched is: the question corresponding to the sub-parameter with the same value in the parameter group.

10. The apparatus according to claim 8, wherein, The parameter segmentation submodule is specifically used for: for each question, in the case of multiple preset number of digits, for each preset number of digits, segmenting the content parameter corresponding to the question into multiple intermediate parameters of the preset number of digits, and determining multiple sub-parameters corresponding to the question based on each intermediate parameter and the order of the intermediate parameters.

11. The apparatus according to claim 8 or 9, wherein, The content conversion module is specifically used to: for each question, perform local sensitive hash (simhash) calculation on the question content to obtain the simhash value of the question, which is used as a content parameter representing the question content.

12. The apparatus according to claim 8 or 9, wherein, It also includes: a source mounting module, used to, after obtaining similar test questions, if there is a first test question in the similar test questions that comes from a solution video, then mount the solution video from the source of the first test question for each second test question; if there is a third test question in the similar test questions that comes from a test paper document, then mount the test paper document from the source of the third test question for each fourth test question, wherein the second test question is: other test questions in the similar test questions besides the first test question, and the fourth test question is: other test questions in the similar test questions besides the third test question.

13. The apparatus according to claim 8, wherein, The question matching module is specifically used to: obtain the similarity between questions with the same sub-parameter; and determine similar questions among those with the same sub-parameter based on the obtained similarity.

14. An electronic device comprising: At least one processor; And a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being 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.

15. 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.

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

Images