CN116341511A - Automatic batch-change appraising method and device based on multiple recognition engines - Google Patents
Automatic batch-change appraising method and device based on multiple recognition engines Download PDFInfo
- Publication number
- CN116341511A CN116341511A CN202310258886.9A CN202310258886A CN116341511A CN 116341511 A CN116341511 A CN 116341511A CN 202310258886 A CN202310258886 A CN 202310258886A CN 116341511 A CN116341511 A CN 116341511A
- Authority
- CN
- China
- Prior art keywords
- answer
- text data
- result
- data
- recognition engine
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 22
- 238000012937 correction Methods 0.000 claims abstract description 48
- 238000006243 chemical reaction Methods 0.000 claims abstract description 18
- 239000004816 latex Substances 0.000 claims description 20
- 238000012545 processing Methods 0.000 claims description 9
- 238000004458 analytical method Methods 0.000 claims description 6
- 230000011218 segmentation Effects 0.000 claims description 6
- 230000014509 gene expression Effects 0.000 claims description 3
- 238000012549 training Methods 0.000 claims description 3
- 238000002790 cross-validation Methods 0.000 claims description 2
- 238000013077 scoring method Methods 0.000 claims 4
- 238000012986 modification Methods 0.000 description 7
- 238000004590 computer program Methods 0.000 description 5
- 230000004048 modification Effects 0.000 description 5
- 238000004422 calculation algorithm Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 230000003287 optical effect Effects 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 239000000203 mixture Substances 0.000 description 2
- 238000003058 natural language processing Methods 0.000 description 2
- 230000000644 propagated effect Effects 0.000 description 2
- 238000012795 verification Methods 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 239000000835 fiber Substances 0.000 description 1
- 229920000126 latex Polymers 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/10—Text processing
- G06F40/194—Calculation of difference between files
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/10—Text processing
- G06F40/12—Use of codes for handling textual entities
- G06F40/151—Transformation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/20—Education
- G06Q50/205—Education administration or guidance
-
- G—PHYSICS
- G09—EDUCATION; CRYPTOGRAPHY; DISPLAY; ADVERTISING; SEALS
- G09B—EDUCATIONAL OR DEMONSTRATION APPLIANCES; APPLIANCES FOR TEACHING, OR COMMUNICATING WITH, THE BLIND, DEAF OR MUTE; MODELS; PLANETARIA; GLOBES; MAPS; DIAGRAMS
- G09B7/00—Electrically-operated teaching apparatus or devices working with questions and answers
- G09B7/02—Electrically-operated teaching apparatus or devices working with questions and answers of the type wherein the student is expected to construct an answer to the question which is presented or wherein the machine gives an answer to the question presented by a student
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Educational Technology (AREA)
- Educational Administration (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Artificial Intelligence (AREA)
- Computational Linguistics (AREA)
- General Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- Economics (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- General Business, Economics & Management (AREA)
- Character Discrimination (AREA)
Abstract
The invention discloses an automatic batch-change appraising method and device based on multiple recognition engines, and relates to the field of intelligent operation, wherein the method comprises the steps of converting question bank answer data into answer text data, and determining multiple recognition engines to be called according to the answer text data obtained by conversion; obtaining answer result data, and identifying through a plurality of identified identification engines to obtain a plurality of answer result text data; and obtaining a correction result based on the comparison between the answer text data and the answer result text data and the comparison between the answer result text data. The invention can improve the identification accuracy, reduce misjudgment to the maximum extent and improve the reading efficiency of teachers.
Description
Technical Field
The invention relates to the field of intelligent operation, in particular to an automatic batch-modification appraising method and device based on a multi-recognition engine.
Background
In order to better master the learning condition of students, a teacher can regularly adopt modes of arranging homework, examination and the like to give questions, the students answer and then the teacher carries out batch-change judgment on the answer results of the students, so that the teacher can master the learning condition of the students. Meanwhile, in order to solve the workload of teachers in the manual judgment, reduce the load of the teachers and the technical gap of judging answering data input by non-traditional keyboards, an automatic correction technology has appeared at present, which can automatically realize correction judgment of answering results of students.
However, the conventional automatic correction technology is limited by the recognition engine and the answer format, the recognition accuracy of the recognition engine has a larger influence on the final correction judgment, in addition, the question bank answer format is more, and more problems exist in matching with the recognition result, so that the final correction result is not satisfactory, the misjudgment rate is higher, the teacher needs to manually modify correction error data, and extra workload is added to the teacher.
Disclosure of Invention
Aiming at the defects in the prior art, the invention aims to provide an automatic batch-modification appraising method and device based on a multi-recognition engine, which can improve recognition accuracy, reduce misappraising to the maximum extent and improve the reading efficiency of teachers.
In order to achieve the above object, the present invention provides an automatic batch-adaptation appraising method based on multiple recognition engines, which specifically comprises the following steps:
converting question bank answer data into answer text data, and determining a plurality of recognition engines to be called according to the answer text data obtained through conversion;
obtaining answer result data, and identifying through a plurality of identified identification engines to obtain a plurality of answer result text data;
and obtaining a correction result based on the comparison between the answer text data and the answer result text data and the comparison between the answer result text data.
On the basis of the technical scheme, the question bank answer data is converted into answer text data, and the method specifically comprises the following steps:
obtaining question bank answer data, removing html tags in the question bank answer data, and reserving the upper and lower marks and formula tags of the html;
converting html upper and lower labels and formula labels into standard LaTex format data;
judging whether MathML labels exist in question bank answer data or not:
if not, carrying out data standardization processing according to the LaTex format data obtained through conversion to obtain answer text data;
and if so, converting the MathML label into LaTex format data, and then carrying out data standardization processing according to the LaTex format data obtained by conversion to obtain answer text data.
On the basis of the technical scheme, the method for determining a plurality of recognition engines to be called according to the converted answer text data specifically comprises the following steps:
and determining a plurality of recognition engines to be invoked based on whether the answer is a number, whether the answer is a formula, whether the answer is Chinese and the discipline corresponding to the answer.
On the basis of the technical scheme, the recognition engine comprises a digital recognition engine, a formula recognition engine, an English recognition engine, a Chinese-English hybrid recognition engine and a Chinese formula hybrid recognition engine.
Based on the technical scheme, for a plurality of determined recognition engines to be called, the recognition engines are obtained through training of different data sets, provided by different manufacturers and capable of cross-validation.
Based on the technical scheme, the correction result is obtained based on the comparison between the answer text data and the answer result text data and the comparison between the answer result text data, and the specific steps comprise:
comparing the answer result text data with the answer text data, and judging whether the answer result text data is identical with the answer text data or not:
if yes, the correction result is correct;
if not, judging whether the text data of each answer result are the same, if so, the correction result is an answer error, and if not, the correction result is undetermined.
Based on the technical scheme, comparing the answer result text data with the answer text data:
if one answer result text data in the answer result text data is the same as the answer text data, or at least two answer result text data are the same as the answer text data, the correction result is correct;
if the answer text data is Chinese numbers, the correction result is correct when the answer result text data contains the corresponding digital answers;
if the answer text data is the Chinese character containing or, correcting the answer result to be correct when the answer result text data contains one or more Chinese characters corresponding to the answer result text data;
if the expressions between the answer result text data and the answer text data are different, but the actual meanings are the same, the correction result is correct;
if the answer text data is Chinese text, comparing the answer result text data with the answer text data by means of word segmentation marking, part-of-speech judgment or semantic similarity.
The invention provides an automatic correction appraising device based on a multi-recognition engine, which comprises:
the multi-recognition engine system is used for converting question bank answer data into answer text data, determining a plurality of recognition engines to be called according to the answer text data obtained through conversion, acquiring answer result data, and recognizing through the determined plurality of recognition engines to obtain a plurality of answer result text data;
and the comparison analysis engine system is used for obtaining the correction result based on the comparison between the answer text data and the answer result text data and the comparison between the answer result text data.
On the basis of the technical proposal, the method comprises the following steps,
the answer text data is plain text data or text data in LaTex format;
the answering result text data is plain text data or text data in LaTex format.
On the basis of the technical scheme, the recognition engine comprises a digital recognition engine, a formula recognition engine, an English recognition engine, a Chinese-English hybrid recognition engine and a Chinese formula hybrid recognition engine.
Compared with the prior art, the invention has the advantages that:
(1) The html labels are removed uniformly from the question bank answer data, if the text data is provided with a formula, the text data is converted into text data with a Latex formula, and the question bank answer data is converted and preprocessed, so that subsequent comparison with the text data of the answer result is facilitated, the text format is uniform, and the comparison accuracy is increased;
(2) Different recognition engines such as numbers, formulas, english, chinese-English mixture, chinese formula mixture and the like are adopted, so that a data set is relatively fixed, the most suitable recognition engine type is selected through judging question bank answer data, student answer data and subject, and accuracy of answer result data recognition is improved;
(3) And identifying the answer result data submitted by the students by determining a plurality of identification engines to be called to obtain a plurality of answer result text data, and then optimizing, comparing and analyzing the answer text data and the plurality of answer result text data to realize the final judgment of answer comparison, and improving the final accuracy, furthest reducing misjudgment and improving the reading efficiency of teachers through the cross verification of different identification engines.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are needed in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of an automatic batch appraising method based on multiple recognition engines according to an embodiment of the present invention;
FIG. 2 is a complete flow chart of an automatic batch appraising method based on multiple recognition engines.
Detailed Description
For the purposes of making the objects, technical solutions and advantages of the embodiments of the present application more clear, the technical solutions of the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is apparent that the described embodiments are some embodiments of the present application, but not all embodiments.
The embodiment of the invention provides an automatic approval judgment method based on multiple recognition engines, which is used for helping teachers to carry out intelligent recognition judgment after handwriting answers of students, after the handwriting or pictures of the students are obtained, the answers in a corresponding answer question library are subjected to format conversion to obtain answer text data, an optimal recognition engine is selected according to the answer text data, the handwriting or pictures are called for at least two different recognition engines to output multiple answer result text data, finally the answer text data and the answer result text data are analyzed and compared, correct, wrong and suspected judgment is carried out according to the modes of making rules, NLP (Natural Language Processing ) semantic understanding, function calculation and the like, and finally an approval result is output, so that the misjudgment is reduced, and the efficiency of the teachers is improved.
Referring to fig. 1, the automatic batch modification appraising method based on multiple recognition engines provided by the embodiment of the invention specifically includes the following steps:
s1: converting question bank answer data into answer text data, and determining a plurality of recognition engines to be called according to the answer text data obtained through conversion; the question bank answer data is the question answer corresponding to the answer result.
In the invention, question bank answer data are converted into answer text data, specifically:
s101: obtaining question bank answer data, removing html (hypertext markup language) tags in the question bank answer data, and reserving html upper and lower marks and formula tags;
s102: converting html upper and lower labels and formula labels into standard LaTex format data;
s103: judging whether MathML (mathematical markup language) labels exist in the question bank answer data or not:
if not, carrying out data standardization processing according to the LaTex format data obtained through conversion to obtain answer text data;
and if so, converting the MathML label into LaTex format data, and then carrying out data standardization processing according to the LaTex format data obtained by conversion to obtain answer text data.
In the invention, a plurality of recognition engines to be called are determined according to the answer text data obtained by conversion, specifically: and determining a plurality of recognition engines to be invoked based on whether the answer is a number, whether the answer is a formula, whether the answer is Chinese and the discipline corresponding to the answer. The recognition engine comprises a digital recognition engine, a formula recognition engine, an English recognition engine, a Chinese-English hybrid recognition engine and a Chinese formula hybrid recognition engine.
Namely, according to whether the answer is a number, whether the answer is a formula, whether the answer is Chinese and discipline corresponding to the answer, the recognition engine which is most matched with the answer result data is found, so that the answer result data can be better recognized.
S2: obtaining answer result data, and identifying through a plurality of identified identification engines to obtain a plurality of answer result text data; the answering result data is handwriting data or picture data submitted after the student answers.
It should be noted that, for a plurality of recognition engines to be called, the recognition engines are obtained through different data set training, provided by different manufacturers and can be cross-validated.
S3: and obtaining a correction result based on the comparison between the answer text data and the answer result text data and the comparison between the answer result text data. Through the cross verification of different recognition engines, the final accuracy is improved, misjudgment is reduced to the maximum extent, and the reading efficiency of teachers is improved.
In the invention, based on the comparison between the answer text data and the answer result text data and the comparison between the answer result text data, the correction result is obtained, and the specific steps comprise:
comparing the answer result text data with the answer text data, and judging whether the answer result text data is identical with the answer text data or not:
if yes, the correction result is correct;
if not, judging whether the text data of each answer result are the same, if so, the correction result is an answer error, and if not, the correction result is pending, namely, doubt is indicated, judgment is not carried out, and then, manual judgment processing is carried out.
In the invention, for comparison between answer result text data and answer text data, the following specific steps are as follows:
if one answer result text data in the answer result text data is the same as the answer text data, or at least two answer result text data are the same as the answer text data, the correction result is correct; namely, for a plurality of pieces of answer result text data, when one piece of answer result text data is the same as the answer text data, the correction result is judged to be correct in answer, if the judgment precision is required, when the plurality of pieces of answer result text data are the same as the answer text data, the correction result is judged to be correct in answer;
if the answer text data is Chinese numbers, the correction result is correct when the answer result text data contains the corresponding digital answers; if the answer text data is a Chinese digital answer, writing a corresponding digital answer in the answer result text data, and judging that the answer is correct by the modification result;
if the answer text data is the Chinese character containing or, correcting the answer result to be correct when the answer result text data contains one or more Chinese characters corresponding to the answer result text data;
if the expressions between the answer result text data and the answer text data are different, but the actual meanings are the same, the correction result is correct; that is, the text data of the answer result is not simplified, and the order is different from that of the answer text data, but the actual meaning is the same, and the answer can be judged to be correct, for example, the answer is 1/2, the answer result is 0.5, and the answer can also be judged to be correct;
if the answer text data is Chinese text, comparing the answer result text data with the answer text data by means of word segmentation marking, part-of-speech judgment or semantic similarity.
The word segmentation can be performed by adopting a CRF word segmentation algorithm, a neural network word segmentation algorithm and other algorithms, and the named entity recognition model can be BiLSTM (two-way long-short-term memory network) +CRF.
When the recognition result is processed, factors such as misrecognition by the recognition engine should be considered, for example, the answer is uppercase letter G, the answer is recognized as lowercase letter a, the answer is uppercase letter D, the answer is recognized as lowercase letter b or p, and at this time, the error judgment process should not be directly performed, but a manual judgment flow is added.
Meanwhile, when the text data of each answer result are different, the data with larger deviation from the standard answer can be processed by calculating the text similarity, if the similarity is low, the error judgment can be carried out, and the common similarity algorithm comprises the following steps: euclidean distance, pearson correlation coefficient, cosine similarity, tanimoto coefficient.
Furthermore, the accuracy of recognition modification can be improved by adjusting the judging rule or adding the recognition engine.
Referring to fig. 2, a complete flow of the automatic approval appraising method based on the multi-recognition engine of the present invention will be described.
A: the student submits the answering result data in handwriting or picture form, and goes to B:
b: converting the question bank answer data into answer text data, determining a plurality of recognition engines, and turning to C;
c: the recognition is carried out through the recognition engine, a plurality of answering result text data are obtained, and the process is transferred to D;
d: judging whether the answer result text data is the same as the answer text data, if so, turning to E, otherwise, turning to F;
e: the correction result is correct;
f: judging whether the text data of each answer result are the same or not, if so, turning to G, and if not, turning to H;
g: the correction result is a reply error;
h: the result of the correction is pending.
In a possible implementation manner, the embodiment of the present invention further provides a readable storage medium, where the readable storage medium is located in a PLC (Programmable Logic Controller ) controller, and a computer program is stored on the readable storage medium, where the program is executed by a processor to implement the following steps of the automatic batch modification and judgment method based on multiple recognition engines:
converting question bank answer data into answer text data, and determining a plurality of recognition engines to be called according to the answer text data obtained through conversion;
obtaining answer result data, and identifying through a plurality of identified identification engines to obtain a plurality of answer result text data;
and obtaining a correction result based on the comparison between the answer text data and the answer result text data and the comparison between the answer result text data.
The storage media may take the form of any combination of one or more computer-readable media. The computer readable medium may be a computer readable signal medium or a computer readable storage medium. The computer readable storage medium may be, for example, but not limited to: an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or a combination of any of the foregoing. More specific examples (a non-exhaustive list) of the computer-readable storage medium would include the following: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In this document, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device.
The computer readable signal medium may include a propagated data signal with computer readable program code embodied therein, either in baseband or as part of a carrier wave. Such a propagated data signal may take any of a variety of forms, including, but not limited to, electro-magnetic, optical, or any suitable combination of the foregoing. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, fiber optic cable, RF, etc., or any suitable combination of the foregoing.
Computer program code for carrying out operations of the present invention may be written in one or more programming languages, including an object oriented programming language such as Java, smalltalk, C ++ and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any kind of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or may be connected to an external computer (for example, through the Internet using an Internet service provider).
The embodiment of the invention provides an automatic correction appraising device based on a multi-recognition engine, which comprises a multi-recognition engine system and a comparison analysis engine system.
The multi-recognition engine system is used for converting the question bank answer data into answer text data, determining a plurality of recognition engines to be called according to the answer text data obtained by conversion, acquiring answer result data, and recognizing through the determined plurality of recognition engines to obtain a plurality of answer result text data; the comparison analysis engine system is used for obtaining the correction result based on the comparison between the answer text data and the answer result text data and the comparison between the answer result text data.
In the invention, answer text data is plain text data or text data in LaTex format; the answering result text data is plain text data or text data in LaTex format. The recognition engine comprises a digital recognition engine, a formula recognition engine, an English recognition engine, a Chinese-English hybrid recognition engine and a Chinese formula hybrid recognition engine.
The automatic correction appraising device is applied to an intelligent operation scene and is used for reducing daily read-and-write work of teachers, and comprises a multi-recognition engine system and a comparison analysis engine system; the multi-recognition engine system is used for finding out the most suitable multiple recognition engines to recognize answer result data according to the answer text data obtained through conversion; the comparison analysis engine system is used for comparing and analyzing the answer text data with the answer result text data, and correcting errors.
According to the automatic correction appraising device based on the multi-recognition engine, students submit answer result data by using an ink screen end or a tablet end, after receiving the data submitted by the students, the multi-recognition engine system sends the question bank answer data and the answer result data submitted by the students to the multi-recognition engine system, and the multi-recognition engine system performs format conversion on the question bank answer data to obtain answer text data and selects an optimal recognition engine according to the answer text data; and calling the answer result data at least twice by different recognition engines, outputting at least two pieces of answer result text data of plain texts or LaTex texts, finally analyzing and comparing the answer result text data obtained by recognition with the answer text data, judging correctness, mistakes and suspicion, and finally outputting a correction result.
The foregoing is merely a specific embodiment of the application to enable one skilled in the art to understand or practice the application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the application. Thus, the present application is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
Claims (10)
1. An automatic batch-adaptation appraising method based on a multi-recognition engine is characterized by comprising the following steps:
converting question bank answer data into answer text data, and determining a plurality of recognition engines to be called according to the answer text data obtained through conversion;
obtaining answer result data, and identifying through a plurality of identified identification engines to obtain a plurality of answer result text data;
and obtaining a correction result based on the comparison between the answer text data and the answer result text data and the comparison between the answer result text data.
2. The automatic approval scoring method based on multiple recognition engines of claim 1, wherein the converting question bank answer data into answer text data is specifically as follows:
obtaining question bank answer data, removing html tags in the question bank answer data, and reserving the upper and lower marks and formula tags of the html;
converting html upper and lower labels and formula labels into standard LaTex format data;
judging whether MathML labels exist in question bank answer data or not:
if not, carrying out data standardization processing according to the LaTex format data obtained through conversion to obtain answer text data;
and if so, converting the MathML label into LaTex format data, and then carrying out data standardization processing according to the LaTex format data obtained by conversion to obtain answer text data.
3. The automatic approval judgment method based on multiple recognition engines as set forth in claim 1, wherein the determining the multiple recognition engines to be invoked according to the converted answer text data includes:
and determining a plurality of recognition engines to be invoked based on whether the answer is a number, whether the answer is a formula, whether the answer is Chinese and the discipline corresponding to the answer.
4. The automatic approval scoring method based on a multi-recognition engine of claim 3, wherein: the recognition engine comprises a digital recognition engine, a formula recognition engine, an English recognition engine, a Chinese-English hybrid recognition engine and a Chinese formula hybrid recognition engine.
5. The automatic approval scoring method based on multiple recognition engines of claim 4, wherein: for a plurality of determined recognition engines to be called, the recognition engines are obtained through training of different data sets, provided by different manufacturers and capable of cross-validation.
6. The automatic approval and judgment method based on multiple recognition engines as set forth in claim 1, wherein the step of obtaining the correction result based on the comparison between the answer text data and the answer result text data and the comparison between the answer result text data comprises the steps of:
comparing the answer result text data with the answer text data, and judging whether the answer result text data is identical with the answer text data or not:
if yes, the correction result is correct;
if not, judging whether the text data of each answer result are the same, if so, the correction result is an answer error, and if not, the correction result is undetermined.
7. The automatic approval scoring method based on a multi-recognition engine of claim 6, wherein for the comparison between answer result text data and answer text data:
if one answer result text data in the answer result text data is the same as the answer text data, or at least two answer result text data are the same as the answer text data, the correction result is correct;
if the answer text data is Chinese numbers, the correction result is correct when the answer result text data contains the corresponding digital answers;
if the answer text data is the Chinese character containing or, correcting the answer result to be correct when the answer result text data contains one or more Chinese characters corresponding to the answer result text data;
if the expressions between the answer result text data and the answer text data are different, but the actual meanings are the same, the correction result is correct;
if the answer text data is Chinese text, comparing the answer result text data with the answer text data by means of word segmentation marking, part-of-speech judgment or semantic similarity.
8. An automatic correction appraising device based on a multi-recognition engine, comprising:
the multi-recognition engine system is used for converting question bank answer data into answer text data, determining a plurality of recognition engines to be called according to the answer text data obtained through conversion, acquiring answer result data, and recognizing through the determined plurality of recognition engines to obtain a plurality of answer result text data;
and the comparison analysis engine system is used for obtaining the correction result based on the comparison between the answer text data and the answer result text data and the comparison between the answer result text data.
9. The automatic correction and scoring device based on multiple recognition engines of claim 8, wherein:
the answer text data is plain text data or text data in LaTex format;
the answering result text data is plain text data or text data in LaTex format.
10. The automatic correction and scoring device based on multiple recognition engines of claim 8, wherein: the recognition engine comprises a digital recognition engine, a formula recognition engine, an English recognition engine, a Chinese-English hybrid recognition engine and a Chinese formula hybrid recognition engine.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310258886.9A CN116341511A (en) | 2023-03-14 | 2023-03-14 | Automatic batch-change appraising method and device based on multiple recognition engines |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310258886.9A CN116341511A (en) | 2023-03-14 | 2023-03-14 | Automatic batch-change appraising method and device based on multiple recognition engines |
Publications (1)
Publication Number | Publication Date |
---|---|
CN116341511A true CN116341511A (en) | 2023-06-27 |
Family
ID=86890910
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202310258886.9A Pending CN116341511A (en) | 2023-03-14 | 2023-03-14 | Automatic batch-change appraising method and device based on multiple recognition engines |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN116341511A (en) |
-
2023
- 2023-03-14 CN CN202310258886.9A patent/CN116341511A/en active Pending
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110110585B (en) | Intelligent paper reading implementation method and system based on deep learning and computer program | |
CN111597908A (en) | Test paper correcting method and test paper correcting device | |
CN116629275B (en) | Intelligent decision support system and method based on big data | |
CN110781672B (en) | Question bank production method and system based on machine intelligence | |
CN116861014B (en) | Image information extraction method and device based on pre-training language model | |
CN116543404A (en) | Table semantic information extraction method, system, equipment and medium based on cell coordinate optimization | |
CN112036184A (en) | Entity identification method, device, computer device and storage medium based on BilSTM network model and CRF model | |
CN112464927B (en) | Information extraction method, device and system | |
CN114913729A (en) | Question selection method and device, computer equipment and storage medium | |
CN112926345A (en) | Multi-feature fusion neural machine translation error detection method based on data enhancement training | |
CN111651994B (en) | Information extraction method and device, electronic equipment and storage medium | |
CN112613293A (en) | Abstract generation method and device, electronic equipment and storage medium | |
CN114416991A (en) | Method and system for analyzing text emotion reason based on prompt | |
CN113505786A (en) | Test question photographing and judging method and device and electronic equipment | |
CN115617961A (en) | Question answering method and device | |
CN113822040A (en) | Subjective question marking and scoring method and device, computer equipment and storage medium | |
CN113705207A (en) | Grammar error recognition method and device | |
CN112528628A (en) | Text processing method and device and electronic equipment | |
CN117235233A (en) | Automatic financial report question-answering method and device based on large model | |
CN116704508A (en) | Information processing method and device | |
CN116451646A (en) | Standard draft detection method, system, electronic equipment and storage medium | |
CN116341511A (en) | Automatic batch-change appraising method and device based on multiple recognition engines | |
Nguyen et al. | Handwriting recognition and automatic scoring for descriptive answers in Japanese language tests | |
KR20230013849A (en) | Character recognition method and system robust to errors of character recognition that recognize information included in tables | |
CN113868372A (en) | Statistical communique index extraction method based on rules and text sequence labeling |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination |