CN112818133A - Depth knowledge tracking method and system - Google Patents
Depth knowledge tracking method and system Download PDFInfo
- Publication number
- CN112818133A CN112818133A CN202110152127.5A CN202110152127A CN112818133A CN 112818133 A CN112818133 A CN 112818133A CN 202110152127 A CN202110152127 A CN 202110152127A CN 112818133 A CN112818133 A CN 112818133A
- Authority
- CN
- China
- Prior art keywords
- knowledge
- data
- character
- content
- digital
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
- G06F16/367—Ontology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/3331—Query processing
- G06F16/334—Query execution
- G06F16/3344—Query execution using natural language analysis
-
- 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
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/768—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using context analysis, e.g. recognition aided by known co-occurring patterns
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
- G06V20/62—Text, e.g. of license plates, overlay texts or captions on TV images
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
- G06V30/28—Character recognition specially adapted to the type of the alphabet, e.g. Latin alphabet
- G06V30/287—Character recognition specially adapted to the type of the alphabet, e.g. Latin alphabet of Kanji, Hiragana or Katakana characters
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Multimedia (AREA)
- Health & Medical Sciences (AREA)
- Data Mining & Analysis (AREA)
- General Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- Software Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Animal Behavior & Ethology (AREA)
- Medical Informatics (AREA)
- Computing Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention provides a deep knowledge tracking method and a system, which can convert the knowledge content existing on a paper textbook into digital knowledge data in an editable form, utilize a corresponding knowledge database and a character semantic recognition result of the digital knowledge data to carry out deep mining and extension on the content of knowledge points so as to obtain associated knowledge point data, and finally carry out permutation and combination on the associated knowledge point data so as to obtain a deep knowledge map, so that the textbook knowledge content can be purposefully mined and expanded, and the deep knowledge map based on the content of a certain knowledge point is formed, so that the corresponding divergent teaching can be conveniently carried out on the knowledge points in the teaching process, and the teaching quality and the divergent thinking ability of students are improved.
Description
Technical Field
The invention relates to the technical field of intelligent teaching, in particular to a depth knowledge tracking method and a depth knowledge tracking system.
Background
At present, knowledge teaching of students is realized in offline classes or online classes, and both the offline classes and the online classes are performed according to textbook contents, so that corresponding knowledge data cannot be effectively expanded and mined in the teaching process, and the teaching quality is seriously influenced and the divergent thinking ability of the students cannot be improved. Although the related content can be searched for a certain knowledge point through a corresponding search engine in the prior art, the searching mode is not targeted, effective associated searching cannot be performed according to the current knowledge point, and a knowledge graph related to the current knowledge point is formed according to the searching result, so that the accuracy and reliability of tracking and expanding the knowledge point are greatly reduced, and meanwhile, the intuition and the accuracy of knowledge content in knowledge teaching are not improved.
Disclosure of Invention
Aiming at the defects in the prior art, the invention provides a depth knowledge tracking method and a system, which scan a paper textbook to obtain a corresponding paper textbook image and process the paper textbook image, thereby converting the knowledge content contained in the paper textbook into digital knowledge data in an editable form, judging the language type corresponding to the characters contained in the digital knowledge data, and according to the language type, selecting a matched knowledge database, then carrying out character semantic recognition processing on the digital knowledge data, determining all knowledge point contents contained in the digital knowledge data, finally selecting knowledge point data associated with the knowledge point contents from the knowledge database, and according to the difficulty degree of the selected associated knowledge point data, arranging all the associated knowledge points, and forming a depth knowledge map about the content of the knowledge points according to an arrangement result; therefore, the deep knowledge tracking method and the deep knowledge tracking system can convert knowledge contents existing on a paper textbook into digital knowledge data in an editable form, utilize a corresponding knowledge database and a character semantic recognition result of the digital knowledge data to carry out deep mining and extension on the contents of knowledge points so as to obtain associated knowledge point data, and finally carry out permutation and combination on the associated knowledge point data so as to obtain a deep knowledge map, so that the textbook knowledge contents can be purposefully mined and expanded, and the deep knowledge map based on the contents of the knowledge points is formed, so that corresponding divergent teaching can be conveniently carried out on the knowledge points in the teaching process, and the teaching quality and the divergent thinking ability of students are improved.
The invention provides a depth knowledge tracking method, which is characterized by comprising the following steps:
step S1, scanning the paper textbook to obtain the corresponding paper textbook image, and processing the paper textbook image, so as to convert the knowledge content contained in the paper textbook into digital knowledge data in editable form;
step S2, judging the language type corresponding to the character contained in the digital knowledge data, selecting a matched knowledge database according to the language type, and then performing character semantic recognition processing on the digital knowledge data to determine all knowledge point contents contained in the digital knowledge data;
step S3, selecting knowledge point data associated with the knowledge point content from the knowledge database, arranging all associated knowledge points according to the difficulty degree of the selected associated knowledge point data, and forming a depth knowledge map about the knowledge point content according to the arrangement result;
further, in step S1, scanning the paper textbook to obtain a corresponding paper textbook image, and processing the paper textbook image so as to convert the knowledge content contained in the paper textbook into digital knowledge data in an editable form specifically includes:
step S101, scanning and shooting the paper textbook to obtain a corresponding paper textbook image;
step S102, sequentially carrying out image Kalman filtering processing and pixel binarization processing on the paper textbook image, thereby converting the paper textbook image into a binarization image;
step S103, recognizing all character contents contained in the binary image, and converting the recognized character contents into digital knowledge data in an editable electronic text form;
further, in step S2, the determining the language type corresponding to the characters contained in the digital knowledge data, selecting a matching knowledge database according to the language type, and performing character semantic recognition processing on the digital knowledge data, wherein the determining all knowledge point contents contained in the digital knowledge data specifically includes:
step S201, randomly selecting a plurality of nonadjacent characters from the digital knowledge data, determining whether each selected character is a Chinese character or a foreign character, determining whether the actual proportion of the Chinese characters in all the selected characters exceeds a preset proportion threshold value, and if so, determining that the digital knowledge data is Chinese type knowledge data; otherwise, determining that the digital knowledge data is foreign language type knowledge data;
step S202, when the digital knowledge data is Chinese type knowledge data, selecting a corresponding Chinese knowledge database, and when the digital knowledge data is foreign type knowledge data, selecting a corresponding foreign knowledge database;
step S203, performing character semantic recognition processing on the digital knowledge data to obtain character semantic information corresponding to the digital knowledge data, and extracting all knowledge point contents from the character semantic information;
further, in step S3, selecting knowledge point data associated with the knowledge point content from the knowledge database, ranking all associated knowledge points according to the difficulty level of the selected associated knowledge point data, and forming a deep knowledge graph about the knowledge point content according to the ranking result specifically includes:
step S301, extracting all subject knowledge data belonging to the same subject as the knowledge point content from the knowledge database, determining the character content similarity between each subject knowledge data and the knowledge point content in the character content, and taking the subject knowledge data with the character content similarity exceeding a preset similarity threshold as the associated knowledge point data;
step S302, arranging all the associated knowledge points determined in the step S301 according to the sequence of the associated knowledge point data determined in the step S301 from easy to difficult, thereby obtaining a corresponding knowledge point data sequence;
step S303, labeling a formula annotation or a background material annotation corresponding to each knowledge point content contained in the knowledge point data sequence, thereby forming a depth knowledge map about the knowledge point content;
further, in step S301, extracting all subject knowledge data belonging to the same subject as the knowledge point content from the knowledge database, and determining the character content similarity between each subject knowledge data and the knowledge point content in the character content specifically includes:
step S3011, encoding all subject knowledge data in the knowledge database according to character content and character sequence in the subject knowledge data by using the following formula (1),
in the above-mentioned formula (1),
indicating the code number of the k-th character corresponding to the j-th subject knowledge data of the ith subject in the knowledge database, CkThe method comprises the steps that a K-th character corresponding to j-th subject knowledge data of an ith subject in a knowledge database is represented, K represents the total number of characters contained in j-th subject knowledge data of the ith subject in the knowledge database, and ASCII () represents ASCII code value calculation on characters in parentheses;
step S3012, using the following formula (2), extracting all subject knowledge data belonging to the same subject as the knowledge point content from the knowledge database,
in the above-mentioned formula (2),
a k-th character of j-th subject knowledge data representing an ith subject in the knowledge database and an extraction item character of the knowledge point content belonging to the same subject, wherein the extraction item character specifically comprises: when in use
When the current knowledge point content is not in the same discipline, the k character of the j discipline knowledge data of the i discipline in the knowledge database does not belong to the same discipline as the knowledge point content, and the code number of the k character of the j discipline knowledge data of the i discipline which does not belong to the same discipline as the knowledge point content is reset to zero, and the current code number is given to the k character of the j discipline knowledge data of the i discipline
The value is zero; when in use
When the current knowledge point content is in the knowledge database, the kth character of the jth subject knowledge data of the ith subject in the knowledge database belongs to the same subject as the knowledge point content, and the kth character of the jth subject knowledge data of the ith subject belonging to the same subject as the knowledge point content is kept unchanged in numerical value, and then the current knowledge point content is given to the current knowledge database
Taking the value as the code number of the kth character; i represents that the knowledge point content belongs to the I-th subject, delta () represents a unit impulse function, the value of the unit impulse function is 1 when the value in the bracket is equal to 0, and the value of the unit impulse function is 1 when the value in the bracket is not equal to 0;
step S3013, determining the character content similarity between each extracted subject knowledge data and the knowledge point content in the character content by using the following formula (3),
in the above formula (3), QjRepresenting the similarity of character contents between jth subject knowledge data extracted from the knowledge database and the knowledge point content, L representing the total number of characters contained in the knowledge point content, n representing the total number of subject types contained in the knowledge database, and ZlThe number of codes for the first character representing the content of the knowledge point, and the number of codes ZlThe determination of (2) is the same as the process of determining the number of codes in step S3011 described above.
The invention also provides a depth knowledge tracking system which is characterized by comprising a paper textbook scanning module, a textbook image processing module, a knowledge point content acquisition module and a depth knowledge map forming module; wherein the content of the first and second substances,
the paper textbook scanning module is used for scanning the paper textbook so as to obtain a corresponding paper textbook image;
the textbook image processing module is used for processing the paper textbook image so as to convert the knowledge content contained in the paper textbook into digital knowledge data in an editable form;
the knowledge point content acquisition module is used for judging the language type corresponding to the characters contained in the digital knowledge data, selecting a matched knowledge database according to the language type, and performing character semantic recognition processing on the digital knowledge data so as to determine all knowledge point contents contained in the digital knowledge data;
the depth knowledge map forming module is used for selecting knowledge point data associated with the knowledge point content from the knowledge database, arranging all associated knowledge points according to the difficulty degree of the selected associated knowledge point data, and forming a depth knowledge map related to the knowledge point content according to the arrangement result;
further, the scanning of the paper textbook by the paper textbook scanning module to obtain the corresponding image of the paper textbook specifically includes:
scanning and shooting the paper textbook so as to obtain a corresponding paper textbook image;
and the number of the first and second groups,
the processing module processes the image of the paper textbook, so as to convert the knowledge content contained in the paper textbook into digital knowledge data in an editable form, specifically comprising:
sequentially carrying out image Kalman filtering processing and pixel binarization processing on the paper textbook image, thereby converting the paper textbook image into a binarization image;
then identifying all character contents contained in the binary image, and converting the identified character contents into digital knowledge data in an editable electronic text form;
further, the knowledge point content acquiring module judges a language type corresponding to a character contained in the digital knowledge data, selects a matched knowledge database according to the language type, and performs character semantic recognition processing on the digital knowledge data, wherein the step of determining all knowledge point contents contained in the digital knowledge data specifically comprises the steps of:
randomly selecting a plurality of nonadjacent characters from the digital knowledge data, determining whether each selected character is a Chinese character or a foreign character, determining whether the actual proportion of the Chinese characters in all the selected characters exceeds a preset proportion threshold, and if so, determining that the digital knowledge data is Chinese type knowledge data; otherwise, determining that the digital knowledge data is foreign language type knowledge data;
when the digital knowledge data is Chinese type knowledge data, selecting a corresponding Chinese knowledge database, and when the digital knowledge data is foreign type knowledge data, selecting a corresponding foreign knowledge database;
finally, carrying out character semantic recognition processing on the digital knowledge data so as to obtain character semantic information corresponding to the digital knowledge data, and extracting all knowledge point contents from the character semantic information;
further, the forming module of the deep knowledge map selects knowledge point data associated with the knowledge point content from the knowledge database, ranks all associated knowledge points according to the difficulty level of the selected associated knowledge point data, and forms the deep knowledge map related to the knowledge point content according to the ranking result specifically includes:
extracting all subject knowledge data belonging to the same subject as the knowledge point content from the knowledge database, determining the character content similarity between each subject knowledge data and the knowledge point content in the character content, and taking the subject knowledge data with the character content similarity exceeding a preset similarity threshold as the associated knowledge point data;
arranging all the associated knowledge points determined in the above according to the sequence of the associated knowledge point data determined in the above from easy to difficult, thereby obtaining a corresponding knowledge point data sequence;
and finally, labeling the corresponding formula annotation or background data annotation of each knowledge point content contained in the knowledge point data sequence, thereby forming a depth knowledge map about the knowledge point content.
Compared with the prior art, the deep knowledge tracking method and the deep knowledge tracking system scan the paper textbook to obtain the corresponding paper textbook image and process the paper textbook image, thereby converting the knowledge content contained in the paper textbook into digital knowledge data in an editable form, judging the language type corresponding to the characters contained in the digital knowledge data, and according to the language type, selecting a matched knowledge database, then carrying out character semantic recognition processing on the digital knowledge data, determining all knowledge point contents contained in the digital knowledge data, finally selecting knowledge point data associated with the knowledge point contents from the knowledge database, and according to the difficulty degree of the selected associated knowledge point data, arranging all the associated knowledge points, and forming a depth knowledge map about the content of the knowledge points according to an arrangement result; therefore, the deep knowledge tracking method and the deep knowledge tracking system can convert knowledge contents existing on a paper textbook into digital knowledge data in an editable form, utilize a corresponding knowledge database and a character semantic recognition result of the digital knowledge data to carry out deep mining and extension on the contents of knowledge points so as to obtain associated knowledge point data, and finally carry out permutation and combination on the associated knowledge point data so as to obtain a deep knowledge map, so that the textbook knowledge contents can be purposefully mined and expanded, and the deep knowledge map based on the contents of the knowledge points is formed, so that corresponding divergent teaching can be conveniently carried out on the knowledge points in the teaching process, and the teaching quality and the divergent thinking ability of students are improved.
Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings.
The technical solution of the present invention is further described in detail by the accompanying drawings and embodiments.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
Fig. 1 is a schematic flow chart of a depth knowledge tracking method according to the present invention.
Fig. 2 is a schematic structural diagram of a depth knowledge tracking system according to the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Fig. 1 is a schematic flow chart of a depth knowledge tracking method according to an embodiment of the present invention. The depth knowledge tracking method comprises the following steps:
step S1, scanning the paper textbook to obtain the corresponding paper textbook image, and processing the paper textbook image, thereby converting the knowledge content contained in the paper textbook into digital knowledge data in editable form;
step S2, judging the language type corresponding to the character contained in the digital knowledge data, selecting the matched knowledge database according to the language type, and then performing the character semantic recognition processing on the digital knowledge data to determine all knowledge point contents contained in the digital knowledge data;
step S3, selecting knowledge point data associated with the knowledge point content from the knowledge database, arranging all associated knowledge points according to the difficulty of the selected associated knowledge point data, and forming a depth knowledge map about the knowledge point content according to the arrangement result.
The beneficial effects of the above technical scheme are: the deep knowledge tracking method can convert the knowledge content existing in a paper textbook into digital knowledge data in an editable form, deeply excavates and extends the content of knowledge points by utilizing a corresponding knowledge database and a character semantic recognition result of the digital knowledge data to obtain associated knowledge point data, and finally arranges and combines the associated knowledge point data to obtain a deep knowledge map, so that the textbook knowledge content can be excavated and expanded in a targeted manner to form the deep knowledge map based on the content of a certain knowledge point, and the knowledge points can be conveniently and correspondingly divergently taught in the teaching process, thereby improving the teaching quality and the divergent thinking ability of students.
Preferably, in step S1, scanning the paper textbook to obtain a corresponding paper textbook image, and processing the paper textbook image so as to convert the knowledge content contained in the paper textbook into digital knowledge data in editable form specifically includes:
step S101, scanning and shooting the paper textbook to obtain a corresponding paper textbook image;
step S102, sequentially carrying out image Kalman filtering processing and pixel binarization processing on the paper textbook image, thereby converting the paper textbook image into a binarization image;
step S103, recognizing all character contents contained in the binary image, and converting the recognized character contents into digital knowledge data in an editable electronic text form.
The beneficial effects of the above technical scheme are: the mode of scanning and shooting is adopted, so that different paper courseware can be conveniently shot according to actual requirements, the acquisition convenience of knowledge content is improved, image Kalman filtering processing and pixel binarization processing are carried out on the paper courseware image, noise components in the image can be effectively removed, and the accuracy of subsequent character content identification and conversion of the character content into editable electronic text form digital knowledge data is improved.
Preferably, in step S2, the determining the language type corresponding to the characters contained in the digital knowledge data, selecting a matching knowledge database according to the language type, and performing character semantic recognition processing on the digital knowledge data, wherein the determining all knowledge point contents contained in the digital knowledge data specifically includes:
step S201, randomly selecting a plurality of nonadjacent characters from the digital knowledge data, determining whether each selected character is a Chinese character or a foreign character, determining whether the actual ratio of the Chinese characters in all the selected characters exceeds a preset ratio threshold, and if so, determining that the digital knowledge data is Chinese type knowledge data; otherwise, determining the digital knowledge data as foreign language type knowledge data;
step S202, when the digital knowledge data is Chinese type knowledge data, selecting a corresponding Chinese knowledge database, and when the digital knowledge data is foreign type knowledge data, selecting a corresponding foreign knowledge database;
step S203, performing character semantic identification processing on the digital knowledge data to obtain character semantic information corresponding to the digital knowledge data, and extracting all knowledge point contents from the character semantic information.
The beneficial effects of the above technical scheme are: because the knowledge contents corresponding to the digital knowledge data of different languages are different, a plurality of nonadjacent characters are randomly selected from the digital knowledge data, each selected character is determined to be a Chinese character or a foreign character, and then whether the actual proportion of the Chinese characters in all the selected characters exceeds a preset proportion threshold value is determined, so that the condition of character language type misjudgment can be effectively avoided, the character language type corresponding to the digital knowledge data can be rapidly and accurately determined, and the matched knowledge database can be determined in a targeted manner; in addition, the existing character recognition software can be adopted to carry out character semantic recognition on the digital knowledge data, so that the recognition accuracy of the knowledge point content is improved.
Preferably, in step S3, selecting knowledge point data associated with the knowledge point content from the knowledge database, ranking all associated knowledge points according to the difficulty level of the selected associated knowledge point data, and forming a deep knowledge graph about the knowledge point content according to the ranking result specifically includes:
step S301, extracting all subject knowledge data belonging to the same subject as the knowledge point content from the knowledge database, determining the character content similarity between each subject knowledge data and the knowledge point content in the character content, and taking the subject knowledge data with the character content similarity exceeding a preset similarity threshold as the associated knowledge point data;
step S302, arranging all the associated knowledge points determined in the step S301 according to the sequence of the associated knowledge point data determined in the step S301 from easy to difficult, thereby obtaining a corresponding knowledge point data sequence;
step S303, labeling each knowledge point content included in the knowledge point data sequence with its corresponding formula annotation or background annotation, thereby forming a depth knowledge map related to the knowledge point content.
The beneficial effects of the above technical scheme are: by extracting subject knowledge data from a data set belonging to the same subject as the knowledge point content in the knowledge database, the search range of the subject knowledge data can be effectively reduced, and the efficiency of determining the associated knowledge point data can be improved; in addition, the associated knowledge point data are arranged according to the sequence from easy to difficult, and corresponding formula annotations or background material annotations are marked for each knowledge point data, so that the knowledge coverage and the knowledge content display intuitiveness of the deep knowledge map can be improved.
Preferably, in step S301, extracting all subject knowledge data belonging to the same subject as the knowledge point content from the knowledge database, and determining the character content similarity between each subject knowledge data and the knowledge point content in the character content specifically includes:
step S3011, encoding all subject knowledge data in the knowledge database according to the character content and character sequence in the subject knowledge data by using the following formula (1),
in the above-mentioned formula (1),
indicating the code number of the k-th character corresponding to the j-th subject knowledge data of the i-th subject in the knowledge database, CkThe device comprises a knowledge database, a first input module, a second input module, a first output module and a second output module, wherein the knowledge database comprises a knowledge data base, wherein the knowledge data base comprises a jth subject of an ith subject;
step S3012, using the following formula (2), extracting all subject knowledge data belonging to the same subject as the content of the knowledge point from the knowledge database,
in the above-mentioned formula (2),
the k-th character of the j-th subject knowledge data representing the ith subject in the knowledge database and the extraction item character of the knowledge point content belonging to the same subject specifically include: when in use
When the current knowledge point content is not in the same discipline, the k character of the j discipline knowledge data of the i discipline in the knowledge database does not belong to the same discipline as the knowledge point content, and the code number of the k character of the j discipline knowledge data of the i discipline which does not belong to the same discipline as the knowledge point content is reset to zero, and the code number is given to the k character of the j discipline knowledge data of the i discipline in the current moment
The value is zero; when in use
When the current knowledge point content is in the same discipline as the current discipline, the k character of the j discipline knowledge data of the i discipline in the knowledge database and the content of the knowledge point belong to the same discipline, and the k character of the j discipline knowledge data of the i discipline belonging to the same discipline as the current discipline is kept unchanged, and the current character is given to the current discipline knowledge data
The value is the code number of the kth character; i represents that the content of the knowledge point belongs to the I-th subject, delta () represents a unit impulse function, the value of the unit impulse function is 1 when the value in the bracket is equal to 0, and the value of the unit impulse function is 1 when the value in the bracket is not equal to 0;
step S3013, using the following formula (3), determining the character content similarity between the extracted knowledge data of each subject and the content of the knowledge point in the character content,
in the above formula (3), QjIndicate from thatThe character content similarity between the jth subject knowledge data extracted from the knowledge database and the knowledge point content, L represents the total number of characters contained in the knowledge point content, n represents the total number of subject types contained in the knowledge database, and Z represents the character content similarity between the jth subject knowledge data and the knowledge point contentlThe number of codes for the first character representing the content of the knowledge point, and the number of codes ZlThe determination of (2) is the same as the process of determining the number of codes in step S3011 described above.
The beneficial effects of the above technical scheme are: all subject knowledge data in the knowledge database are encoded according to the character content and the character sequence in the subject knowledge data by using the formula (1), so that the character content of the knowledge data is converted into the number of codes which can be identified by a computer, and subsequent calculation and similarity calculation are facilitated; then, all subject knowledge data belonging to the same subject as the knowledge point content are extracted from the knowledge database by using the formula (2), so that all subject knowledge data belonging to the same subject as the knowledge point content are accurately extracted, and the reliability and the accuracy of the extracted data are ensured; and finally, obtaining the actual similarity of each extracted subject knowledge data and the knowledge point content on the character content by using the formula (3), and taking the subject knowledge data with the actual similarity exceeding a preset similarity threshold as associated knowledge point data to ensure the stability of the system and the accuracy of the data.
Fig. 2 is a schematic structural diagram of a depth knowledge tracking system according to an embodiment of the present invention. The depth knowledge tracking system comprises a paper textbook scanning module, a textbook image processing module, a knowledge point content acquisition module and a depth knowledge map forming module; wherein the content of the first and second substances,
the paper textbook scanning module is used for scanning the paper textbook so as to obtain a corresponding paper textbook image;
the textbook image processing module is used for processing the paper textbook image so as to convert the knowledge content contained in the paper textbook into digital knowledge data in an editable form;
the knowledge point content acquisition module is used for judging the language type corresponding to the characters contained in the digital knowledge data, selecting a matched knowledge database according to the language type, and performing character semantic recognition processing on the digital knowledge data so as to determine all knowledge point contents contained in the digital knowledge data;
the depth knowledge map forming module is used for selecting knowledge point data associated with the knowledge point content from the knowledge database, arranging all associated knowledge points according to the difficulty degree of the selected associated knowledge point data, and forming a depth knowledge map related to the knowledge point content according to the arrangement result.
The beneficial effects of the above technical scheme are: the deep knowledge tracking system can convert knowledge contents existing in a paper textbook into digital knowledge data in an editable form, deeply excavates and extends knowledge point contents by utilizing a corresponding knowledge database and a character semantic recognition result of the digital knowledge data to obtain associated knowledge point data, and finally arranges and combines the associated knowledge point data to obtain a deep knowledge map, so that the textbook knowledge contents can be excavated and expanded in a targeted manner to form the deep knowledge map based on the knowledge point contents, and the corresponding divergent teaching of the knowledge points in the teaching process is facilitated, so that the teaching quality and the divergent thinking ability of students are improved.
Preferably, the scanning the paper textbook by the paper textbook scanning module to obtain the corresponding image of the paper textbook specifically includes:
scanning and shooting the paper textbook so as to obtain a corresponding paper textbook image;
and the number of the first and second groups,
the processing module processes the image of the paper textbook, so as to convert the knowledge content contained in the paper textbook into digital knowledge data in an editable form, specifically comprising:
sequentially carrying out image Kalman filtering processing and pixel binarization processing on the paper textbook image so as to convert the paper textbook image into a binarization image;
and then recognizing all character contents contained in the binary image, and converting the recognized character contents into digital knowledge data in an editable electronic text form.
The beneficial effects of the above technical scheme are: the mode of scanning and shooting is adopted, so that different paper courseware can be conveniently shot according to actual requirements, the acquisition convenience of knowledge content is improved, image Kalman filtering processing and pixel binarization processing are carried out on the paper courseware image, noise components in the image can be effectively removed, and the accuracy of subsequent character content identification and conversion of the character content into editable electronic text form digital knowledge data is improved.
Preferably, the knowledge point content acquiring module judges a language type corresponding to a character contained in the digital knowledge data, selects a matched knowledge database according to the language type, and performs character semantic recognition processing on the digital knowledge data, so as to determine all knowledge point contents contained in the digital knowledge data specifically include:
randomly selecting a plurality of nonadjacent characters from the digital knowledge data, determining whether each selected character is a Chinese character or a foreign character, determining whether the actual proportion of the Chinese characters in all the selected characters exceeds a preset proportion threshold, and if so, determining that the digital knowledge data is Chinese type knowledge data; otherwise, determining the digital knowledge data as foreign language type knowledge data;
when the digital knowledge data is Chinese type knowledge data, selecting a corresponding Chinese knowledge database, and when the digital knowledge data is foreign type knowledge data, selecting a corresponding foreign knowledge database;
and finally, carrying out character semantic recognition processing on the digital knowledge data so as to obtain character semantic information corresponding to the digital knowledge data, and extracting all knowledge point contents from the character semantic information.
The beneficial effects of the above technical scheme are: because the knowledge contents corresponding to the digital knowledge data of different languages are different, a plurality of nonadjacent characters are randomly selected from the digital knowledge data, each selected character is determined to be a Chinese character or a foreign character, and then whether the actual proportion of the Chinese characters in all the selected characters exceeds a preset proportion threshold value is determined, so that the condition of character language type misjudgment can be effectively avoided, the character language type corresponding to the digital knowledge data can be rapidly and accurately determined, and the matched knowledge database can be determined in a targeted manner; in addition, the existing character recognition software can be adopted to carry out character semantic recognition on the digital knowledge data, so that the recognition accuracy of the knowledge point content is improved.
Preferably, the forming module of the deep knowledge map selects knowledge point data associated with the content of the knowledge point from the knowledge database, ranks all the associated knowledge points according to the difficulty level of the selected associated knowledge point data, and forms the deep knowledge map about the content of the knowledge point according to the ranking result specifically includes:
extracting all subject knowledge data belonging to the same subject as the knowledge point content from the knowledge database, determining the character content similarity between each subject knowledge data and the knowledge point content in the character content, and taking the subject knowledge data with the character content similarity exceeding a preset similarity threshold as the associated knowledge point data;
arranging all the associated knowledge points determined in the above according to the sequence of the associated knowledge point data determined in the above from easy to difficult, thereby obtaining a corresponding knowledge point data sequence;
and finally, labeling the corresponding formula annotation or background data annotation of each knowledge point content contained in the knowledge point data sequence, thereby forming a depth knowledge map about the knowledge point content.
The beneficial effects of the above technical scheme are: by extracting subject knowledge data from a data set belonging to the same subject as the knowledge point content in the knowledge database, the search range of the subject knowledge data can be effectively reduced, and the efficiency of determining the associated knowledge point data can be improved; in addition, the associated knowledge point data are arranged according to the sequence from easy to difficult, and corresponding formula annotations or background material annotations are marked for each knowledge point data, so that the knowledge coverage and the knowledge content display intuitiveness of the deep knowledge map can be improved.
As can be seen from the above description of the embodiments, the method and system for tracking deep knowledge obtains the corresponding image of the paper textbook by scanning the paper textbook, and processes the image of the paper textbook, thereby converting the knowledge content contained in the paper textbook into digital knowledge data in an editable form, judging the language type corresponding to the characters contained in the digital knowledge data, and according to the language type, selecting a matched knowledge database, then carrying out character semantic recognition processing on the digital knowledge data, determining all knowledge point contents contained in the digital knowledge data, finally selecting knowledge point data associated with the knowledge point contents from the knowledge database, and according to the difficulty degree of the selected associated knowledge point data, arranging all the associated knowledge points, and forming a depth knowledge map about the content of the knowledge points according to an arrangement result; therefore, the deep knowledge tracking method and the deep knowledge tracking system can convert knowledge contents existing on a paper textbook into digital knowledge data in an editable form, utilize a corresponding knowledge database and a character semantic recognition result of the digital knowledge data to carry out deep mining and extension on the contents of knowledge points so as to obtain associated knowledge point data, and finally carry out permutation and combination on the associated knowledge point data so as to obtain a deep knowledge map, so that the textbook knowledge contents can be purposefully mined and expanded, and the deep knowledge map based on the contents of the knowledge points is formed, so that corresponding divergent teaching can be conveniently carried out on the knowledge points in the teaching process, and the teaching quality and the divergent thinking ability of students are improved.
It will be apparent to those skilled in the art that various changes and modifications may be made in the present invention without departing from the spirit and scope of the invention. Thus, if such modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include such modifications and variations.
Claims (9)
1. The depth knowledge tracking method is characterized by comprising the following steps:
step S1, scanning the paper textbook to obtain the corresponding paper textbook image, and processing the paper textbook image, so as to convert the knowledge content contained in the paper textbook into digital knowledge data in editable form;
step S2, judging the language type corresponding to the character contained in the digital knowledge data, selecting a matched knowledge database according to the language type, and then performing character semantic recognition processing on the digital knowledge data to determine all knowledge point contents contained in the digital knowledge data;
step S3, selecting knowledge point data associated with the knowledge point content from the knowledge database, arranging all associated knowledge points according to the difficulty degree of the selected associated knowledge point data, and forming a depth knowledge map about the knowledge point content according to the arrangement result.
2. The depth knowledge tracking method of claim 1, wherein:
in step S1, scanning the paper textbook to obtain a corresponding paper textbook image, and processing the paper textbook image so as to convert the knowledge content contained in the paper textbook into digital knowledge data in an editable form specifically includes:
step S101, scanning and shooting the paper textbook to obtain a corresponding paper textbook image;
step S102, sequentially carrying out image Kalman filtering processing and pixel binarization processing on the paper textbook image, thereby converting the paper textbook image into a binarization image;
step S103, recognizing all character contents contained in the binary image, and converting the recognized character contents into digital knowledge data in an editable electronic text form.
3. The depth knowledge tracking method of claim 1, wherein:
in step S2, the determining the language type corresponding to the characters contained in the digital knowledge data, selecting a matching knowledge database according to the language type, and performing character semantic recognition processing on the digital knowledge data, wherein the determining the content of all knowledge points contained in the digital knowledge data specifically includes:
step S201, randomly selecting a plurality of nonadjacent characters from the digital knowledge data, determining whether each selected character is a Chinese character or a foreign character, determining whether the actual proportion of the Chinese characters in all the selected characters exceeds a preset proportion threshold value, and if so, determining that the digital knowledge data is Chinese type knowledge data; otherwise, determining that the digital knowledge data is foreign language type knowledge data;
step S202, when the digital knowledge data is Chinese type knowledge data, selecting a corresponding Chinese knowledge database, and when the digital knowledge data is foreign type knowledge data, selecting a corresponding foreign knowledge database;
step S203, performing character semantic recognition processing on the digital knowledge data to obtain character semantic information corresponding to the digital knowledge data, and extracting all knowledge point contents from the character semantic information.
4. The depth knowledge tracking method of claim 1, wherein:
in step S3, selecting knowledge point data associated with the knowledge point content from the knowledge database, ranking all associated knowledge points according to the difficulty of the selected associated knowledge point data, and forming a deep knowledge map about the knowledge point content according to the ranking result specifically includes:
step S301, extracting all subject knowledge data belonging to the same subject as the knowledge point content from the knowledge database, determining the character content similarity between each subject knowledge data and the knowledge point content in the character content, and taking the subject knowledge data with the character content similarity exceeding a preset similarity threshold as the associated knowledge point data;
step S302, arranging all the associated knowledge points determined in the step S301 according to the sequence of the associated knowledge point data determined in the step S301 from easy to difficult, thereby obtaining a corresponding knowledge point data sequence;
step S303, labeling each knowledge point content included in the knowledge point data sequence with a corresponding formula annotation or background annotation, thereby forming a depth knowledge map related to the knowledge point content.
5. The depth knowledge tracking method of claim 4, wherein:
in step S301, extracting all subject knowledge data belonging to the same subject as the knowledge point content from the knowledge database, and determining the character content similarity between each subject knowledge data and the knowledge point content in the character content specifically includes:
step S3011, encoding all subject knowledge data in the knowledge database according to character content and character sequence in the subject knowledge data by using the following formula (1),
in the above-mentioned formula (1),
indicating the code number of the k-th character corresponding to the j-th subject knowledge data of the ith subject in the knowledge database, CkThe method comprises the steps that a K-th character corresponding to j-th subject knowledge data of an ith subject in a knowledge database is represented, K represents the total number of characters contained in j-th subject knowledge data of the ith subject in the knowledge database, and ASCII () represents ASCII code value calculation on characters in parentheses;
step S3012, using the following formula (2), extracting all subject knowledge data belonging to the same subject as the knowledge point content from the knowledge database,
in the above-mentioned formula (2),
a k-th character of j-th subject knowledge data representing an ith subject in the knowledge database and an extraction item character of the knowledge point content belonging to the same subject, wherein the extraction item character specifically comprises: when in use
When the current knowledge point content is not in the same discipline, the k character of the j discipline knowledge data of the i discipline in the knowledge database does not belong to the same discipline as the knowledge point content, and the code number of the k character of the j discipline knowledge data of the i discipline which does not belong to the same discipline as the knowledge point content is reset to zero, and the current code number is given to the k character of the j discipline knowledge data of the i discipline
The value is zero; when in use
When the current knowledge point content is in the knowledge database, the kth character of the jth subject knowledge data of the ith subject in the knowledge database belongs to the same subject as the knowledge point content, and the code number of the kth character of the jth subject knowledge data of the ith subject belonging to the same subject as the knowledge point content is kept unchanged, and the current knowledge point content is given to the current knowledge database
Taking the value as the code number of the kth character; i represents that the knowledge point content belongs to the I-th subject, and delta () represents a unit impulse function,When the value in the bracket is equal to 0, the value of the unit impulse function is 1, and when the value in the bracket is not equal to 0, the value of the unit impulse function is 1;
step S3013, determining the character content similarity between each extracted subject knowledge data and the knowledge point content in the character content by using the following formula (3),
in the above formula (3), QjRepresenting the similarity of character contents between jth subject knowledge data extracted from the knowledge database and the knowledge point content, L representing the total number of characters contained in the knowledge point content, n representing the total number of subject types contained in the knowledge database, and ZlThe number of codes for the first character representing the content of the knowledge point, and the number of codes ZlThe determination of (2) is the same as the process of determining the number of codes in step S3011 described above.
6. The depth knowledge tracking system is characterized by comprising a paper textbook scanning module, a textbook image processing module, a knowledge point content acquisition module and a depth knowledge map forming module; wherein the content of the first and second substances,
the paper textbook scanning module is used for scanning the paper textbook so as to obtain a corresponding paper textbook image;
the textbook image processing module is used for processing the paper textbook image so as to convert the knowledge content contained in the paper textbook into digital knowledge data in an editable form;
the knowledge point content acquisition module is used for judging the language type corresponding to the characters contained in the digital knowledge data, selecting a matched knowledge database according to the language type, and performing character semantic recognition processing on the digital knowledge data so as to determine all knowledge point contents contained in the digital knowledge data;
the depth knowledge map forming module is used for selecting knowledge point data associated with the knowledge point content from the knowledge database, arranging all associated knowledge points according to the difficulty degree of the selected associated knowledge point data, and forming a depth knowledge map related to the knowledge point content according to the arrangement result.
7. The depth knowledge tracking system of claim 6, wherein:
the scanning module scans the paper textbook, and the obtaining of the corresponding image of the paper textbook specifically includes:
scanning and shooting the paper textbook so as to obtain a corresponding paper textbook image;
and the number of the first and second groups,
the processing module processes the image of the paper textbook, so as to convert the knowledge content contained in the paper textbook into digital knowledge data in an editable form, specifically comprising:
sequentially carrying out image Kalman filtering processing and pixel binarization processing on the paper textbook image, thereby converting the paper textbook image into a binarization image;
and then recognizing all character contents contained in the binary image, and converting the recognized character contents into digital knowledge data in an editable electronic text form.
8. The depth knowledge tracking system of claim 6, wherein:
the knowledge point content acquisition module judges the language type corresponding to the characters contained in the digital knowledge data, selects a matched knowledge database according to the language type, and performs character semantic recognition processing on the digital knowledge data, so as to specifically determine all knowledge point contents contained in the digital knowledge data, including:
randomly selecting a plurality of nonadjacent characters from the digital knowledge data, determining whether each selected character is a Chinese character or a foreign character, determining whether the actual proportion of the Chinese characters in all the selected characters exceeds a preset proportion threshold, and if so, determining that the digital knowledge data is Chinese type knowledge data; otherwise, determining that the digital knowledge data is foreign language type knowledge data;
when the digital knowledge data is Chinese type knowledge data, selecting a corresponding Chinese knowledge database, and when the digital knowledge data is foreign type knowledge data, selecting a corresponding foreign knowledge database;
and finally, carrying out character semantic recognition processing on the digital knowledge data so as to obtain character semantic information corresponding to the digital knowledge data, and extracting all knowledge point contents from the character semantic information.
9. The depth knowledge tracking system of claim 6, wherein:
the depth knowledge map forming module selects knowledge point data associated with the knowledge point content from the knowledge database, ranks all associated knowledge points according to the difficulty level of the selected associated knowledge point data, and forms a depth knowledge map about the knowledge point content according to a ranking result, which specifically includes:
extracting all subject knowledge data belonging to the same subject as the knowledge point content from the knowledge database, determining the character content similarity between each subject knowledge data and the knowledge point content in the character content, and taking the subject knowledge data with the character content similarity exceeding a preset similarity threshold as the associated knowledge point data;
arranging all the associated knowledge points determined in the above according to the sequence of the associated knowledge point data determined in the above from easy to difficult, thereby obtaining a corresponding knowledge point data sequence; and finally, labeling the corresponding formula annotation or background data annotation of each knowledge point content contained in the knowledge point data sequence, thereby forming a depth knowledge map about the knowledge point content.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110152127.5A CN112818133A (en) | 2021-02-04 | 2021-02-04 | Depth knowledge tracking method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110152127.5A CN112818133A (en) | 2021-02-04 | 2021-02-04 | Depth knowledge tracking method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112818133A true CN112818133A (en) | 2021-05-18 |
Family
ID=75861181
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110152127.5A Pending CN112818133A (en) | 2021-02-04 | 2021-02-04 | Depth knowledge tracking method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112818133A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114219224A (en) * | 2021-11-24 | 2022-03-22 | 慧之安信息技术股份有限公司 | Teaching quality detection method and system for intelligent classroom |
| CN114372460A (en) * | 2021-12-08 | 2022-04-19 | 北京金山数字娱乐科技有限公司 | Character distinguishing method and device, electronic equipment and storage medium |
-
2021
- 2021-02-04 CN CN202110152127.5A patent/CN112818133A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114219224A (en) * | 2021-11-24 | 2022-03-22 | 慧之安信息技术股份有限公司 | Teaching quality detection method and system for intelligent classroom |
| CN114372460A (en) * | 2021-12-08 | 2022-04-19 | 北京金山数字娱乐科技有限公司 | Character distinguishing method and device, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US6501855B1 (en) | Manual-search restriction on documents not having an ASCII index | |
| CN1282072A (en) | Error correcting method for voice identification result and voice identification system | |
| CN101520783B (en) | Method and device for searching keywords based on image content | |
| CN112818133A (en) | Depth knowledge tracking method and system | |
| CN102073884A (en) | Handwriting recognition method, system and handwriting recognition terminal | |
| CN113033198B (en) | Similar text pushing method and device, electronic equipment and computer storage medium | |
| CN111046201A (en) | Garbage classification simulation system based on computer software | |
| CN113378970A (en) | Sentence similarity detection method and device, electronic equipment and storage medium | |
| CN111538836A (en) | Method for identifying financial advertisements in text advertisements | |
| CN113282754A (en) | Public opinion detection method, device, equipment and storage medium for news events | |
| CN114090736A (en) | Enterprise industry identification system and method based on text similarity | |
| Lu et al. | Retrieval of machine-printed latin documents through word shape coding | |
| CN115759071A (en) | Government affair sensitive information identification system and method based on big data | |
| CN109344233B (en) | Chinese name recognition method | |
| CN102243708B (en) | Handwriting recognition method, handwriting recognition system and handwriting recognition terminal | |
| US20240037328A1 (en) | Method and system for filtering ill corpus | |
| Hirayama et al. | Development of template-free form recognition system | |
| CN112925874B (en) | Similar code searching method and system based on case marks | |
| CN115373982A (en) | Test report analysis method, device, equipment and medium based on artificial intelligence | |
| CN111985193A (en) | Method and system for automatically labeling and classifying questions | |
| CN109344254B (en) | Address information classification method and device | |
| CN112818132A (en) | Method and system for complementing knowledge point relevance based on cognitive map | |
| CN111783786A (en) | Picture identification method and system, electronic equipment and storage medium | |
| CN111460206A (en) | Image processing method, image processing device, electronic equipment and computer readable storage medium | |
| CN111460821A (en) | Entity identification and linking method and device |
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 |