Online intelligent reviewing system based on big data

Abstract

The invention belongs to the technical field of online intelligent marking of student homework, and particularly discloses and provides an online intelligent marking system based on big data, which comprises the steps of identifying each fuzzy character in each student homework, and comparing the stroke writing characteristics and arrangement modes of each fuzzy character and each similar character with each fuzzy character, determining calibration characters of each fuzzy character in the homework of each student; by comparing the corresponding answer writing contour of the subjective question answer of each student with each standard answer in the network public platform and comparing with the reference answer, identifying the contradictory index of the similar part of the subjective question answer of each student and the network standard answer, and determining the network reference score of the student according to the contradictory index of the similar part of the subjective question answer of each student and the network standard answer; by comparing the homework answers among the students, all suspected students with the homework answers possibly having interactive reference are screened, then the homework progress indexes of all the suspected students are detected by identifying the historical homework submission records of all the suspected students, and the homework interactive reference scores of all the students are determined.

Description

Technical Field

[0001] The present invention belongs to the technical field of online intelligent marking of student assignments, and relates to an online intelligent marking system based on big data. Background Art

[0002] With the booming development of online education, traditional teaching models and homework grading methods can no longer meet the growing needs of personalized learning and efficient management. In the traditional education system, teachers need to spend a lot of time manually grading homework, which not only affects teaching efficiency, but also limits teachers' investment in teaching innovation, personalized tutoring, etc. In this context, the "online intelligent grading of students' homework" technology came into being. Online intelligent grading technology can automatically handle a large number of basic grading tasks, effectively reducing the burden on teachers, allowing them to focus more on improving teaching quality and guiding students' personalized development, and has become an important invention and breakthrough in the field of education.

[0003] Some related solutions for online homework marking in the prior art, for example, a Chinese invention patent application for an online homework intelligent marking system and method with publication number CN109903014A, which plans the homework format so that the big data analysis module can accurately extract data from the data storage module for processing, and quickly generate different charts, so that students, parents, teachers and principals can quickly grasp the learning status. In the wrong question bank, students can search according to chapters, units, sections or time sequence, and retrieve the knowledge points that have not been passed for review; log in through different ports such as student login port, teacher login port, parent login port and principal login port to meet the operation needs of different personnel and facilitate use. This can provide materials for big data analysis, collect data in a concomitant manner, and accurately portray the student's learning status, thereby forming a learning status report for a certain time or stage, so that students, parents, teachers and principals can clearly understand the situation of learning and education.

[0004] Another Chinese invention patent application with patent number CN114298875A is a method and system for automatic scoring of objective questions based on blockchain. It creates an automatic scoring platform based on the blockchain framework. The homework node reads the student information and homework information, screens out the student information and objective question information to be reviewed, and writes the student information and objective question information to be reviewed into the automatic scoring platform; the review node reads the objective question information to be reviewed and the correct answer, and corrects the objective question information to be reviewed based on bit operations to obtain the correction results; the statistical node calculates the objective question scores based on the correction results, generates an objective question score table based on the student information to be reviewed, and writes the score table into the automatic scoring platform.

[0005] Although the above scheme proposes some solutions for online marking of students' homework, it still has the following limitations, specifically: 1. The existing technology is relatively lacking in the marking and evaluation of subjective questions, and when judging the answers to the questions, the student's written answers are more often compared with the reference answers, without considering the impact of the student's own blurred handwriting on the answer evaluation, which may cause students to ignore the value of handwriting.

[0006] 2. Failure to identify online answers to students' assignments and references to answers between students may cause students to ignore the value of academic integrity. Summary of the invention

[0007] In view of this, in order to solve the problems raised in the above background technology, an online intelligent review system based on big data is proposed.

[0008] The purpose of the present invention can be achieved through the following technical solutions: The present invention provides an online intelligent marking system based on big data, the system comprising: a homework information extraction module: used to count the homework sets of all students and classify the homework of each student, including subjective questions and objective questions, wherein the objective questions include filling-in-the-blank questions and multiple-choice questions, and the students are numbered 1, 2, ...r..., n.

[0009] Writing recognition module: used to scan and obtain the answers to various questions in each student's homework, extract the writing outline of the answers, and then identify the written scores of each student's homework.

[0010] Recognition accuracy correction module: used to screen out the blurred text and blurred letters that do not meet the standard in the assignments of each student, and then accurately calibrate the outlines of the fonts that do not meet the standard.

[0011] Network citation scoring module: used to match the answers to the subjective questions of each student's homework with the public answers on the Internet, list the similar parts of the subjective questions of each student's homework and their sources, evaluate the paradoxical index of the similar parts of the subjective questions of each student's homework, and determine the network citation score D′ of the subjective questions of each student's homework r .

[0012] Student interactive scoring module: used to compare the answers of each student's subjective questions and determine the interactive reference score D" of each student's homework r .

[0013] Result feedback module: used to obtain the outline similarity between the answers to various questions in each student's homework and the reference answers, and determine the final score of each student's homework.

[0014] Compared with the prior art, the beneficial effects of the present invention are as follows: (1) The present invention identifies the ambiguous characters that do not meet the standard in the homework of each student, and compares the stroke writing features and arrangement methods of each ambiguous character with its similar characters to determine the calibration text of each ambiguous character in the homework of each student, thereby reducing the influence of the student's own ambiguous handwriting on the answer judgment, thereby increasing the accuracy of the student's homework marking.

[0015] (2) The present invention determines the written score of each student's homework by identifying the writing standardization indicators of the answers to each student's homework. Students can better understand their own writing habits and weaknesses, and then cultivate the ability of self-reflection, which is very important for personal growth and development.

[0016] (3) The present invention compares the writing outlines of the answers to the subjective questions of each student's homework with the standard answers in the online public platform, and at the same time compares the writing outlines of the reference answers, thereby identifying the paradoxical indicators of the similar parts between the answers to the subjective questions of each student's homework and the standard answers in the online public platform. This can objectively evaluate the students' citations, avoid errors caused by subjective judgment, and make the evaluation process more fair. Based on this, the students' online citation scores are determined. This evaluation method encourages students to rely more on their own understanding and analysis when completing their homework, and reduce their reliance on direct copying and pasting, thereby promoting the improvement of critical thinking and independent thinking ability, and effectively detecting whether there is plagiarism or dishonesty in students' homework.

[0017] (4) The present invention compares the homework answers of different students to screen out the suspected students whose homework answers may have cross-references, and then detects their homework progress indicators by identifying their historical homework submission records. Based on this, the homework cross-reference score of each student is determined, thereby increasing the reliability of the final review and scoring of the students' homework. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for describing the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0019] Figure 1 This is a schematic diagram of the connection of the system modules of the present invention.

[0020] Figure 2 It is a schematic diagram of the blurred text of the present invention.

[0021] Figure 3 It is a schematic diagram of the blurred letters of the present invention. DETAILED DESCRIPTION

[0022] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0023] See also Figure 1 As shown, the present invention provides an online intelligent marking system based on big data, which includes: a homework information extraction module, a handwriting recognition module, a recognition accuracy correction module, a network citation scoring module, a student interaction scoring module, and result feedback.

[0024] The homework information extraction module is connected to the handwriting recognition module, the handwriting recognition module is connected to the recognition accuracy correction module, the recognition accuracy correction module is connected to the network citation scoring module, the network citation scoring module is connected to the student interaction scoring module, and the student interaction scoring module is connected to the result feedback.

[0025] The homework information extraction module is used to count the homework sets of all students and classify the homework of each student into subjective questions and objective questions, wherein the objective questions include filling-in-the-blank questions and multiple-choice questions. The students are numbered 1, 2, ...r..., n.

[0026] The writing recognition module is used to scan and obtain the answers to various questions in each student's homework, and extract the writing outline of the answer, and then identify the written score of each student's homework.

[0027] In a preferred embodiment, the written score of each student's homework is identified in the following manner: the writing outlines of the answers to each type of question in each student's homework are matched with all the characters and letters in the character library; if a certain character or a certain letter in the writing outlines of the answers to a certain type of question in a student's homework has a unique matching character or letter in the character library, the normative weight of the character or letter is recorded as A1, such as A1=2; if there are multiple matching characters or letters, the normative weight of the character or letter is recorded as A2, such as A2=1; if there is no matching character or letter, the normative weight of the character or letter is recorded as A3, such as A3=0.

[0028] The recognition standard for matching the characters or letters in the answer writing outline with the character library is that the similarity of the stroke writing outline is within a specified range.

[0029] The similarity of the stroke writing contour is obtained by matching the strokes of each character in the answer writing contour with the strokes of each Chinese character in the font library, extracting the comprehensive difference value of the length, width, inclination angle and other characteristics of the strokes, and then obtaining the inverse ratio of the comprehensive difference value to the preset reference benchmark difference value, which is recorded as the similarity between the stroke writing contour of each character in the answer writing contour and the stroke writing contour of each Chinese character in the font library.

[0030] The comprehensive normative weights corresponding to all the words and letters in the writing outlines of the answers to all the questions in each student's assignment are obtained statistically, and the ratio of the comprehensive normative weights to the preset reference normative weights is obtained and recorded as the writing normative index of the answers to each student's assignment.

[0031] The writing standardization index of the answer of each student's homework is further matched with the corresponding written score of each preset writing standardization index range to determine the written score of each student's homework.

[0032] The present invention determines the written score of each student's homework by identifying the writing standardization index of the answer of each student's homework. Students can better understand their own writing habits and weaknesses, and then cultivate the ability of self-reflection, which is very important for personal growth and development.

[0033] See also Figure 2 , 3 As shown, the recognition accuracy correction module is used to screen out the fuzzy characters and letters that do not meet the standard in the homework of each student, and then perform accuracy calibration on the outlines of the fonts that do not meet the standard.

[0034] In a preferred embodiment, the specific screening method for the fuzzy words and fuzzy letters that do not meet the standardization standards in the assignments of each student is: extracting the words or letters with standardization weights of A2 and A3 from the writing outlines of the corresponding answers to various types of questions in the assignments of each student, and then marking the words with standardization weights of A2 and A3 as fuzzy words, and marking the letters with standardization weights of A2 and A3 as fuzzy letters.

[0035] In a further preferred embodiment, the accuracy calibration of the font outlines that do not meet the standardization standards includes: extracting each character and its outline in the character library that matches each fuzzy character that does not meet the standardization standards in each student's homework, recording them as similar characters, and extracting the similarity between each fuzzy character in each student's homework and its similar characters from the similarity between each character in the writing outline of the answer and the writing outline of each Chinese character stroke in the character library.

[0036] Based on the contour edge detection technology, the habitual writing features and habitual arrangements of each stroke of each student are identified from the writing contours of the corresponding answers to various questions in the students' homework, such as horizontal, vertical, dot, left-falling stroke, right-falling stroke, hook, and turn strokes, writing features such as contour length, width, and inclination angle, arrangement methods such as the distance between adjacent strokes, the position of the starting and ending points of the pen compared to the overall area of ​​the text, and the thickness and length of the connecting contour with adjacent strokes.

[0037] Obtain the writing contours corresponding to each fuzzy character that does not meet the standardization standards in each student's homework, extract the fixed-frame strokes corresponding to each fuzzy character in each student's homework, identify the writing features to which they belong, compare them with the inertial writing features of each stroke by the corresponding student, obtain the difference values ​​between the fixed-frame strokes corresponding to each fuzzy character in each student's homework and the inertial writing features of each stroke, sum them up to obtain the comprehensive difference values ​​between the fixed-frame strokes corresponding to each fuzzy character in each student's homework and each stroke, screen out the strokes with the minimum comprehensive difference value, and determine them as the sub-categorized writing trend strokes of the fixed-frame strokes corresponding to the corresponding fuzzy character in the corresponding student's homework, obtain the sub-categorized writing trend strokes of the fixed-frame strokes corresponding to each fuzzy character in each student's homework, reorganize and construct the sub-categorized writing trend strokes of the fixed-frame strokes corresponding to each fuzzy character in each student's homework, and obtain the sub-categorized writing trend characters of each fuzzy character in each student's homework.

[0038] From the writing contours corresponding to the fuzzy characters that do not meet the standardization standards in the assignments of each student, identify the arrangement modes of the fixed-frame strokes corresponding to each fuzzy character in each student's assignment, match them with the inertial arrangement modes of the strokes of the corresponding students, assign weights to the matched arrangement modes, and count the total weight assignments of the matches between the fixed-frame strokes corresponding to each fuzzy character in each student's assignment and the arrangement modes of each stroke, and then determine the stroke to which the maximum total weight assignment belongs as the supplementary writing trend stroke of the fixed-frame stroke of the corresponding fuzzy character in the assignment of the corresponding student, and obtain the supplementary writing trend strokes of the fixed-frame strokes corresponding to each fuzzy character in the assignments of each student, and then reconstruct and obtain the supplementary writing trend Chinese characters of each fuzzy character in the assignments of each student.

[0039] Specifically, the matching arrangement modes are weighted, such as: if an arrangement mode of a fixed-frame stroke corresponding to a fuzzy word in a student's homework matches a habitual arrangement mode of a stroke of the corresponding student, the weight of the arrangement mode is assigned to 1, and if it does not match, the weight is assigned to 0.

[0040] Similarity values ​​are assigned to the subcategorized Chinese characters with writing trends and the supplementary Chinese characters with writing trends, and the values ​​are added to the similarities between each fuzzy character and its similar characters in each student's assignment to obtain a final similarity value between each fuzzy character and its similar characters in each student's assignment. Then, the similar character to which the maximum final similarity value of each fuzzy character in each student's assignment belongs is used as the calibration text of the corresponding fuzzy character in the corresponding student's assignment.

[0041] The similarity assignment for the detailed writing trend Chinese characters and the supplementary writing trend Chinese characters is as follows: the similarity of the detailed writing trend Chinese characters is assigned to z1, and the similarity of the supplementary writing trend Chinese characters is assigned to z2, where z1 and z2 are set constants, such as z1=0.4 and z2=0.6.

[0042] In a further preferred embodiment, the accuracy calibration of the font outlines that do not meet the standardization standards also includes: extracting the letters and their outlines in the character library that match the fuzzy letters that do not meet the standardization standards in the homework of each student, and recording them as similar letters.

[0043] According to the calibration text recognition method of each blurred text in each student's homework, the calibration letters of each blurred letter in each student's homework are determined similarly.

[0044] The present invention determines the calibration text of each ambiguous character in each student's homework by identifying the ambiguous characters that do not meet the standardization standards and comparing the stroke writing features and arrangement methods of each ambiguous character with similar characters, thereby reducing the influence of the students' own ambiguous handwriting on the judgment of the answers, thereby increasing the accuracy of marking the students' homework.

[0045] The network citation scoring module is used to match the answers to the subjective questions of each student's homework with the public answers on the Internet, list the similar parts of the subjective questions of each student's homework and their sources, evaluate the paradoxical index of the similar parts of the subjective questions of each student's homework, and determine the network citation score D′ of the subjective questions of each student's homework. r .

[0046] In a preferred embodiment, the process of listing the network similar parts and their sources in the answers to the subjective questions of each student's homework is as follows: the writing contours of the corresponding answers to the subjective questions of each student's homework are divided into text groups to obtain the writing contours of the corresponding text groups of the answers to the subjective questions of each student's homework, and the standard answers to the subjective questions in the online public platform are counted, such as the answers obtained by searching chat according to different combination patterns of the keywords in the questions and the answers obtained by searching various search software, and the text contours of the standard answers to the subjective questions are integrated, and they are matched with the writing contours of the corresponding text groups of the answers to the subjective questions of each student's homework, and the similarity between the writing contours of the corresponding text groups of the answers to the subjective questions of each student's homework and the text contours of the standard answers of the corresponding questions are obtained in a similar way according to the similarity of the writing contours of the strokes, and the text contours of the text groups of the corresponding answers to the subjective questions of each student's homework are screened out, that is, the network similar parts in the answers to the subjective questions of each student's homework, and the text contours of the standard answers to which they belong are marked as the sources of the corresponding network similar parts.

[0047] In a further preferred embodiment, the paradoxical index for evaluating the similar parts of the answers to the subjective questions of each student's assignment is specifically as follows: obtaining reference answers to various questions submitted in advance by teachers in the marking platform, scanning to obtain the writing outlines of the reference answers to various questions, comparing them with the network-similar parts in the answers to the subjective questions of each student's assignment, and obtaining the similarity of the writing outlines of the strokes. Similarly, identifying the similarity between the writing outlines of the corresponding answers to the network-similar parts in the answers to the subjective questions of each student's assignment and the writing outlines of the corresponding parts in the reference answers to the corresponding questions; if the similarity between the writing outlines of the corresponding answers to a network-similar part in the answers to the subjective questions of a student's assignment and the writing outlines of the corresponding parts in the reference answers to the corresponding questions is within a set standard range, then the network-similar part in the student's answer to the subjective question is marked as similar and reasonable; otherwise, the network-similar part in the student's answer to the subjective question is weighted by citation.

[0048] The corresponding weighted value for citing weighted parts of similar parts of students' answers to subjective questions is set to g, where g is a set constant, such as g=1. Based on this, the citation weighted values ​​of all similar parts of all students' answers to subjective questions are obtained. The comprehensive citation weighted value of each student's answer to the subjective question is summed up, and its ratio to the preset benchmark citation weight is obtained to obtain the paradoxical index of the similar parts of each student's answer to the subjective question.

[0049] In a further preferred embodiment, the network citation score of the answer to the subjective question of each student's homework is determined by the formula: set the corresponding citation impact factor of each source of the network similar part, match it with the source of each network similar part in the answer to the subjective question of each student's homework, obtain the citation impact factor of each network similar part in the answer to the subjective question of each student's homework, and sum them to obtain the comprehensive citation impact factor U of the network similar part in the answer to the subjective question of each student's homework r , and then determine the network citation score of each student's answer to the subjective question of the homework Where δ r is the paradoxical index of the similarity of the answers to the subjective questions of the rth student's homework, It means round down, and e is a natural constant.

[0050] The present invention compares the writing outlines of the answers to the subjective questions of each student's homework with the standard answers in the online public platform, and at the same time compares the writing outlines of the reference answers, identifies the paradoxical indicators of the similar parts of the answers to the subjective questions of each student's homework and the standard answers in the online public platform, and can objectively evaluate the students' citations, avoid errors caused by subjective judgment, make the evaluation process more fair, and determine the students' online citation scores accordingly. This evaluation method encourages students to rely more on their own understanding and analysis when completing their homework, and reduce their reliance on direct copying and pasting, thereby promoting the improvement of critical thinking and independent thinking ability, and effectively detecting whether there is plagiarism or dishonesty in students' homework.

[0051] The student interactive scoring module is used to compare the answers of each student's subjective questions and determine the interactive reference score D" of each student's homework. r .

[0052] In a preferred embodiment, the content of determining the interactive reference score of each student's homework is: comparing the writing outlines of the answers to the subjective questions of each student's homework corresponding to the answers to the text groups, identifying the similar text group parts in the writing outlines of the answers to the subjective questions of each student's homework whose similarity with other students' homework is within a specified range, and then analyzing the paradoxical indicators of all similar text group parts of each student's homework with other students in the same way as the paradoxical indicators of the similar parts of the subjective questions of each student's homework, and screening out the students whose paradoxical indicators of all similar text group parts with other students' homework exceed the preset paradoxical indicator threshold, and record them as suspected students.

[0053] Extract the historical homework submission records of each suspected student from the review platform, identify the total number of handwriting erasures M′x of each suspected student in the current submitted homework based on the preset handwriting erasure texture features, and obtain the total number of handwriting erasures M of each suspected student in the historical homework submission records xj , compare and evaluate the progress indicators of each student's work Where M x(j+1) is the total number of handwriting erasures made by the x-th suspected student in the j+1th historical homework submission record, ΔM is the preset reference modification difference of the total number of handwriting erasures, x is the number of the suspected student, x=1,2,...,c, j is the number of historical homework submission records, j=1,2,...,q, q is the number of historical homework submission records.

[0054] Obtain the ease of the current homework submitted by the teacher in advance in the review platform, and analyze the abnormal evaluation coefficient of the homework submitted by each suspected student is the preset benchmark abnormal evaluation coefficient of the student's homework, and the abnormal evaluation coefficients of the homework submitted by all students except the suspected students are recorded as

[0055] Get the abnormal evaluation coefficient of each student's submitted assignment or It is matched with the homework interaction reference scores corresponding to the preset ranges of various abnormal evaluation coefficients to determine the homework interaction reference scores of each student.

[0056] The result feedback module is used to obtain the outline similarity between the answers to various questions of each student's homework and the reference answers, and determine the final score of each student's homework.

[0057] In a preferred embodiment, the final score of each student's homework is determined as follows: after the review platform calibrates the accuracy of the writing contours of the answers to various questions in each student's homework, the writing contours of the answers to various questions in each student's homework are matched with the reference answers to the corresponding questions pre-submitted by the teacher in the review platform, and the contour similarity μ between the answers to various questions in each student's homework and the reference answers is obtained in the same way according to the similarity acquisition method of the writing contours of the strokes. ri , i is the number of the homework type, i = 1, 2, ..., h.

[0058] Determine the final grade for each student's assignment In the formula is the written score of the rth student's homework, and h is the number of homework topic types.

[0059] The present invention screens out suspected students whose homework answers may have cross-references by comparing them with each other, and then detects their homework progress indicators by identifying their historical homework submission records, thereby determining the homework cross-reference score of each student, thereby increasing the reliability of the final review and scoring of the students' homework.

[0060] The above contents are merely examples and explanations of the concept of the present invention. Those skilled in the art may make various modifications or additions to the specific embodiments described or replace them in a similar manner. As long as they do not deviate from the concept of the invention or exceed the scope defined by the present invention, they shall all fall within the protection scope of the present invention.

Claims

1. An online intelligent review system based on big data, characterized in that: The system includes: a homework information extraction module: used to count the homework sets of all students and classify the homework of each student into subjective questions and objective questions. The students are numbered 1, 2, ... r..., n; Handwriting recognition module: used to scan and obtain the answers to various questions in each student's homework, extract the outline of the answers, and then identify the written scores of each student's homework; Recognition accuracy correction module: used to screen out the fuzzy words and letters that do not meet the standard in the assignments of each student, and then accurately calibrate the outlines of the fonts that do not meet the standard; Network citation scoring module: used to match the answers to the subjective questions of each student's homework with the public answers on the Internet, list the similar parts of the subjective questions of each student's homework and their sources, evaluate the paradoxical index of the similar parts of the subjective questions of each student's homework, and determine the network citation score D′ of the subjective questions of each student's homework r ; Student interactive scoring module: used to compare the answers of each student's subjective questions and determine the interactive reference score D" of each student's homework r : Result feedback module: used to obtain the outline similarity between the answers to various questions in each student's homework and the reference answers, and determine the final score of each student's homework.

2. According to claim 1, an online intelligent review system based on big data is characterized in that: The specific method of identifying the written score of each student's homework is as follows: matching the writing outlines of the answers to various types of questions in each student's homework with all the characters and letters in the character library; if a certain character or a certain letter in the writing outline of the answer to a certain type of question in a student's homework has a unique matching character or letter in the character library, the standard weight of the character or letter is recorded as A1; if there are multiple matching characters or letters, the standard weight of the character or letter is recorded as A2; if there is no matching character or letter, the standard weight of the character or letter is recorded as A3; Statistically obtain the comprehensive normative weights of all words and letters in the writing outlines of the answers to all questions in each student's homework, obtain the ratio of the comprehensive normative weight to the preset reference normative weight, and record it as the writing normative index of the answers to each student's homework; The writing standardization index of the answer of each student's homework is further matched with the corresponding written score of each preset writing standardization index range to determine the written score of each student's homework.

3. The online intelligent review system based on big data according to claim 2 is characterized in that: The specific screening method for the fuzzy words and fuzzy letters that do not meet the standardization standards in the assignments of each student is: extract the words or letters with standardization weights of A2 and A3 in the writing outlines of the corresponding answers to various questions in the assignments of each student, and then mark the words with standardization weights of A2 and A3 as fuzzy words, and mark the letters with standardization weights of A2 and A3 as fuzzy letters.

4. The online intelligent review system based on big data according to claim 1 is characterized in that: The accuracy calibration of the font outlines that do not meet the standardization standards includes: extracting the characters and their outlines in the character library that match the fuzzy characters that do not meet the standardization standards in the homework of each student, recording them as similar characters, and extracting the similarity between the fuzzy characters in the homework of each student and their similar characters; Identify the habitual writing features and habitual arrangement methods of each stroke of each student from the writing outlines of the answers to various questions in each student's homework; Extract the fixed-frame strokes corresponding to each fuzzy character in each student's homework, identify the writing features thereof, compare them with the habitual writing features of each stroke of the corresponding student, determine the sub-classified writing trend strokes corresponding to each fixed-frame stroke of each fuzzy character in each student's homework, reorganize and construct them, and obtain the sub-classified writing trend characters of each fuzzy character in each student's homework; Identify the arrangement modes of the fixed strokes corresponding to the fuzzy characters in the homework of each student, determine the supplementary writing trend strokes of the fixed strokes corresponding to the fuzzy characters in the homework of each student, and then reconstruct and obtain the supplementary writing trend Chinese characters of the fuzzy characters in the homework of each student; Similarity values ​​are assigned to the subcategorized Chinese characters with writing trends and the supplementary Chinese characters with writing trends, and the values ​​are added to the similarities between each fuzzy character and its similar characters in each student's assignment to obtain a final similarity value between each fuzzy character and its similar characters in each student's assignment. Then, the similar character to which the maximum final similarity value of each fuzzy character in each student's assignment belongs is used as the calibration text of the corresponding fuzzy character in the corresponding student's assignment.

5. The online intelligent review system based on big data according to claim 4 is characterized in that: The accuracy calibration of the font outlines that do not meet the standardization standards also includes: extracting letters and their outlines in the character library that match the fuzzy letters that do not meet the standardization standards in the homework of each student, and recording them as similar letters; According to the calibration text recognition method of each blurred text in each student's homework, the calibration letters of each blurred letter in each student's homework are determined similarly.

6. The online intelligent review system based on big data according to claim 1 is characterized in that: The process of listing the network-similar parts and their sources in the answers to the subjective questions of each student's homework is as follows: dividing the writing outlines of the corresponding answers to the subjective questions of each student's homework into text groups, obtaining the writing outlines of the corresponding text groups of the answers to the subjective questions of each student's homework, counting the standard answers to the subjective questions in the online public platform, integrating them into the text outlines of the standard answers to the subjective questions, matching them with the writing outlines of the corresponding text groups of the answers to the subjective questions of each student's homework, obtaining the text group outline parts whose similarity between the subjective questions of each student's homework and the text outlines of the standard answers to the corresponding questions is within a preset range, namely, the network-similar parts in the answers to the subjective questions of each student's homework, and marking the standard answer text outlines to which they belong as the sources of the corresponding network-similar parts.

7. The online intelligent review system based on big data according to claim 1 is characterized in that: The paradoxical index for evaluating the similar parts of the answers to the subjective questions of each student's homework is specifically as follows: obtaining reference answers to various questions submitted in advance by teachers in the review platform, scanning and obtaining the writing outlines of the reference answers to various questions, comparing them with the network similar parts of the answers to the subjective questions of each student's homework, identifying the similarity between the writing outlines of the corresponding answers to the network similar parts of the answers to the subjective questions of each student's homework and the writing outlines of the corresponding parts of the reference answers to the corresponding questions, and judging whether the network similar parts in the answers to the subjective questions of each student are similar and reasonable or citation weighted; The comprehensive citation weighted value of each student's answer to the subjective question is calculated, and its ratio to the preset benchmark citation weight is obtained to obtain the paradoxical index of the similar parts of the answers to the subjective questions of each student's homework.

8. The online intelligent review system based on big data according to claim 1 is characterized in that: The formula for determining the network citation score of the answers to the subjective questions of each student's homework is as follows: Determine the citation impact factor of each network-similar part of the answers to the subjective questions of each student's homework, and sum them to obtain the comprehensive citation impact factor U of the network-similar part of the answers to the subjective questions of each student's homework r , and then determine the network citation score of each student's answer to the subjective question of the homework Where δ r is the paradoxical index of the similarity of the answers to the subjective questions of the rth student's homework, It means round down, and e is a natural constant.

9. The online intelligent review system based on big data according to claim 6, characterized in that: The content of determining the interactive reference score of each student's homework is: comparing the writing outlines of the answers to the subjective questions of each student's homework and the answers to each text group with each other to identify each suspected student; Extract the historical homework submission records of each suspected student, identify the total number of handwriting erasures M′x of each suspected student in the current submitted homework, and obtain the total number of handwriting erasures M of each suspected student in the historical homework submission records xj , compare and evaluate the progress indicators of each student's work x , where x is the number of the suspected student, x = 1, 2, ..., c, j is the number of the number of historical homework submission records, j = 1, 2, ..., q, q is the number of historical homework submission records; Obtain the ease of the current homework submitted by the teacher in advance in the review platform, and analyze the abnormal evaluation coefficient of the homework submitted by each suspected student is the preset benchmark abnormal evaluation coefficient of the student's homework, and the abnormal evaluation coefficients of the homework submitted by all students except the suspected students are recorded as Get the abnormal evaluation coefficient of each student's submitted assignment or It is matched with the homework interaction reference scores corresponding to the preset ranges of various abnormal evaluation coefficients to determine the homework interaction reference scores of each student.

10. The online intelligent review system based on big data according to claim 7, characterized in that:The final score of each student's homework is determined as follows: after the review platform calibrates the accuracy of the writing outlines of the answers to various questions in each student's homework, it autonomously identifies the outline similarity μ between the answers to various questions in each student's homework and the reference answers. ri , i is the number of the homework type, i = 1, 2, ..., h; Determine the final grade for each student's assignment In the formula is the written score of the rth student's homework, and h is the number of homework topic types.

Images