CN113343972A - Paper marking method and device, electronic equipment and storage medium - Google Patents
Paper marking method and device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN113343972A CN113343972A CN202110726513.0A CN202110726513A CN113343972A CN 113343972 A CN113343972 A CN 113343972A CN 202110726513 A CN202110726513 A CN 202110726513A CN 113343972 A CN113343972 A CN 113343972A
- Authority
- CN
- China
- Prior art keywords
- paper
- scoring
- content
- marking
- result
- 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 60
- 238000012360 testing method Methods 0.000 claims abstract description 109
- 238000012545 processing Methods 0.000 claims abstract description 101
- 238000004458 analytical method Methods 0.000 claims description 91
- 238000012549 training Methods 0.000 claims description 47
- 238000002372 labelling Methods 0.000 claims description 30
- 238000013077 scoring method Methods 0.000 claims description 18
- 230000004044 response Effects 0.000 claims description 14
- 230000011218 segmentation Effects 0.000 claims description 13
- 238000004590 computer program Methods 0.000 claims description 11
- 238000007689 inspection Methods 0.000 claims description 9
- 238000000605 extraction Methods 0.000 claims description 8
- 238000005211 surface analysis Methods 0.000 claims description 2
- 230000008569 process Effects 0.000 description 9
- 238000004364 calculation method Methods 0.000 description 4
- 230000008901 benefit Effects 0.000 description 3
- 238000004891 communication Methods 0.000 description 3
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 230000018109 developmental process Effects 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 238000011161 development Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000012015 optical character recognition Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 230000001413 cellular effect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000014509 gene expression Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/205—Parsing
- G06F40/211—Syntactic parsing, e.g. based on context-free grammar [CFG] or unification grammars
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/289—Phrasal analysis, e.g. finite state techniques or chunking
- G06F40/295—Named entity recognition
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/30—Semantic analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Abstract
The disclosure provides a paper marking method, a paper marking device, electronic equipment and a storage medium, wherein the method comprises the steps of obtaining a test paper image; identifying answering contents corresponding to the test paper questions in the test paper images; determining a target marking processing mode corresponding to the type of the test paper question based on the corresponding relation between the type of the test paper question and the marking processing mode; and adopting a target scoring processing mode to perform scoring processing on the answering content to obtain a scoring result. According to the method and the device, different marking processing modes can be selected for automatic marking based on different question types, the marking efficiency is high, the marking processing modes can be processing modes appointed for the question types, the marking uniformity of the same question type is improved to a certain extent, the marking result is more objective, the transverse comparison among different students is convenient to realize, and the examination has more reference significance.
Description
Technical Field
The disclosure relates to the technical field of multimedia teaching, in particular to a paper marking method, a paper marking device, electronic equipment and a storage medium.
Background
In a training, education, examination and other systems, in order to assess the learning condition of a student, the student is usually required to be examined, and the examination result is scored to know the mastery degree of the student on the learned knowledge.
The manual scoring mode is a commonly adopted scoring mode, particularly for subjective questions, students usually adopt a discussion mode to solve questions in examination questions, and multiple solution modes are usually provided, so that answers of the subjective questions are only used as references, and the scores can be scored by referring to scoring rules manually.
However, the efficiency of manual scoring is low, and the scoring standards of manual scoring are difficult to unify, so that it is difficult to assess the knowledge point mastering conditions of students under the unified standards.
Disclosure of Invention
The embodiment of the disclosure at least provides an examination paper marking method, an examination paper marking device, an electronic device and a storage medium, and improves examination paper marking efficiency under the condition of ensuring a unified examination paper marking standard.
In a first aspect, an embodiment of the present disclosure provides an examination paper scoring method, including:
acquiring a test paper image;
identifying answering contents corresponding to the test paper questions in the test paper images;
determining a target examination paper marking processing mode corresponding to the type of the examination paper question based on the corresponding relation between the type of the examination paper question and the examination paper marking processing mode;
and adopting the target scoring processing mode to perform scoring processing on the answering content to obtain a scoring result.
By adopting the paper marking method, under the condition that answer content corresponding to the test paper questions is identified from the test paper images, the target paper marking processing mode corresponding to the identified test paper question type can be determined based on the corresponding relation between the test paper question type and the paper marking processing mode, and then the answer content is subjected to paper marking processing based on the target paper marking processing mode to obtain a paper marking result. Therefore, different examination paper marking processing modes can be selected for automatic examination paper marking based on different question types, examination paper marking efficiency is high, the examination paper marking processing mode can be a processing mode appointed for the question types, examination paper marking uniformity of the same question type is improved to a certain extent, examination paper marking results are more objective, transverse comparison among different students is facilitated, and assessment is more meaningful.
In a possible embodiment, identifying the answering content of the corresponding test paper in the test paper image comprises:
performing paragraph identification on the test paper image, and dividing the test paper image into a plurality of paragraph areas;
for each paragraph region in the plurality of paragraph regions, identifying content attribute information corresponding to the paragraph region;
based on the obtained content attribute information, taking a plurality of paragraph areas corresponding to the same test paper question as an area set; the area set comprises a question stem area and a response area;
associating the answering areas corresponding to the same area set with the question stem areas, and respectively identifying the content of each paragraph area in the area set aiming at each area set to obtain the answering content of the answering area corresponding to the question stem content of the question stem areas.
Here, based on paragraph identification, a region set including the question stem region and the answer region may be implemented, and then the answer content of the answer region is determined, which further improves the scoring efficiency.
In one possible embodiment, the type of the test paper topic comprises an objective topic type; the method for performing the paper marking processing on the answering content by adopting the target paper marking processing mode to obtain a paper marking result comprises the following steps:
comparing the identified answering content with a first reference answer to obtain a comparison result;
in response to the comparison result indicating that the answering content matches the first reference answer, determining that the scoring result of the test paper question comprises correct answers;
and in response to the comparison result indicating that the answering content does not match the first reference answer, determining that the marking result of the test paper question comprises an answer error.
Here, for the objective question type, the examination paper marking result of the examination paper question can be determined based on the matching degree between the answering content and the first reference answer, the examination paper marking is automatically completed, and the examination paper marking efficiency is higher.
In one possible embodiment, the type of test paper title comprises a subjective title type; the method for performing the paper marking processing on the answering content by adopting the target paper marking processing mode to obtain a paper marking result comprises the following steps:
determining semantic similarity between the identified answering content and a second reference answer;
and determining the scoring result based on the semantic similarity.
Here, for the subjective question type, the scoring result can be determined based on the semantic similarity between the answering content and the second reference answer, and the scoring is automatically completed, so that the scoring efficiency is higher.
In one possible embodiment, the determining semantic similarity between the identified answering content and the second reference answer includes at least one of:
semantic feature extraction is respectively carried out on the identified answering content and the second reference answer to obtain a first semantic feature vector and a second semantic feature vector; determining a vector similarity between the first semantic feature vector and the second semantic feature vector; determining the vector similarity as the semantic similarity;
respectively carrying out entity extraction on the identified answering content and the second reference answer to obtain a first entity and a second entity; the second entity comprises answer keywords; determining a degree of match between the first entity and a second entity; determining the matching degree as the semantic similarity;
determining semantic similarity between the identified answering content and a second reference answer by using a trained semantic analysis network; the semantic analysis network is obtained by training similar sentences as training positive samples and other non-similar sentences as training negative samples, or by training two paired sentences and labeling similarity between the two sentences.
Here, the semantic similarity between the answer content and the second reference answer can be considered from multiple layers, which mainly considers that some key semantic information exists for the test paper question, and the key semantic information can be the semantic information of the whole content, the semantic information of a local sentence, and the semantic information of a local entity word, and is determined by multiple ways, so that the accuracy of the subsequent examination result can be improved to a certain extent.
In one possible embodiment, the method further comprises:
carrying out sentence smoothness inspection on the identified answering content by utilizing a trained sentence smoothness inspection network to obtain sentence smoothness; the sentence smoothness checking network is obtained by training a correct sentence as a training positive sample and an error sentence obtained by sequentially exchanging the correct sentence as a training negative sample;
the determining the scoring result based on the semantic similarity comprises:
and determining the scoring result based on the semantic similarity and the sentence passing degree.
Here, the scoring result can be determined by combining the semantic similarity and the sentence smoothness, so that the accuracy of the scoring result is further improved.
In one possible implementation, the type of the test paper title comprises a writing title type; the method for performing the paper marking processing on the answering content by adopting the target paper marking processing mode to obtain a paper marking result comprises the following steps:
identifying answering content corresponding to the test paper questions from the test paper area;
and performing at least one of paragraph analysis, sentence analysis and scroll analysis on the identified answering content to obtain the scoring result.
Here, for the writing question types, the examination paper analysis can be performed in a combination of a plurality of ways of improving the variety of analysis, so as to improve the accuracy of the examination paper reading result.
In a possible embodiment, performing paragraph analysis on the identified answer content to obtain the scoring result includes:
writing element analysis is carried out on each paragraph included in the identified answering content by utilizing a trained first paragraph analysis network, a first score is determined, and the first score condition is determined as the scoring result;
the first paragraph analysis network is obtained by training an image sample comprising article content and a labeling result of labeling preset writing elements for the article content.
In a possible embodiment, performing paragraph analysis on the identified answer content to obtain the scoring result includes:
performing central sentence analysis on each paragraph included in the identified answering content by using a trained second paragraph analysis network, determining a second score, and determining the second score condition as the scoring result;
the second paragraph analysis network is obtained by training a labeling result of a central sentence labeling on the basis of an image sample including article content and each paragraph included in the article content.
In a possible implementation manner, performing sentence analysis on the identified answering content to obtain the scoring result includes:
performing word segmentation processing on the identified answering content to obtain a word segmentation result;
and comparing a preset reference word and sentence aiming at the test paper question with the word segmentation result to obtain the paper marking result.
In a possible implementation manner, performing a volume analysis on the identified answering content to obtain the scoring result includes:
performing volume analysis on the identified answering content by using a volume analysis network, determining a third score, and determining the third score as the scoring result;
the scroll analysis network is obtained by training based on image samples with various writing styles and labeling results for performing scroll score labeling on the image samples.
In a possible implementation manner, the writing question type points to a target writing type in multiple preset writing types, and the target marking processing manner includes a target writing processing manner matched with the target writing type;
the method for performing the paper marking processing on the answering content by adopting the target paper marking processing mode to obtain a paper marking result comprises the following steps:
and processing the test paper questions by adopting the target writing processing mode to obtain the examination paper reading result.
In the method, different writing processing modes can be adopted for processing different writing types so as to provide a targeted scoring scheme, and the scoring accuracy is improved on the premise of ensuring the scoring objectivity.
In a second aspect, an embodiment of the present disclosure further provides an examination paper reading apparatus, including:
the acquisition module is used for acquiring a test paper image;
the identification module is used for identifying the answering content of the corresponding test paper in the test paper image;
the determining module is used for determining a target examination paper marking processing mode corresponding to the type of the examination paper question based on the corresponding relation between the type of the examination paper question and the examination paper marking processing mode;
and the scoring module is used for scoring the answering content by adopting the target scoring processing mode to obtain a scoring result.
In a third aspect, an embodiment of the present disclosure further provides an electronic device, including: a processor, a memory and a bus, the memory storing machine-readable instructions executable by the processor, the processor and the memory communicating via the bus when the electronic device is running, the machine-readable instructions when executed by the processor performing the steps of the scoring method according to the first aspect and any of its various embodiments.
In a fourth aspect, the disclosed embodiments also provide a computer-readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, performs the steps of the paper marking method according to the first aspect and any one of the various implementation manners thereof.
For the description of the effects of the above scoring device, the electronic device, and the computer-readable storage medium, reference is made to the description of the scoring method, which is not repeated herein.
In order to make the aforementioned objects, features and advantages of the present disclosure more comprehensible, preferred embodiments accompanied with figures are described in detail below.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present disclosure, the drawings required for use in the embodiments will be briefly described below, and the drawings herein incorporated in and forming a part of the specification illustrate embodiments consistent with the present disclosure and, together with the description, serve to explain the technical solutions of the present disclosure. It is appreciated that the following drawings depict only certain embodiments of the disclosure and are therefore not to be considered limiting of its scope, for those skilled in the art will be able to derive additional related drawings therefrom without the benefit of the inventive faculty.
FIG. 1 is a flow chart illustrating a scoring method provided by an embodiment of the present disclosure;
FIG. 2 shows a schematic view of an examination paper scoring device provided by an embodiment of the disclosure;
fig. 3 shows a schematic diagram of an electronic device provided by an embodiment of the present disclosure.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present disclosure more clear, the technical solutions of the embodiments of the present disclosure will be described clearly and completely with reference to the drawings in the embodiments of the present disclosure, and it is obvious that the described embodiments are only a part of the embodiments of the present disclosure, not all of the embodiments. The components of the embodiments of the present disclosure, generally described and illustrated in the figures herein, can be arranged and designed in a wide variety of different configurations. Thus, the following detailed description of the embodiments of the present disclosure, presented in the figures, is not intended to limit the scope of the claimed disclosure, but is merely representative of selected embodiments of the disclosure. All other embodiments, which can be derived by a person skilled in the art from the embodiments of the disclosure without making creative efforts, shall fall within the protection scope of the disclosure.
It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, it need not be further defined and explained in subsequent figures.
The term "and/or" herein merely describes an associative relationship, meaning that three relationships may exist, e.g., a and/or B, may mean: a exists alone, A and B exist simultaneously, and B exists alone. In addition, the term "at least one" herein means any one of a plurality or any combination of at least two of a plurality, for example, including at least one of A, B, C, and may mean including any one or more elements selected from the group consisting of A, B and C.
According to researches, the manual marking mode is a commonly adopted marking mode, particularly for subjective questions, students usually adopt a discussion mode to answer questions in the examination questions, and multiple answer modes often exist, so that answers of the subjective questions are only used as references, and the scores can be scored by manually referring to scoring rules.
However, the efficiency of manual scoring is low, and the scoring standards of manual scoring are difficult to unify, so that it is difficult to assess the knowledge point mastering conditions of students under the unified standards.
Based on the research, the present disclosure provides an examination paper marking method, an examination paper marking device, an electronic device, and a storage medium, which improve examination paper marking efficiency while ensuring a uniform standard for examination paper marking.
To facilitate understanding of the present embodiment, first, a detailed description is given to an examination paper reading method disclosed in an embodiment of the present disclosure, where an execution subject of the examination paper reading method provided in the embodiment of the present disclosure is generally an electronic device with a certain computing capability, and the electronic device includes, for example: a terminal device, which may be a User Equipment (UE), a mobile device, a User terminal, a cellular phone, a cordless phone, a Personal Digital Assistant (PDA), a handheld device, a computing device, a vehicle mounted device, a wearable device, or a server or other processing device. In some possible implementations, the scoring method may be implemented by a processor calling computer readable instructions stored in a memory.
Referring to fig. 1, which is a flowchart of an examination paper marking method provided in the embodiment of the present disclosure, the method includes steps S101 to S104, where:
s101: acquiring a test paper image;
s102: identifying answering contents corresponding to the test paper questions in the test paper images;
s103: determining a target marking processing mode corresponding to the type of the test paper question based on the corresponding relation between the type of the test paper question and the marking processing mode;
s104: and adopting a target scoring processing mode to perform scoring processing on the answering content to obtain a scoring result.
Here, in order to facilitate understanding of the scoring method provided by the embodiments of the present disclosure, an application scenario of the scoring method may be described in detail next. The paper marking method can be applied to any scene needing academic level examination, for example, daily examinations of primary and secondary school students, national unified examinations of ordinary college students of high school, and other scenes needing examination results, such as in-house tests and the like.
In the related art, a manual paper marking mode can be adopted, for example, for subjective questions, students usually adopt a discussion mode to solve problems in examination questions, multiple answer modes often exist, so that answers of the subjective questions are only used as references, and the students can score according to scoring rules, so that even if the answers correspond to the same reference answer, scores of different teachers for the same student are possibly different, the manual paper marking standard is difficult to unify, the students are difficult to assess the knowledge points of the students under the unified standard, and the manual paper marking mode is adopted, so that the efficiency is low, and the current requirements for performing various academic assessments on various personnel cannot be met.
In order to solve the above problem, the embodiment of the present disclosure provides an automatic paper marking processing scheme for different test paper item types based on image recognition, and improves paper marking efficiency under the condition of ensuring a unified standard for paper marking.
The test paper image may be obtained by scanning an answer sheet of a relevant person by using an image scanning technology, may be a scanned image of a Chinese test paper of a primary school student, or may be a scanned image of a mathematical test paper of a high-grade student, and is not particularly limited herein, and may be determined based on different application scenarios.
In practical applications, the content in one test paper image will usually contain various elements, such as titles, stems, etc. In order to facilitate the examination paper reading process, the answering content corresponding to the examination paper subject can be identified from the examination paper image, and the answering content can be the content handwritten by the examinee or the content printed by the examinee.
In the embodiment of the present disclosure, the answering content may be recognized from the test paper image based on Optical Character Recognition (OCR). In specific application, the identification of the response content of the corresponding test paper question can be realized by combining the binding relationship between the response area and the question stem area. The method can be realized by the following steps:
the method comprises the following steps of firstly, carrying out paragraph identification on a test paper image, and dividing the test paper image into a plurality of paragraph areas;
identifying content attribute information corresponding to the paragraph areas aiming at each paragraph area in the plurality of paragraph areas;
taking a plurality of paragraph areas corresponding to the same test paper question as an area set based on the obtained content attribute information; the region set comprises a question stem region and a response region;
and step four, associating the answering areas corresponding to the same area set with the question stem areas, and identifying the content of each paragraph area in the area set aiming at each area set to obtain the answering content of the answering area corresponding to the question stem content of the question stem areas.
A paragraph region can point to a minimum recognition unit of the test paper content, and the position range of each paragraph region in the test paper image can be determined by combining the test paper template. For example, a first-level caption blank filling question of the test paper may be used as a paragraph region, or a specific blank filling question in a second-level caption may be used as a paragraph region.
For different paragraph regions, the content attribute information corresponding to the paragraph region may be the same or different, for example, for a first-level title gap filling question and a first-level title selection question, the content attribute information may be titles, and for two specific gap filling questions of the first-level title gap filling question, the content attribute information may be titles.
Here, the area set may be performed based on the content attribute information, and the answer area corresponding to the same area set may be associated with the question stem area, that is, the same area set includes both the paragraph area corresponding to the question stem and the paragraph area corresponding to the answer. In this way, the content of each paragraph area in the area set is identified to obtain the answering content of the answering area corresponding to the question stem content of the question stem area. Therefore, based on the binding relationship between the answer area and the question stem area, the identified answer content points to the corresponding test paper question, and the accuracy of the paper marking result is improved on the premise of ensuring the paper marking efficiency.
In order to further improve the efficiency of paper marking, different paper marking processing modes can be set based on different types of test paper questions, so that under the condition of determining answer contents of the test paper questions, automatic paper marking processing can be performed according to the corresponding paper marking processing modes to improve the efficiency of paper marking, and the corresponding paper marking processing modes are set for different types of questions, so that the examination standards for uniform types of questions are unified by the paper marking processing modes, and examination results have reference significance.
The types of the test paper questions mainly include objective question types, subjective question types, writing question types and the like, and can also include other question types. For objective question types, the matching degree between the answering content and the reference answer can be used to determine the scoring result; for subjective item types, semantic similarity comparison can be adopted to determine the scoring result; for the writing question types, analysis methods including paragraph analysis, sentence analysis and scroll analysis can be used to determine the scoring result, and the three question types can be described separately.
In a first aspect: in the case where the identified type of the test paper topic includes an objective topic type, the paper marking process may be performed as follows:
step one, comparing the identified answering content with a first reference answer to obtain a comparison result;
step two, responding to the comparison result to indicate that the answering content is matched with the first reference answer, and determining that the examination paper marking result of the examination paper question comprises correct answers; and in response to the comparison result indicating that the answering content does not match the first reference answer, determining that the marking result of the test paper question comprises an answer error.
The first reference answer may be preset for the answer content of the objective question. By comparing the answering content with the first reference answer, whether the examination result of the examination question is a correct answer or a wrong answer can be determined.
In a specific application, the determination result of the matching degree between the answering content and the first reference answer may be determined in combination with the scoring rule. Here, the matching may be a narrow or broad concept of matching. In one implementation, a match is determined only if there is absolute agreement, e.g., for a radio topic, the answer choice is exactly the same as the first reference answer and the answer is determined to be correct. In another implementation, a match may be determined in the case of partially identical or similar expressions (i.e., identical interpretations), e.g., for partially filled-in questions, there may also be alternative reply content, at least two of which may be considered to be matching; for another example, for a multiple choice question, a few-choice score may be involved in multiple choice, and a wrong choice may not be involved in multiple choice, and then the few-choice score may be regarded as partial matching, and may be determined by combining different application requirements, which is not described herein again.
In a second aspect: in the case where the identified type of the test paper topic includes a subjective topic type, the paper marking process may be performed as follows:
step one, determining semantic similarity between the identified answering content and a second reference answer;
and step two, determining an examination paper marking result based on the semantic similarity.
The semantic similarity may be global semantic similarity, that is, the answer content is taken as a whole to be subjected to semantic similarity calculation with the second reference answer, or local semantic similarity, that is, a key entity is extracted from the answer content, and the key entity is subjected to semantic similarity calculation with the reference entity in the second reference answer, or a determination mode combining the two semantic similarities.
In the embodiment of the present disclosure, the global semantic similarity may be determined according to two ways, one of which may be to first perform semantic feature extraction on the identified answering content and the second reference answer, respectively, to obtain a first semantic feature vector and a second semantic feature vector, and further determine the vector similarity between the first semantic feature vector and the second semantic feature vector, and determine the vector similarity as the semantic similarity.
The semantic feature vector may be obtained by analyzing the response content and the second reference answer based on a semantic dimension, and the corresponding vector similarity may be determined based on a calculation method of cosine similarity and the like.
Secondly, the trained semantic analysis network can be used for determining the semantic similarity between the identified answering content and the second reference answer.
The semantic analysis network may be obtained by training similar sentences as training positive samples and other non-similar sentences as training negative samples, for example, 10 persons describe the same sentence in different ways to obtain similar sentences, and in the training stage, it can be known that the semantic analysis network maximizes the similarity score obtained by the similar sentences and minimizes the non-similar sentences, and in specific applications, the semantic analysis network can be implemented by using two classifications.
In addition, the semantic analysis network can be obtained by training two sentences which are paired and the labeling similarity between the two sentences. Before training, the relevance between two sentences can be judged manually, and then the score of labeling similarity is determined. And training the corresponding relation between the input two sentences and the scores to obtain the network parameter values of the semantic analysis network, and further guiding the calculation of the semantic similarity between the answering content and the second reference answer.
In the embodiment of the present disclosure, the local semantic similarity may be determined by performing entity extraction on the identified answering content and the second reference answer to obtain a first entity and a second entity, where the second entity includes an answering keyword, and then determining a matching degree between the first entity and the second entity, and determining the matching degree as the semantic similarity.
The matching degree between the two entities (i.e. the first entity and the second entity) may be determined by vector similarity after vector representation is performed on the two entities, which is not described herein again.
It should be noted that, for the three semantic similarity determination methods mentioned above, in a specific application, the semantic similarity between the answering content and the second reference answer may be determined by any one determination method, or two or all of the three determination methods may be considered in combination, for example, a global determination method and a local determination method may be combined, and then the final semantic similarity may be an overall result obtained by weighting the results of the two determination methods differently.
In order to cover the answer mode of the subjective question type as much as possible so as to further improve the accuracy of the paper marking, a trained sentence smoothness inspection network can be used for performing sentence smoothness inspection on the identified answering content to obtain sentence smoothness, and then the paper marking result is determined by combining the semantic similarity and the sentence smoothness.
In a specific application, the more smooth the answering content is, and the higher the semantic similarity between the answering content and the second reference answer is, the more accurate the answer can be described to a certain extent, and the corresponding subjective question score is also higher.
The sentence smoothness checking network can be obtained by training a correct sentence as a training positive sample and an error sentence obtained by sequentially exchanging the correct sentence as a training negative sample. In specific application, network training can be realized based on two classifications, that is, correct sentences in correct sequence are marked as 1, wrong sentences obtained by exchanging the sequence of the correct sentences are marked as 0, and then training of a network is realized by checking the smoothness of the sentences.
In a third aspect: in the case that the identified type of the test paper title includes the writing title type, the paper marking process can be performed according to the following steps:
step one, identifying answering contents corresponding to test paper questions from a test paper area;
and step two, performing at least one of paragraph analysis, sentence analysis and scroll analysis on the identified answering content to obtain a scoring result.
For writing questions, the scoring method provided by the embodiment of the disclosure can analyze answering content from multiple analysis dimensions, such as paragraph analysis, sentence analysis, and scroll analysis, to determine scoring results. The following can be described separately with respect to the above three analysis modes.
In a first aspect: embodiments of the present disclosure may determine paragraph scores through paragraph analysis. The paragraph analysis here may be written element analysis embodied by all paragraphs, or may be central sentence analysis embodied by individual paragraphs.
In the case of performing the writing element analysis, the first paragraph analysis network may be used to perform the writing element analysis on each paragraph included in the identified answering content, determine a first score, and use the score as a scoring result.
The first paragraph analysis network may be obtained by training based on an image sample including article content and a labeling result obtained by labeling preset writing elements for the article content. The preset writing elements may be elements including arguments, comments, and demonstrations. By labeling the writing elements in advance for each paragraph, the article characteristics of the article content with the preset writing elements can be obtained through training, and the article characteristics of the article content without the preset writing elements can also be obtained through training.
In the case of performing the writing element analysis, the second paragraph analysis network may be used to perform a clause analysis on each paragraph included in the identified answering content, and determine a second score as a result of scoring.
The second paragraph analysis network can be trained based on the image sample including the article content and the labeling result of the central sentence labeling for each paragraph included in the article content. Through the marking of the central sentence, whether the central sentence can be selected from the identified answering contents can be judged, and whether the description contents of the central sentence are related to the writing subject can be determined so as to determine the final score.
In a second aspect: the embodiment of the disclosure can determine the scoring result through statement analysis.
Here, the word segmentation processing may be performed on the identified answering content to obtain a word segmentation result, and then the reference word sentence preset for the question of the test paper is compared with the word segmentation result to obtain an examination paper reading result.
For example, an answer sentence may be preset for the question of the test paper, and when a plurality of wrongly written characters appear in the answer content, the answer result may be determined to be closer to a low score, whereas when no wrongly written characters or fewer wrongly written characters appear in the answer content, the answer result may be determined to be closer to a high score.
For another example, a good word sentence can be preset for the test paper question, the obtained word segmentation result is compared with the preset good word sentence, and the answer result is closer to a high score under the condition of higher relevance, and conversely, the answer result is closer to a low score under the condition of lower relevance.
In a third aspect: the embodiment of the disclosure can determine the scoring result through the scroll analysis.
Here, the identified answering content may be subjected to a volume analysis using a volume analysis network, and a third score may be determined as a result of scoring.
The scroll analysis network can be obtained by training based on image samples with various writing styles and labeling results for performing scroll score labeling on the image samples. That is, the above-mentioned scroll analysis network can learn out different writing styles in advance and correspond the corresponding relation between the scroll score, for example, can correspond high scroll score to the examination paper of scroll work, can correspond low scroll score to the examination paper of the sloppy face, and then can confirm the score of the answer content that is currently discerned according to this scroll analysis network.
Considering that the types of writing questions are rich and diverse, different writing processing modes can be adopted for different writing types, and here, under the condition that the writing questions are identified, the test paper questions can be processed based on the corresponding target writing processing mode to obtain the reading result.
The writing types may include a narrative, a lyric, a treatise, a description, an application, etc. For example, for a narrative, the development process experienced by the narrative subject can be determined, and the development process can be extracted in a manner of reverse narrative, narrative and the like to realize the writing process.
It will be understood by those skilled in the art that in the method of the present invention, the order of writing the steps does not imply a strict order of execution and any limitations on the implementation, and the specific order of execution of the steps should be determined by their function and possible inherent logic.
Based on the same inventive concept, the embodiment of the present disclosure further provides a paper marking device corresponding to the paper marking method, and as the principle of solving the problem of the device in the embodiment of the present disclosure is similar to that of the paper marking method in the embodiment of the present disclosure, the implementation of the device can refer to the implementation of the method, and repeated details are not repeated.
Referring to fig. 2, a schematic diagram of an examination paper reading apparatus provided in an embodiment of the present disclosure is shown, where the apparatus includes: an acquisition module 201, an identification module 202, a determination module 203 and an examination paper marking module 204; wherein the content of the first and second substances,
an obtaining module 201, configured to obtain a test paper image;
the identification module 202 is used for identifying answering contents corresponding to the test paper questions in the test paper images;
the determining module 203 is configured to determine a target examination paper reading processing mode corresponding to the type of the examination paper question based on a corresponding relationship between the type of the examination paper question and the examination paper reading processing mode;
and the scoring module 204 is configured to perform scoring processing on the answering content by using a target scoring processing mode to obtain a scoring result.
With the paper marking device, when the answering content corresponding to the test paper questions is identified from the test paper images, the target paper marking processing mode corresponding to the identified test paper question type can be determined based on the corresponding relation between the test paper question type and the paper marking processing mode, and then the answering content is subjected to paper marking processing based on the target paper marking processing mode to obtain the paper marking result. Therefore, different examination paper marking processing modes can be selected for automatic examination paper marking based on different question types, examination paper marking efficiency is high, the examination paper marking processing mode can be a processing mode appointed for the question types, examination paper marking uniformity of the same question type is improved to a certain extent, examination paper marking results are more objective, transverse comparison among different students is facilitated, and assessment is more meaningful.
In one possible implementation, the identifying module 202 is configured to identify the answering content of the corresponding test paper topic in the test paper image according to the following steps:
performing paragraph identification on the test paper image, and dividing the test paper image into a plurality of paragraph areas;
identifying content attribute information corresponding to the paragraph areas for each of the plurality of paragraph areas;
based on the obtained content attribute information, taking a plurality of paragraph areas corresponding to the same test paper question as an area set; the region set comprises a question stem region and a response region;
and associating the answering areas corresponding to the same area set with the question stem areas, and respectively identifying the content of each paragraph area in the area set aiming at each area set to obtain the answering content of the answering area corresponding to the question stem content of the question stem areas.
Here, based on paragraph identification, a region set including the question stem region and the answer region may be implemented, and then the answer content of the answer region is determined, which further improves the scoring efficiency.
In one possible embodiment, the type of test paper topic comprises an objective topic type; the scoring module 204 is configured to perform scoring processing on the answering content in a target scoring processing manner according to the following steps to obtain a scoring result:
comparing the identified answering content with the first reference answer to obtain a comparison result;
in response to the comparison result indicating that the answering content is matched with the first reference answer, determining that the marking result of the test paper question comprises correct answers;
and in response to the comparison result indicating that the answering content does not match the first reference answer, determining that the marking result of the test paper question comprises an answer error.
In one possible embodiment, the types of test paper questions include a subjective question type; the scoring module 204 is configured to perform scoring processing on the answering content in a target scoring processing manner according to the following steps to obtain a scoring result:
determining semantic similarity between the identified answering content and the second reference answer;
and determining an examination paper marking result based on the semantic similarity.
In one possible embodiment, the scoring module 204 is configured to determine the semantic similarity between the identified answering content and the second reference answer according to the following steps, including at least one of the following:
semantic feature extraction is respectively carried out on the identified answering content and the second reference answer to obtain a first semantic feature vector and a second semantic feature vector; determining a vector similarity between the first semantic feature vector and the second semantic feature vector; determining the vector similarity as semantic similarity;
respectively carrying out entity extraction on the identified answering content and the second reference answer to obtain a first entity and a second entity; the second entity comprises answer keywords; determining a matching degree between the first entity and the second entity; determining the matching degree as semantic similarity;
determining semantic similarity between the identified answering content and a second reference answer by using the trained semantic analysis network; the semantic analysis network is obtained by training similar sentences as training positive samples and other non-similar sentences as training negative samples, or by training two paired sentences and labeling similarity between the two sentences.
In a possible implementation manner, the scoring module 204 is configured to determine a scoring result based on the semantic similarity according to the following steps:
carrying out sentence smoothness inspection on the identified answering content by utilizing the trained sentence smoothness inspection network to obtain sentence smoothness; the sentence smoothness inspection network is obtained by training a correct sentence as a training positive sample and an error sentence obtained by sequentially exchanging the correct sentence as a training negative sample;
and determining the scoring result based on the semantic similarity and the sentence smoothness.
In one possible embodiment, the types of test paper titles include a writing title type; the scoring module 204 is configured to perform scoring processing on the answering content in a target scoring processing manner according to the following steps to obtain a scoring result:
identifying answering contents corresponding to test paper questions from the test paper area;
and performing at least one of paragraph analysis, sentence analysis and scroll analysis on the identified answering content to obtain a scoring result.
In a possible implementation manner, the scoring module 204 is configured to perform paragraph analysis on the identified answer content according to the following steps to obtain a scoring result:
writing element analysis is carried out on each paragraph included in the identified answering content by utilizing the trained first paragraph analysis network, a first score is determined, and the first score condition is determined as a paper marking result;
the first paragraph analysis network is obtained by training an image sample comprising article content and a labeling result of labeling preset writing elements for the article content.
In a possible implementation manner, the scoring module 204 is configured to perform paragraph analysis on the identified answer content according to the following steps to obtain a scoring result:
performing central sentence analysis on each paragraph included in the identified answering content by using a trained second paragraph analysis network, determining a second score, and determining the second score condition as a paper marking result;
the second paragraph analysis network is obtained by training a labeling result of central sentence labeling on the basis of an image sample including article content and each paragraph included in the article content.
In a possible implementation manner, the scoring module 204 is configured to perform sentence analysis on the identified answering content according to the following steps to obtain a scoring result:
performing word segmentation processing on the identified answering content to obtain a word segmentation result;
and comparing the preset reference words and sentences aiming at the examination question with the word segmentation result to obtain an examination result.
In a possible implementation manner, the scoring module 204 is configured to perform a surface-to-surface analysis on the identified answering content to obtain a scoring result, and includes:
performing volume analysis on the identified answering content by using a volume analysis network, determining a third score, and determining the third score as a result of scoring;
the scroll analysis network is obtained by training based on image samples with various writing styles and labeling results for performing scroll score labeling on the image samples.
In one possible implementation mode, the writing question type points to a target writing type in multiple preset writing types, and the target marking processing mode comprises a target writing processing mode matched with the target writing type;
the scoring module 204 is configured to perform scoring processing on the answering content in a target scoring processing manner according to the following steps to obtain a scoring result, and includes:
and processing the test paper questions by adopting a target writing processing mode to obtain an examination paper reading result.
The description of the processing flow of each module in the device and the interaction flow between the modules may refer to the related description in the above method embodiments, and will not be described in detail here.
An embodiment of the present disclosure further provides an electronic device, as shown in fig. 3, which is a schematic structural diagram of the electronic device provided in the embodiment of the present disclosure, and the electronic device includes: a processor 301, a memory 302, and a bus 303. The memory 302 stores machine-readable instructions executable by the processor 301 (for example, corresponding execution instructions of the acquisition module 201, the identification module 202, the determination module 203, the scoring module 204, and the like in the apparatus in fig. 2), when the electronic device is operated, the processor 301 and the memory 302 communicate via the bus 303, and when the processor 301 executes the following processing:
acquiring a test paper image;
identifying answering contents corresponding to the test paper questions in the test paper images;
determining a target marking processing mode corresponding to the type of the test paper question based on the corresponding relation between the type of the test paper question and the marking processing mode;
and adopting a target scoring processing mode to perform scoring processing on the answering content to obtain a scoring result.
The embodiment of the present disclosure further provides a computer-readable storage medium, where a computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, the computer program performs the steps of the paper marking method described in the above method embodiment. The storage medium may be a volatile or non-volatile computer-readable storage medium.
The embodiments of the present disclosure also provide a computer program product, where the computer program product carries a program code, and instructions included in the program code may be used to execute the steps of the scoring method described in the foregoing method embodiments, which may be referred to specifically in the foregoing method embodiments, and are not described herein again.
The computer program product may be implemented by hardware, software or a combination thereof. In an alternative embodiment, the computer program product is embodied in a computer storage medium, and in another alternative embodiment, the computer program product is embodied in a Software product, such as a Software Development Kit (SDK), or the like.
It is clear to those skilled in the art that, for convenience and brevity of description, the specific working processes of the system and the apparatus described above may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again. In the several embodiments provided in the present disclosure, it should be understood that the disclosed system, apparatus, and method may be implemented in other ways. The above-described embodiments of the apparatus are merely illustrative, and for example, the division of the units is only one logical division, and there may be other divisions when actually implemented, and for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection of devices or units through some communication interfaces, and may be in an electrical, mechanical or other form.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, functional units in the embodiments of the present disclosure may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit.
The functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a non-volatile computer-readable storage medium executable by a processor. Based on such understanding, the technical solution of the present disclosure may be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for causing an electronic device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present disclosure. And the aforementioned storage medium includes: various media capable of storing program codes, such as a usb disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk.
Finally, it should be noted that: the above-mentioned embodiments are merely specific embodiments of the present disclosure, which are used for illustrating the technical solutions of the present disclosure and not for limiting the same, and the scope of the present disclosure is not limited thereto, and although the present disclosure is described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: any person skilled in the art can modify or easily conceive of the technical solutions described in the foregoing embodiments or equivalent technical features thereof within the technical scope of the present disclosure; such modifications, changes or substitutions do not depart from the spirit and scope of the embodiments of the present disclosure, and should be construed as being included therein. Therefore, the protection scope of the present disclosure shall be subject to the protection scope of the claims.
Claims (15)
1. An examination paper marking method, comprising:
acquiring a test paper image;
identifying answering contents corresponding to the test paper questions in the test paper images;
determining a target examination paper marking processing mode corresponding to the type of the examination paper question based on the corresponding relation between the type of the examination paper question and the examination paper marking processing mode;
and adopting the target scoring processing mode to perform scoring processing on the answering content to obtain a scoring result.
2. The paper marking method as claimed in claim 1, wherein the step of identifying the answering content corresponding to the test paper in the test paper image comprises the following steps:
performing paragraph identification on the test paper image, and dividing the test paper image into a plurality of paragraph areas;
for each paragraph region in the plurality of paragraph regions, identifying content attribute information corresponding to the paragraph region;
based on the obtained content attribute information, taking a plurality of paragraph areas corresponding to the same test paper question as an area set; the area set comprises a question stem area and a response area;
associating the answering areas corresponding to the same area set with the question stem areas, and respectively identifying the content of each paragraph area in the area set aiming at each area set to obtain the answering content of the answering area corresponding to the question stem content of the question stem areas.
3. The paper marking method according to claim 1 or 2, wherein the types of test paper questions comprise objective question types; the method for performing the paper marking processing on the answering content by adopting the target paper marking processing mode to obtain a paper marking result comprises the following steps:
comparing the identified answering content with a first reference answer to obtain a comparison result;
in response to the comparison result indicating that the answering content matches the first reference answer, determining that the scoring result of the test paper question comprises correct answers;
and in response to the comparison result indicating that the answering content does not match the first reference answer, determining that the marking result of the test paper question comprises an answer error.
4. The paper marking method according to claim 1 or 2, wherein the types of test paper questions comprise subjective question types; the method for performing the paper marking processing on the answering content by adopting the target paper marking processing mode to obtain a paper marking result comprises the following steps:
determining semantic similarity between the identified answering content and a second reference answer;
and determining the scoring result based on the semantic similarity.
5. The scoring method according to claim 4, wherein the determining semantic similarity between the identified answering content and the second reference answer comprises at least one of:
semantic feature extraction is respectively carried out on the identified answering content and the second reference answer to obtain a first semantic feature vector and a second semantic feature vector; determining a vector similarity between the first semantic feature vector and the second semantic feature vector; determining the vector similarity as the semantic similarity;
respectively carrying out entity extraction on the identified answering content and the second reference answer to obtain a first entity and a second entity; the second entity comprises answer keywords; determining a degree of match between the first entity and a second entity; determining the matching degree as the semantic similarity;
determining semantic similarity between the identified answering content and a second reference answer by using a trained semantic analysis network; the semantic analysis network is obtained by training similar sentences as training positive samples and other non-similar sentences as training negative samples, or by training two paired sentences and labeling similarity between the two sentences.
6. The scoring method according to claim 4 or 5, characterized in that the method further comprises:
carrying out sentence smoothness inspection on the identified answering content by utilizing a trained sentence smoothness inspection network to obtain sentence smoothness; the sentence smoothness checking network is obtained by training a correct sentence as a training positive sample and an error sentence obtained by sequentially exchanging the correct sentence as a training negative sample;
the determining the scoring result based on the semantic similarity comprises:
and determining the scoring result based on the semantic similarity and the sentence passing degree.
7. The paper marking method according to claim 1 or 2, wherein the type of the test paper topic comprises a writing topic type; the method for performing the paper marking processing on the answering content by adopting the target paper marking processing mode to obtain a paper marking result comprises the following steps:
identifying answering content corresponding to the test paper questions from the test paper area;
and performing at least one of paragraph analysis, sentence analysis and scroll analysis on the identified answering content to obtain the scoring result.
8. The scoring method as claimed in claim 7, wherein performing paragraph analysis on the identified answer content to obtain the scoring result comprises:
writing element analysis is carried out on each paragraph included in the identified answering content by utilizing a trained first paragraph analysis network, a first score is determined, and the first score condition is determined as the scoring result;
the first paragraph analysis network is obtained by training an image sample comprising article content and a labeling result of labeling preset writing elements for the article content.
9. The scoring method according to claim 7 or 8, wherein performing paragraph analysis on the identified answer content to obtain the scoring result comprises:
performing central sentence analysis on each paragraph included in the identified answering content by using a trained second paragraph analysis network, determining a second score, and determining the second score condition as the scoring result;
the second paragraph analysis network is obtained by training a labeling result of a central sentence labeling on the basis of an image sample including article content and each paragraph included in the article content.
10. The scoring method according to any one of claims 7-9, wherein performing sentence analysis on the identified answering content to obtain the scoring result comprises:
performing word segmentation processing on the identified answering content to obtain a word segmentation result;
and comparing a preset reference word and sentence aiming at the test paper question with the word segmentation result to obtain the paper marking result.
11. The scoring method according to any one of claims 7-10, wherein performing a surface analysis on the identified answer content to obtain the scoring result comprises:
performing volume analysis on the identified answering content by using a volume analysis network, determining a third score, and determining the third score as the scoring result;
the scroll analysis network is obtained by training based on image samples with various writing styles and labeling results for performing scroll score labeling on the image samples.
12. The scoring method according to claim 7, wherein the writing question type points to a target writing type of a plurality of preset writing types, and the target writing processing manner includes a target writing processing manner matched with the target writing type;
the method for performing the paper marking processing on the answering content by adopting the target paper marking processing mode to obtain a paper marking result comprises the following steps:
and processing the test paper questions by adopting the target writing processing mode to obtain the examination paper reading result.
13. An examination paper reading device, comprising:
the acquisition module is used for acquiring a test paper image;
the identification module is used for identifying the answering content of the corresponding test paper in the test paper image;
the determining module is used for determining a target examination paper marking processing mode corresponding to the type of the examination paper question based on the corresponding relation between the type of the examination paper question and the examination paper marking processing mode;
and the scoring module is used for scoring the answering content by adopting the target scoring processing mode to obtain a scoring result.
14. An electronic device, comprising: a processor, a memory and a bus, the memory storing machine-readable instructions executable by the processor, the processor and the memory communicating over the bus when the electronic device is operating, the machine-readable instructions when executed by the processor performing the steps of the scoring method of any one of claims 1 to 12.
15. A computer-readable storage medium, in which a computer program is stored which, when being executed by a processor, carries out the steps of the scoring method according to any one of claims 1 to 12.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110726513.0A CN113343972A (en) | 2021-06-29 | 2021-06-29 | Paper marking method and device, electronic equipment and storage medium |
| PCT/CN2022/089787 WO2023273583A1 (en) | 2021-06-29 | 2022-04-28 | Exam-marking method and apparatus, electronic device, and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110726513.0A CN113343972A (en) | 2021-06-29 | 2021-06-29 | Paper marking method and device, electronic equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113343972A true CN113343972A (en) | 2021-09-03 |
Family
ID=77481420
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110726513.0A Pending CN113343972A (en) | 2021-06-29 | 2021-06-29 | Paper marking method and device, electronic equipment and storage medium |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN113343972A (en) |
| WO (1) | WO2023273583A1 (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113780997A (en) * | 2021-09-16 | 2021-12-10 | 湖北天天数链技术有限公司 | Evaluation method and device for people and sentry matching, electronic equipment and storage medium |
| WO2023273583A1 (en) * | 2021-06-29 | 2023-01-05 | 上海商汤智能科技有限公司 | Exam-marking method and apparatus, electronic device, and storage medium |
| CN116304067A (en) * | 2023-05-24 | 2023-06-23 | 广州宏途数字科技有限公司 | Cloud paper reading data analysis method, system, equipment and medium |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116595129B (en) * | 2023-06-12 | 2023-10-27 | 广州市南方人力资源评价中心有限公司 | Subjective question scoring method and device based on knowledge point labeling |
| CN116798036B (en) * | 2023-06-27 | 2024-04-02 | 网才科技(广州)集团股份有限公司 | Method and device for identifying and checking answer sheet objective question identification result |
| CN116778032B (en) * | 2023-07-03 | 2024-04-16 | 北京博思创成技术发展有限公司 | Answer sheet generation method, device, equipment and storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108764074A (en) * | 2018-05-14 | 2018-11-06 | 山东师范大学 | Subjective item intelligently reading method, system and storage medium based on deep learning |
| CN110096702A (en) * | 2019-04-22 | 2019-08-06 | 安徽省泰岳祥升软件有限公司 | A kind of subjective item methods of marking and device |
| CN110348400A (en) * | 2019-07-15 | 2019-10-18 | 京东方科技集团股份有限公司 | A kind of scoring acquisition methods, device and electronic equipment |
| CN112527968A (en) * | 2020-12-22 | 2021-03-19 | 大唐融合通信股份有限公司 | Composition review method and system based on neural network |
Family Cites Families (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2012088494A (en) * | 2010-10-19 | 2012-05-10 | Kinki Univ | Examination question evaluation system and method for controlling the same, program and recording medium |
| CN107977637B (en) * | 2017-12-11 | 2021-09-28 | 上海启思教育科技服务有限公司 | Multi-question intelligent paper marking system |
| CN110110581B (en) * | 2019-03-14 | 2021-06-04 | 杭州笔声智能科技有限公司 | Test paper correcting method and system based on artificial intelligence |
| CN112115736B (en) * | 2019-06-19 | 2022-07-01 | 广东小天才科技有限公司 | Job correction method and system based on image recognition and intelligent terminal |
| CN110322379A (en) * | 2019-07-08 | 2019-10-11 | 深圳中兴网信科技有限公司 | Paper processing method and paper processing system |
| CN112580503A (en) * | 2020-12-17 | 2021-03-30 | 深圳市元德教育科技有限公司 | Operation correction method, device, equipment and storage medium |
| CN113343972A (en) * | 2021-06-29 | 2021-09-03 | 北京市商汤科技开发有限公司 | Paper marking method and device, electronic equipment and storage medium |
-
2021
- 2021-06-29 CN CN202110726513.0A patent/CN113343972A/en active Pending
-
2022
- 2022-04-28 WO PCT/CN2022/089787 patent/WO2023273583A1/en unknown
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108764074A (en) * | 2018-05-14 | 2018-11-06 | 山东师范大学 | Subjective item intelligently reading method, system and storage medium based on deep learning |
| CN110096702A (en) * | 2019-04-22 | 2019-08-06 | 安徽省泰岳祥升软件有限公司 | A kind of subjective item methods of marking and device |
| CN110348400A (en) * | 2019-07-15 | 2019-10-18 | 京东方科技集团股份有限公司 | A kind of scoring acquisition methods, device and electronic equipment |
| CN112527968A (en) * | 2020-12-22 | 2021-03-19 | 大唐融合通信股份有限公司 | Composition review method and system based on neural network |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2023273583A1 (en) * | 2021-06-29 | 2023-01-05 | 上海商汤智能科技有限公司 | Exam-marking method and apparatus, electronic device, and storage medium |
| CN113780997A (en) * | 2021-09-16 | 2021-12-10 | 湖北天天数链技术有限公司 | Evaluation method and device for people and sentry matching, electronic equipment and storage medium |
| CN116304067A (en) * | 2023-05-24 | 2023-06-23 | 广州宏途数字科技有限公司 | Cloud paper reading data analysis method, system, equipment and medium |
| CN116304067B (en) * | 2023-05-24 | 2023-09-12 | 广州宏途数字科技有限公司 | Cloud paper reading data analysis method, system, equipment and medium |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2023273583A1 (en) | 2023-01-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11508251B2 (en) | Method and system for intelligent identification and correction of questions | |
| CN113343972A (en) | Paper marking method and device, electronic equipment and storage medium | |
| CN110110585B (en) | Intelligent paper reading implementation method and system based on deep learning and computer program | |
| US11790641B2 (en) | Answer evaluation method, answer evaluation system, electronic device, and medium | |
| CN109710590B (en) | Error problem book generation method and device | |
| CN109359215B (en) | Video intelligent pushing method and system | |
| CN109271401B (en) | Topic searching and correcting method and device, electronic equipment and storage medium | |
| CN109817046B (en) | Learning auxiliary method based on family education equipment and family education equipment | |
| CN109815491B (en) | Answer scoring method, device, computer equipment and storage medium | |
| US20200286402A1 (en) | Method and system for assisting with math problem | |
| CN110110581B (en) | Test paper correcting method and system based on artificial intelligence | |
| CN111159356B (en) | Knowledge graph construction method based on teaching content | |
| Doolan | Generation 1.5 writing compared to L1 and L2 writing in first-year composition | |
| JP7077483B2 (en) | Problem correction methods, devices, electronic devices and storage media for mental arithmetic problems | |
| CN112347997A (en) | Test question detection and identification method and device, electronic equipment and medium | |
| CN113157867A (en) | Question answering method and device, electronic equipment and storage medium | |
| CN111767307A (en) | Object processing method and system, and mathematic subject correcting method and system | |
| CN113888757A (en) | Examination paper intelligent analysis method, examination paper intelligent analysis system and storage medium based on benchmarking evaluation | |
| CN112700203B (en) | Intelligent marking method and device | |
| CN112860845A (en) | Test question retrieval method and device, electronic equipment and storage medium | |
| CN115599962B (en) | Data resource cloud storage sharing method based on education field | |
| Ullah et al. | An E-Assessment Methodology Based on Artificial Intelligence Techniques to Determine Students' Language Quality and Programming Assignments' Plagiarism. | |
| CN111104503A (en) | Construction engineering quality acceptance standard question-answering system and construction method thereof | |
| Ibrahim et al. | A data mining framework for analyzing students’ feedback of assessment | |
| CN115050039A (en) | Automatic test paper analysis method, system, electronic device and storage medium |
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 | ||
| REG | Reference to a national code |
Ref country code: HK Ref legal event code: DE Ref document number: 40049358 Country of ref document: HK |
|
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210903 |