CN117544831A - Automatic decomposing method and system for classroom teaching links - Google Patents
Automatic decomposing method and system for classroom teaching links Download PDFInfo
- Publication number
- CN117544831A CN117544831A CN202311314047.0A CN202311314047A CN117544831A CN 117544831 A CN117544831 A CN 117544831A CN 202311314047 A CN202311314047 A CN 202311314047A CN 117544831 A CN117544831 A CN 117544831A
- Authority
- CN
- China
- Prior art keywords
- teaching
- text
- organization
- video
- link
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 74
- 230000008520 organization Effects 0.000 claims abstract description 159
- 230000008569 process Effects 0.000 claims abstract description 37
- 238000000605 extraction Methods 0.000 claims abstract description 19
- 238000005516 engineering process Methods 0.000 claims abstract description 15
- 230000005236 sound signal Effects 0.000 claims abstract description 9
- 238000002372 labelling Methods 0.000 claims abstract description 4
- 238000012549 training Methods 0.000 claims description 27
- 230000011218 segmentation Effects 0.000 claims description 21
- 238000012706 support-vector machine Methods 0.000 claims description 13
- 238000012805 post-processing Methods 0.000 claims description 12
- 238000004422 calculation algorithm Methods 0.000 claims description 11
- 238000013145 classification model Methods 0.000 claims description 9
- 238000012545 processing Methods 0.000 claims description 9
- 238000012360 testing method Methods 0.000 claims description 9
- 238000010801 machine learning Methods 0.000 claims description 7
- 238000007781 pre-processing Methods 0.000 claims description 7
- 238000011156 evaluation Methods 0.000 claims description 6
- 238000012552 review Methods 0.000 claims description 6
- 238000007596 consolidation process Methods 0.000 claims description 3
- 239000011159 matrix material Substances 0.000 claims description 3
- 238000005065 mining Methods 0.000 claims description 3
- 238000012216 screening Methods 0.000 claims description 2
- 238000004458 analytical method Methods 0.000 abstract description 15
- 230000000694 effects Effects 0.000 abstract description 13
- 238000011160 research Methods 0.000 abstract description 9
- 238000003058 natural language processing Methods 0.000 description 6
- 238000000354 decomposition reaction Methods 0.000 description 4
- 238000013473 artificial intelligence Methods 0.000 description 2
- 230000006399 behavior Effects 0.000 description 2
- 238000004140 cleaning Methods 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 230000000877 morphologic effect Effects 0.000 description 2
- 230000009467 reduction Effects 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 1
- 238000007635 classification algorithm Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000001035 drying Methods 0.000 description 1
- 239000012634 fragment Substances 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000001939 inductive effect Effects 0.000 description 1
- 238000011068 loading method Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 238000004445 quantitative analysis Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/80—Generation or processing of content or additional data by content creator independently of the distribution process; Content per se
- H04N21/83—Generation or processing of protective or descriptive data associated with content; Content structuring
- H04N21/845—Structuring of content, e.g. decomposing content into time segments
- H04N21/8456—Structuring of content, e.g. decomposing content into time segments by decomposing the content in the time domain, e.g. in time segments
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- 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
- G06F40/284—Lexical analysis, e.g. tokenisation or collocates
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/20—Education
- G06Q50/205—Education administration or guidance
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/26—Speech to text systems
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/43—Processing of content or additional data, e.g. demultiplexing additional data from a digital video stream; Elementary client operations, e.g. monitoring of home network or synchronising decoder's clock; Client middleware
- H04N21/435—Processing of additional data, e.g. decrypting of additional data, reconstructing software from modules extracted from the transport stream
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/43—Processing of content or additional data, e.g. demultiplexing additional data from a digital video stream; Elementary client operations, e.g. monitoring of home network or synchronising decoder's clock; Client middleware
- H04N21/44—Processing of video elementary streams, e.g. splicing a video clip retrieved from local storage with an incoming video stream, rendering scenes according to MPEG-4 scene graphs
- H04N21/44008—Processing of video elementary streams, e.g. splicing a video clip retrieved from local storage with an incoming video stream, rendering scenes according to MPEG-4 scene graphs involving operations for analysing video streams, e.g. detecting features or characteristics in the video stream
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/47—End-user applications
- H04N21/472—End-user interface for requesting content, additional data or services; End-user interface for interacting with content, e.g. for content reservation or setting reminders, for requesting event notification, for manipulating displayed content
- H04N21/47205—End-user interface for requesting content, additional data or services; End-user interface for interacting with content, e.g. for content reservation or setting reminders, for requesting event notification, for manipulating displayed content for manipulating displayed content, e.g. interacting with MPEG-4 objects, editing locally
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Multimedia (AREA)
- Business, Economics & Management (AREA)
- Signal Processing (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Tourism & Hospitality (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Human Computer Interaction (AREA)
- Strategic Management (AREA)
- Educational Technology (AREA)
- Software Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Educational Administration (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Acoustics & Sound (AREA)
- Economics (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- General Business, Economics & Management (AREA)
- Medical Informatics (AREA)
- Databases & Information Systems (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to the technical field of intelligent education, and discloses a method and a system for automatically decomposing classroom teaching links, wherein the method comprises the following steps: defining teaching type classification and teaching organization links corresponding to the classification, wherein each teaching organization link has attribute characteristics; the audio signal in the teaching process is converted into a teaching text record, and the whole video teaching process is recorded in a text form; based on the teaching text record, the teaching type determination and the text content feature extraction of the teaching organization link are completed; and positioning the starting time and the ending time of the corresponding organization links in the corresponding teaching video by utilizing text contents of the links in the teaching organization links, and completing the extraction of video contents in the links of the teaching organization by utilizing video editing and editing technology to form corresponding video clips so as to complete the labeling of the video clip contents. Compared with the prior art, the method and the device for constructing the video clip according to the actual teaching organization links efficiently provide teaching videos of specific teaching organization links for teaching activity analysis and research.
Description
Technical Field
The invention relates to the technical field of intelligent education, in particular to a method and a system for automatically decomposing classroom teaching links.
Background
Classroom teaching quality is the subject of primary attention of teaching research, and various teaching activities of teacher organization in the teaching process are main ways for cultivating learning interests of students and improving learning quality. In the course of teaching in the classroom, the teacher plans all kinds of teaching activity links in the classroom through the organization, and the guide student classroom study is based on the development of teaching activity link, and the content organization of the teaching and learning of every teaching activity link all influences the quality and the learning efficiency of classroom teaching quality. The video recording and broadcasting of classroom teaching is a longer video, how to automatically analyze and decompose teaching organization links from the teacher teaching video process, and output various teaching activity short videos based on the teaching organization links for teaching scientific research and classroom teaching quality analysis, which is a problem to be solved urgently.
At present, the comprehensive promotion of education informatization is realized by means of modern information technology (video recording and broadcasting, voice recognition, natural language processing, video analysis and the like), and the video recording and broadcasting of the teaching process of a teacher is realized, and the video segment decomposition and the quantitative analysis of the content of each teaching organization link in the teaching process in the teaching video of the teacher are used for helping the teacher to actively optimize and adjust the organization links and the classroom behaviors of the teaching of the classroom, so that the teaching quality of the classroom and the learning efficiency of students are improved. Meanwhile, through analysis of the video clips of all the organization links of the classroom teaching of the excellent teacher, the implementation method of the classroom teaching activity organization and the teaching task of the excellent teacher can be learned, and the organization implementation level of the teacher teaching is improved.
The speech recognition technology is that a machine converts a speech signal into text through a recognition and understanding process, and four methods are commonly used in the speech recognition technology: 1. linguistic and acoustic based methods, 2, stochastic modeling, 3, methods using artificial neural networks, 4, probabilistic grammar analysis. The most dominant method is the stochastic model method.
Natural Language Processing (NLP) is a branch of Artificial Intelligence (AI) that enables computers to understand, generate, and process human language. Modern Natural Language Processing (NLP) is highly dependent on this artificial intelligence technique of machine learning. Machine learning can generalize from examples in a dataset to make predictions. The data set is called training data, and the machine learning algorithm is trained by using the training data to generate a machine learning model capable of completing a target task.
Video analysis techniques use computer image visual analysis techniques to analyze and track objects within a video scene by separating the background from the objects in the scene. And the system can automatically trigger linkage related execution actions once the target has the behavior of the predefined rule in the scene by presetting different rules in the scenes of different videos according to the analysis module of the user.
The video slicing is to slice the video stream into a series of short video files conforming to the rule setting according to the preset rule, and generate an index file corresponding to the short video, index record the short video file searching environment parameters, and provide support for the subsequent searching of the short video.
Disclosure of Invention
The invention aims to: aiming at the problems existing in the prior art, the invention provides a method and a system for automatically decomposing the teaching links of a teacher, which realize the deconstruction of video clips according to the actual teaching organization links, and efficiently provide teaching videos of specific teaching organization links for teaching activity analysis and research, thus being particularly critical to the research of the teaching organization process of the teacher and improving the teaching quality of the teacher in the classroom.
The technical scheme is as follows: the invention provides a method for automatically decomposing classroom teaching links, which comprises the following steps:
step 1: defining teaching type classification and teaching organization links corresponding to the classification, wherein each teaching organization link has attribute characteristics;
step 2: the audio signal in the teaching process is converted into a teaching text record, and the whole video teaching process is recorded in a text form;
step 3: based on the teaching text record, the teaching type determination and the text content feature extraction of the teaching organization link are completed;
step 4: and positioning the starting time and the ending time of the corresponding organization links in the corresponding teaching video by utilizing text contents of the links in the teaching organization links, and completing the extraction of video contents in the links of the teaching organization by utilizing video editing and editing technology to form corresponding video clips so as to complete the labeling of the video clip contents.
Further, the class types defined in the step 1 include a new teaching class, a review class, and a comment class; the teaching organization links of the review lessons comprise problem driving- > important difficult discussion- > knowledge carding- > classical evaluation- > consolidation perfection; the attribute characteristics of each teaching organization link are characteristic keywords used in the implementation process of the teaching organization link.
Further, the step 2 of converting the audio signal of the teaching process into the teaching text record specifically includes the following steps:
step 2.1: video decoding, namely, restoring video data into an original image frame sequence by decoding a video file;
step 2.2: in the process of video decoding, extracting audio data in video, wherein the audio data exists in a specific audio coding format and is stored as an independent audio file;
step 2.3: the audio preprocessing, which includes removing noise, reducing the influence of background music and balancing volume, is carried out on the audio data before the audio file is further processed;
step 2.4: voice recognition, which converts audio data into characters by using a voice recognition technology;
step 2.5: post-processing the text, namely, after converting the text into the text, performing post-processing operation on the text, wherein the post-processing comprises the steps of removing repeated content, correcting errors and adding punctuation marks; and after the word post-processing is finished, forming a text record of the classroom teaching.
Further, the specific process of completing the teaching type determination and the text content feature extraction of the teaching organization link based on the teaching text record in the step 3 is as follows:
step 3.1, preprocessing a classroom teaching text, processing text information, converting the text into a format suitable for a machine learning algorithm to process, wherein the text comprises word segmentation, stop word removal, part-of-speech tagging and word desiccation;
step 3.2, teaching text representation, wherein after the teaching document is segmented, the text forms a set of characteristic words, and a vector space model VSM is adopted to express the document into teaching text representation;
step 3.3, extracting characteristics of teaching texts, extracting characteristics of text classification training and testing, and reducing characteristic dimensions;
step 3.4, characteristic weight distribution, namely, carrying out weight distribution on characteristic words of each teaching type and teaching organization links for carrying out subsequent classification training and teaching type classification;
step 3.5, teaching text classification training and teaching type classification; the classification of the teaching type is based on the identification classification of the teaching text content, and the classifier model of the teaching type is obtained through training by screening teaching organization link characteristics existing in the teaching text, so that the automatic classification of the teaching type based on the teaching text is finally realized;
and 3.6, outputting a teaching organization link characteristic text, matching and classifying the teaching text according to the teaching class definition standard and the organization link characteristic corresponding to the teaching classifier model, and outputting link text content information in the teaching organization link.
Further, the specific operations in the step 3.2 include:
the teaching text set is expressed in a matrix form, namely, one row represents a document and one column represents a characteristic word; assuming that the set of teaching texts is denoted as D (D1, D2, D3 … dn), the feature space is denoted as T (T1, T2, T3 … tn), where tn represents one dimension of all features in the space vector, the index term is called a feature term, and one teaching text di is denoted as an n-dimensional vector (W1, W2, W3 … … Wn), where Wj represents a weight value of the teaching text on the j-th feature, the weight value is a feature value of a feature term, and the weight value is obtained through a weight allocation algorithm.
Further, in the step 3.3 and the step 3.4, a vector space of the teaching text is constructed by calculating weights of vocabularies with higher occurrence frequency in the teaching text in the whole corpus by using a TF-IDF algorithm; the method comprises the following specific steps:
1) For each teaching text d, calculating word frequency TF (t, d) of each word t in the text, namely the number of times/total word number of the word in the document;
2) Calculating document frequency DF (t) of each vocabulary t in the whole teaching text library, namely the number of documents containing the vocabulary/total number of the documents, and calculating Inverse Document Frequency, namely IDF (t) =log (N/(1+DF (t))), wherein N is the total number of the documents in the teaching text library;
3) For each teaching text d, calculating TF-IDF value of each vocabulary t, i.e., TF-IDF (t, d) =tf (t, d) ×idf (t);
4) Taking the TF-IDF values of all words in each teaching text as the feature vector of the document, namely, the feature vector V (d) of the document d= [ TF-IDF (t 1, d), TF-IDF (t 2, d) ], wherein t1, t2, & tm is all words in the vocabulary;
5) Each text in the teaching text library can be represented as a feature vector, i.e., a text vector space.
Further, the teaching text classification training in the step 3.5 and the teaching type classification specifically include the following steps:
step 3.5.1, selecting a proper number of class teaching text sets of known teaching types as a training set, training a teaching classification model on the training set by using a Support Vector Machine (SVM), determining parameters and threshold values of the teaching classification model, and constructing a teaching type classifier;
step 3.5.2, applying the teaching type classifier to the test set, and verifying the performance and accuracy of the teaching classifier model;
step 3.5.3, evaluating the performance of the teaching type classifier, and evaluating the tested classification result by using an objective classification performance evaluation method and indexes;
and 3.5.4, classifying and identifying the teaching types, namely identifying the teaching types corresponding to the actual teaching texts by utilizing a teaching type classification model, and mining text characteristic information of each teaching organization link in the teaching texts.
Further, the specific format of the output result of the text feature information of each teaching organization link output in the step 3 is as follows:
{
{ text name of teaching, teaching type [ organizing links (feature quantity, …), organizing links (feature quantity, …) … ] },
{ text name of teaching, teaching type [ organizing links (feature quantity, …), organizing links (feature quantity, …) … ], … },
……
}
the teaching text names are used for naming teaching texts classified by teaching types, and the teaching types are the teaching types to which the texts belong; the organizing links in the teaching type are all teaching organizing links forming the teaching type, and the characteristic quantity in the organizing links is the characteristic quantity word segmentation of the teaching text belonging to a certain organizing link.
Further, in the step 4, the steps of extracting video content in each organization link of the lecture, and deconstructing slices of a specific organization link are as follows:
1) Feature quantity word segmentation positioning is carried out on the time stamp positioning of the feature quantity word segmentation of each organization link in the teaching text in the appearance of the teaching video, the audio file is analyzed by adding the audio file of the teaching text, the time stamp of the feature quantity word segmentation in the audio file is searched for, and a time stamp set of the feature quantity word segmentation appearance of each teaching organization link in the teaching text is formed;
2) The method comprises the steps of (1) extracting starting time and ending time of feature quantity occurrence in a feature quantity set of each organization link in a formed time stamp set of the feature quantity occurrence of each organization link in a teaching text, generating corresponding starting time and ending time of each organization link in the teaching text in a teaching video, and completing video slicing and video content extraction of each organization link from the starting time to the ending time of the teaching text by using an FFMPEG video tool;
3) The video content of the teaching organization links of the teaching video is extracted to form the video of each relevant teaching organization link, so that the video content extraction of the teaching organization links of the teaching video is completed.
The invention also discloses a system based on the automatic decomposition method of the classroom teaching links, which comprises a teaching type and teaching organization link definition module, a teaching video teaching text recording module, a teaching organization link deconstructing module in the teaching text recording, and a teaching organization link text deconstructing module corresponding to teaching video content;
the teaching type and teaching organization link definition module defines teaching organization links corresponding to the classification and classification of the teaching type and attribute characteristics of each teaching organization link;
the teaching video teaching text recording module is used for converting an audio signal in the teaching process into a teaching text record and recording the whole video teaching process in a text form;
the deconstructing module of the teaching organization link in the teaching text record is used for completing the teaching type determination and the text content characteristic extraction of the teaching organization link based on the teaching text record;
the deconstructing module of the video content corresponding to the teaching of the organization link text is used for positioning the starting time and the ending time of the corresponding organization link in the corresponding teaching video by utilizing the text content of each link in the teaching organization link, extracting the video content in each organization link of the teaching is completed by utilizing video editing and editing technology, forming the corresponding video short-film, and marking the video short-film content.
Advantageous effects
1. The teaching video content of the invention realizes deconstructing according to the video segments of the actual teaching organization links, and can efficiently provide teaching videos of specific teaching organization links for teaching activity analysis and research, which is particularly critical to the research of teacher teaching organization processes and the improvement of teaching quality of teacher classes. According to the invention, by utilizing the machine deep learning technology, the video analysis technology and other technologies, through the automatic analysis and decomposition of teaching organization links in the teaching video process of a teacher, various teaching activity short videos based on the teaching organization links are output for teaching scientific research and classroom teaching quality analysis.
2. The automatic deconstructing method for the video content of the teaching organization links based on the teaching organization links utilizes an intelligent automatic means, greatly improves deconstructing processing efficiency of video content fragments of various teaching organization links and analysis precision of the content of the teaching organization links, improves short video application scale of the teaching organization links, and provides an intelligent analysis application tool for classroom teaching scientific research and teacher class quality judgment.
Drawings
FIG. 1 is a text form output processing flow of video content of teaching of the invention;
FIG. 2 is a flow chart of teaching text training, test evaluation and teaching type classifier generation in the invention;
FIG. 3 is a feature vector representation of the teaching text of the present invention;
fig. 4 is a process of implementing text expression of the teaching text of the present invention.
Detailed Description
The invention is further described below with reference to the accompanying drawings. The following examples are only for more clearly illustrating the technical aspects of the present invention, and are not intended to limit the scope of the present invention.
The invention discloses a method for automatically decomposing a classroom teaching link, which is shown in fig. 1 and comprises the following steps:
step 1: and defining teaching type classification and teaching organization links corresponding to the classification, wherein each teaching organization link has attribute characteristics.
The defined class types comprise teaching new classes, review classes and comments classes; the teaching organization links of the review lessons comprise problem driving- > important difficulty discussion- > knowledge carding- > classical evaluation- > consolidation perfection; the teaching organization links for teaching new lessons and the teaching organization links for lecturing can be set according to actual needs. The attribute characteristics of each teaching organization link are characteristic keywords used in the implementation process of the teaching organization link.
Step 2: and converting the audio signal of the teaching process into a teaching text record, and recording the whole video teaching process in a text form.
1) Video decoding: the video file is decoded to obtain audio data therein. The implementation is accomplished through the use of video processing libraries or tools, such as FFmpeg, openCV, etc. Programming calls to these tools can open the lecture video file and extract the audio stream during the lecture through the audio decoder in the tool.
2) And (3) generating an audio file: and storing the audio data after audio decoding as an audio file. The audio data exists in a specific audio coding format (such as AAC, MP3, WAV, etc.), and the specific implementation is accomplished using an audio processing library or tool, such as Librosa, pyDub, etc., to convert the audio data into an MP3 format file.
3) Before converting the audio into text, the audio needs to be preprocessed correspondingly.
3.1 The tool Librosa is used for preprocessing the audio file, wherein the preprocessing comprises noise removal or noise reduction, and the influence of background noise is reduced. The audio may also be enhanced to improve sound clarity and audibility.
3.2 Audio voice activity detection is accomplished by the audio processing library Librosa, recognizing and marking portions of the voice's voice activity in the audio.
3.3 Using tools (e.g., soX, FFmpeg) to complete audio volume normalization, ensuring volume level uniformity and consistency of audio.
4) The voice recognition tool converts the preprocessed audio into characters to form the teaching text record file. Common speech recognition tools and services include Google Cloud Speech-to-Text, microsoft Azure Speech to Text, IBM Watson Speech to Text, and the like. Post-processing the text, namely, after converting the text into the text, performing post-processing operation on the text, wherein the post-processing comprises the steps of removing repeated content, correcting errors and adding punctuation marks; and after the word post-processing is finished, forming a text record of the classroom teaching.
Step 3: based on the teaching text record, the teaching type determination and the text content feature extraction of the teaching organization link are completed.
And 3.1, preprocessing a classroom teaching text, processing text information, converting the text into a format suitable for processing by a machine learning algorithm, wherein the format comprises word segmentation, stop word removal, part-of-speech tagging and word desiccation.
The pretreatment of the teaching text can be realized by writing a Python cleaning program, and the teaching text can also be processed by using NLP tools such as NLTK, spaCy and the like, and the cleaning work comprises the following steps:
1) Text cleaning: unstructured text content such as numbers, punctuation marks, HTML tags and the like is removed to better identify the meaning and keywords of the text content.
2) Word segmentation: a continuous text sequence is segmented into meaningful words, chinese word segmentation, english word segmentation and the like.
3) Word drying/morphological reduction: the words are converted to their basic form to eliminate morphological differences.
4) Deactivating the filter term: some common but not practical vocabulary is filtered out.
And 3.2, teaching text representation, wherein after the teaching document is segmented, the text forms a set of characteristic words, and a vector space model VSM is adopted to express the document into the teaching text representation.
The teaching text expression mainly adopts a vector space model VSM form to express the teaching text, and a teaching text set is expressed in a matrix form, namely, one row represents a document, and one column represents a characteristic word. For example, assuming that the set of lecture text is denoted D (D1, D2, D3 … dn), the feature space is denoted T (T1, T2, T3 … tn), where tn represents one dimension of all features in the space vector, these index terms are referred to as feature terms, which can clearly represent the lecture text, so that the machine can understand and process the lecture text. A teaching text di is expressed as an n-dimensional vector (W1, W2, W3 … … Wn), wherein Wj represents the weight value of the teaching text on the j-th feature, and the weight value is the feature value of a feature item and can be obtained through a weight distribution algorithm.
And 3.3, extracting characteristics of teaching texts, extracting characteristics of text classification training and testing, and reducing characteristic dimensions.
And 3.4, distributing the characteristic weights, namely distributing the weights of the characteristic words of each teaching type and the teaching organization link for subsequent classification training and teaching type classification.
Constructing a vector space of the teaching text by calculating the weight of the vocabulary with higher occurrence frequency in the teaching text in the whole corpus by using a TF-IDF algorithm; the method comprises the following specific steps:
1) For each teaching text d, the word frequency TF (t, d) of each word t, i.e. the number of times/total number of words the word appears in the document, is calculated in the text.
2) The document frequency DF (t) of each vocabulary t appearing in the whole teaching text base, i.e. the number of documents containing the vocabulary/the total number of texts, is calculated, and Inverse Document Frequency, i.e. IDF (t) =log (N/(1+df (t))), where N is the total number of documents in the teaching text base.
3) For each teaching text d, a TF-IDF value of each vocabulary t, i.e., TF-IDF (t, d) =tf (t, d) ×idf (t), is calculated.
4) The TF-IDF values of all words in each teaching text are taken as the feature vector of the document, i.e. the feature vector V (d) of the document d= [ TF-IDF (t 1, d), TF-IDF (t 2, d) ], wherein t1, t2, & gt, tm is all words in the vocabulary.
5) Each text in the teaching text library can be represented as a feature vector, i.e., a text vector space.
Step 3.5, teaching text classification training and teaching type classification; the teaching type classification is based on recognition classification of teaching text content, teaching organization link characteristics existing in the teaching text are screened, a classifier model of the teaching type is trained and obtained, and finally automatic classification of the teaching type based on the teaching text is achieved.
And 3.5.1, selecting a proper number of class teaching text sets of known teaching types as a training set, training a teaching classification model on the training set by using a Support Vector Machine (SVM), determining parameters and threshold values of the teaching classification model, and constructing a teaching type classifier. The Support Vector Machine (SVM) classification algorithm maps samples of teaching text into a high-dimensional space, and classification is achieved by finding a hyperplane in the high-dimensional space that maximizes classification boundaries.
The preprocessed features and labeled teaching text data are fed into the SVM for training, and the SVM automatically searches the hyperplane to maximize the interval between classes so as to avoid the risk of overfitting. SVM is typically used to binary classify text data (e.g., new class lead-in/non-lead-in, discussion/non-discussion, etc.), and the training process includes selecting kernel functions, determining hyper-parameters C and gamma, etc.
And 3.5.2, applying the teaching type classifier to the test set, and verifying the performance and accuracy of the teaching classifier model.
The teaching type classifier based on the SVM algorithm is evaluated by using unlabeled data or a group of test sets, and the performance of the teaching type classifier can be evaluated by using evaluation indexes such as accuracy, recall, F1 score and the like. If the lecture type classifier model does not perform well, we can adjust the parameters of the SVM until the lecture type classifier model performs best.
And 3.5.3, evaluating the performance of the teaching type classifier, and evaluating the classification result of the test by using an objective classification performance evaluation method and indexes.
And 3.5.4, classifying and identifying the teaching types, namely identifying the teaching types corresponding to the actual teaching texts by utilizing a teaching type classification model, and mining text characteristic information of each teaching organization link in the teaching texts. The teaching type classifier is used for completing the construction of an actual teaching type classifier model through inductive learning and performance evaluation, and can classify the teaching types of the teaching texts in a vector form.
And 3.6, outputting a teaching organization link characteristic text, matching and classifying the teaching text according to the teaching class definition standard and the organization link characteristic corresponding to the teaching classifier model, and outputting link text content information in the teaching organization link.
The specific format of the output result of the text characteristic information of each teaching organization link is as follows:
{
{ text name of teaching, teaching type [ organizing links (feature quantity, …), organizing links (feature quantity, …) … ] },
{ text name of teaching, teaching type [ organizing links (feature quantity, …), organizing links (feature quantity, …) … ], … },
……
}
the teaching text names are used for naming teaching texts classified by teaching types, and the teaching types are the teaching types to which the texts belong; the organizing links in the teaching type are all teaching organizing links forming the teaching type, and the characteristic quantity in the organizing links is the characteristic quantity word segmentation of the teaching text belonging to a certain organizing link.
Step 4: slicing the teaching video teaching organization links, positioning the starting time and the ending time of the corresponding organization links in the corresponding teaching video by utilizing text contents of each link in the teaching organization links, and completing the extraction of video contents in each teaching organization link by utilizing video editing and editing technology to form corresponding video shortcuts so as to complete the labeling of the video shortcuts.
And loading a classification output result of the teaching type, and extracting video contents of corresponding organization links according to the feature quantity set of each organization link. The specific tissue section deconstructing steps are as follows:
1) Feature quantity word segmentation positioning is carried out on the time stamp positioning of the feature quantity word segmentation of each organization link in the teaching text in the appearance of the teaching video, the audio file is analyzed by adding the audio file of the teaching text, the time stamp of the feature quantity word segmentation in the audio file is searched for, and a time stamp set of the feature quantity word segmentation appearance of each teaching organization link in the teaching text is formed;
2) The method comprises the steps of (1) extracting starting time and ending time of feature quantity occurrence in a feature quantity set of each organization link in a formed time stamp set of the feature quantity occurrence of each organization link in a teaching text, generating corresponding starting time and ending time of each organization link in the teaching text in a teaching video, and completing video slicing and video content extraction of each organization link from the starting time to the ending time of the teaching text by using an FFMPEG video tool;
3) The video content of the teaching organization links of the teaching video is extracted to form the video of each relevant teaching organization link, so that the video content extraction of the teaching organization links of the teaching video is completed.
And finally accessing the teaching organization link video, completing the label construction of the teaching organization link slice video under specific teaching types according to the attribute characteristics of the teaching organization link video segments, wherein the link slice label contains the information of teaching types, course names, teaching links, teaching persons and the like, and is convenient for searching the teaching organization link video segments.
The invention designs a system aiming at the automatic decomposition method of the classroom teaching links, which comprises a teaching type and teaching organization link definition module, a teaching video teaching text recording module, a deconstructing module of the teaching organization links in the teaching text recording, a deconstructing module of the teaching organization links, and a storage inquiry of the video contents of the teaching organization links, wherein the deconstructing module of the teaching video contents corresponds to the organization link text.
The teaching type and teaching organization link definition module defines teaching organization links corresponding to the classification of the teaching type and classification, and attribute characteristics of each teaching organization link.
The teaching video teaching text recording module is used for converting an audio signal in the teaching process into teaching text record and recording the whole video teaching process in a text form.
The deconstructing module of the teaching organization link in the teaching text record is used for completing the teaching type determination and the text content feature extraction of the teaching organization link based on the teaching text record.
The deconstructing module of the video content corresponding to the teaching of the organization link text is used for positioning the starting time and the ending time of the corresponding organization link in the teaching video by utilizing the text content of each link in the teaching organization link, extracting the video content in each organization link of the teaching is completed by utilizing the video editing and editing technology, forming the corresponding video short-film, and completing the video short-film content marking.
The system for automatically decomposing the classroom teaching links disclosed by the invention can be deployed in a server, and can automatically decompose the classroom teaching link videos in a batch and automatic manner.
The foregoing embodiments are merely illustrative of the technical concept and features of the present invention, and are intended to enable those skilled in the art to understand the present invention and to implement the same, not to limit the scope of the present invention. All equivalent changes or modifications made according to the spirit of the present invention should be included in the scope of the present invention.
Claims (10)
1. The automatic decomposing method for the classroom teaching links is characterized by comprising the following steps:
step 1: defining teaching type classification and teaching organization links corresponding to the classification, wherein each teaching organization link has attribute characteristics;
step 2: the audio signal in the teaching process is converted into a teaching text record, and the whole video teaching process is recorded in a text form;
step 3: based on the teaching text record, the teaching type determination and the text content feature extraction of the teaching organization link are completed;
step 4: and positioning the starting time and the ending time of the corresponding organization links in the corresponding teaching video by utilizing text contents of the links in the teaching organization links, and completing the extraction of video contents in the links of the teaching organization by utilizing video editing and editing technology to form corresponding video clips so as to complete the labeling of the video clip contents.
2. The method according to claim 1, wherein the class types defined in the step 1 include teaching new classes, review classes, and comments; the teaching organization links of the review lessons comprise problem driving- > important difficult discussion- > knowledge carding- > classical evaluation- > consolidation perfection; the attribute characteristics of each teaching organization link are characteristic keywords used in the implementation process of the teaching organization link.
3. The method for automatically decomposing a classroom teaching link according to claim 1, wherein the step 2 of converting the audio signal of the teaching process into a teaching text record specifically comprises the steps of:
step 2.1: video decoding, namely, restoring video data into an original image frame sequence by decoding a video file;
step 2.2: in the process of video decoding, extracting audio data in video, wherein the audio data exists in a specific audio coding format and is stored as an independent audio file;
step 2.3: the audio preprocessing, which includes removing noise, reducing the influence of background music and balancing volume, is carried out on the audio data before the audio file is further processed;
step 2.4: voice recognition, which converts audio data into characters by using a voice recognition technology;
step 2.5: post-processing the text, namely, after converting the text into the text, performing post-processing operation on the text, wherein the post-processing comprises the steps of removing repeated content, correcting errors and adding punctuation marks; and after the word post-processing is finished, forming a text record of the classroom teaching.
4. The method for automatically decomposing a classroom teaching link according to claim 1, wherein the specific process of completing the teaching type determination and the text content feature extraction of the teaching organization link based on the teaching text record in the step 3 is as follows:
step 3.1, preprocessing a classroom teaching text, processing text information, converting the text into a format suitable for a machine learning algorithm to process, wherein the text comprises word segmentation, stop word removal, part-of-speech tagging and word desiccation;
step 3.2, teaching text representation, wherein after the teaching document is segmented, the text forms a set of characteristic words, and a vector space model VSM is adopted to express the document into teaching text representation;
step 3.3, extracting characteristics of teaching texts, extracting characteristics of text classification training and testing, and reducing characteristic dimensions;
step 3.4, characteristic weight distribution, namely, carrying out weight distribution on characteristic words of each teaching type and teaching organization links for carrying out subsequent classification training and teaching type classification;
step 3.5, teaching text classification training and teaching type classification; the classification of the teaching type is based on the identification classification of the teaching text content, and the classifier model of the teaching type is obtained through training by screening teaching organization link characteristics existing in the teaching text, so that the automatic classification of the teaching type based on the teaching text is finally realized;
and 3.6, outputting a teaching organization link characteristic text, matching and classifying the teaching text according to the teaching class definition standard and the organization link characteristic corresponding to the teaching classifier model, and outputting link text content information in the teaching organization link.
5. The method for automatically decomposing a classroom teaching link according to claim 4, wherein the specific operations in step 3.2 include:
the teaching text set is expressed in a matrix form, namely, one row represents a document and one column represents a characteristic word; assuming that the set of teaching texts is denoted as D (D1, D2, D3 … dn), the feature space is denoted as T (T1, T2, T3 … tn), where tn represents one dimension of all features in the space vector, the index term is called a feature term, and one teaching text di is denoted as an n-dimensional vector (W1, W2, W3 … … Wn), where Wj represents a weight value of the teaching text on the j-th feature, the weight value is a feature value of a feature term, and the weight value is obtained through a weight allocation algorithm.
6. The method for automatically decomposing a classroom teaching link according to claim 4, wherein the vector space of the teaching text is constructed by calculating weights of words with higher occurrence frequency in the teaching text in the whole corpus by using a TF-IDF algorithm in the steps 3.3 and 3.4; the method comprises the following specific steps:
1) For each teaching text d, calculating word frequency TF (t, d) of each word t in the text, namely the number of times/total word number of the word in the document;
2) Calculating document frequency DF (t) of each vocabulary t in the whole teaching text library, namely the number of documents containing the vocabulary/total number of the documents, and calculating Inverse Document Frequency, namely IDF (t) =log (N/(1+DF (t))), wherein N is the total number of the documents in the teaching text library;
3) For each teaching text d, calculating TF-IDF value of each vocabulary t, i.e., TF-IDF (t, d) =tf (t, d) ×idf (t);
4) Taking the TF-IDF values of all words in each teaching text as the feature vector of the document, namely, the feature vector V (d) of the document d= [ TF-IDF (t 1, d), TF-IDF (t 2, d) ], wherein t1, t2, & tm is all words in the vocabulary;
5) Each text in the teaching text library can be represented as a feature vector, i.e., a text vector space.
7. The method for automatically decomposing a class teaching link according to claim 4, wherein the training of class classification of the teaching text and the class classification of the teaching type in step 3.5 specifically comprise the following steps:
step 3.5.1, selecting a proper number of class teaching text sets of known teaching types as a training set, training a teaching classification model on the training set by using a Support Vector Machine (SVM), determining parameters and threshold values of the teaching classification model, and constructing a teaching type classifier;
step 3.5.2, applying the teaching type classifier to the test set, and verifying the performance and accuracy of the teaching classifier model;
step 3.5.3, evaluating the performance of the teaching type classifier, and evaluating the tested classification result by using an objective classification performance evaluation method and indexes;
and 3.5.4, classifying and identifying the teaching types, namely identifying the teaching types corresponding to the actual teaching texts by utilizing a teaching type classification model, and mining text characteristic information of each teaching organization link in the teaching texts.
8. The method for automatically decomposing the classroom teaching links according to claim 1, wherein the specific format of the output result of the text feature information of each teaching organization link output in the step 3 is:
{
{ text name of teaching, teaching type [ organizing links (feature quantity, …), organizing links (feature quantity, …) … ] },
{ text name of teaching, teaching type [ organizing links (feature quantity, …), organizing links (feature quantity, …) … ], … },
……
}
the teaching text names are used for naming teaching texts classified by teaching types, and the teaching types are the teaching types to which the texts belong; the organizing links in the teaching type are all teaching organizing links forming the teaching type, and the characteristic quantity in the organizing links is the characteristic quantity word segmentation of the teaching text belonging to a certain organizing link.
9. The method for automatically decomposing a classroom teaching link according to claim 1, wherein the step 4 of extracting video content from each organization link of a lecture, and the step of deconstructing slices of a specific organization link is as follows:
1) Feature quantity word segmentation positioning is carried out on the time stamp positioning of the feature quantity word segmentation of each organization link in the teaching text in the appearance of the teaching video, the audio file is analyzed by adding the audio file of the teaching text, the time stamp of the feature quantity word segmentation in the audio file is searched for, and a time stamp set of the feature quantity word segmentation appearance of each teaching organization link in the teaching text is formed;
2) The method comprises the steps of (1) extracting starting time and ending time of feature quantity occurrence in a feature quantity set of each organization link in a formed time stamp set of the feature quantity occurrence of each organization link in a teaching text, generating corresponding starting time and ending time of each organization link in the teaching text in a teaching video, and completing video slicing and video content extraction of each organization link from the starting time to the ending time of the teaching text by using an FFMPEG video tool;
3) The video content of the teaching organization links of the teaching video is extracted to form the video of each relevant teaching organization link, so that the video content extraction of the teaching organization links of the teaching video is completed.
10. A system based on the automatic decomposing method of the classroom teaching links of any one of claims 1 to 9, characterized by comprising a teaching type and teaching organization link definition module, a teaching video teaching text recording module, a teaching organization link deconstructing module in the teaching text recording, and a teaching organization link text deconstructing module corresponding to teaching video content;
the teaching type and teaching organization link definition module defines teaching organization links corresponding to the classification and classification of the teaching type and attribute characteristics of each teaching organization link;
the teaching video teaching text recording module is used for converting an audio signal in the teaching process into a teaching text record and recording the whole video teaching process in a text form;
the deconstructing module of the teaching organization link in the teaching text record is used for completing the teaching type determination and the text content characteristic extraction of the teaching organization link based on the teaching text record;
the deconstructing module of the video content corresponding to the teaching of the organization link text is used for positioning the starting time and the ending time of the corresponding organization link in the corresponding teaching video by utilizing the text content of each link in the teaching organization link, extracting the video content in each organization link of the teaching is completed by utilizing video editing and editing technology, forming the corresponding video short-film, and marking the video short-film content.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311314047.0A CN117544831B (en) | 2023-10-11 | 2023-10-11 | Automatic decomposing method and system for classroom teaching links |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311314047.0A CN117544831B (en) | 2023-10-11 | 2023-10-11 | Automatic decomposing method and system for classroom teaching links |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117544831A true CN117544831A (en) | 2024-02-09 |
| CN117544831B CN117544831B (en) | 2024-05-07 |
Family
ID=89786880
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311314047.0A Active CN117544831B (en) | 2023-10-11 | 2023-10-11 | Automatic decomposing method and system for classroom teaching links |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117544831B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6608930B1 (en) * | 1999-08-09 | 2003-08-19 | Koninklijke Philips Electronics N.V. | Method and system for analyzing video content using detected text in video frames |
| CN106128195A (en) * | 2016-08-10 | 2016-11-16 | 中山可可网络科技有限公司 | Instant messaging teaching assisting system |
| CN109284390A (en) * | 2018-11-29 | 2019-01-29 | 北京师范大学 | A kind of teaching scene codes method based on classroom log |
| CN110910691A (en) * | 2019-11-28 | 2020-03-24 | 深圳市木愚科技有限公司 | Personalized course generation method and system |
| CN116403445A (en) * | 2023-04-12 | 2023-07-07 | 北京高思博乐教育科技股份有限公司 | Self-adaptive auxiliary teaching method and device, electronic equipment and storage medium |
-
2023
- 2023-10-11 CN CN202311314047.0A patent/CN117544831B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6608930B1 (en) * | 1999-08-09 | 2003-08-19 | Koninklijke Philips Electronics N.V. | Method and system for analyzing video content using detected text in video frames |
| CN106128195A (en) * | 2016-08-10 | 2016-11-16 | 中山可可网络科技有限公司 | Instant messaging teaching assisting system |
| CN109284390A (en) * | 2018-11-29 | 2019-01-29 | 北京师范大学 | A kind of teaching scene codes method based on classroom log |
| CN110910691A (en) * | 2019-11-28 | 2020-03-24 | 深圳市木愚科技有限公司 | Personalized course generation method and system |
| CN116403445A (en) * | 2023-04-12 | 2023-07-07 | 北京高思博乐教育科技股份有限公司 | Self-adaptive auxiliary teaching method and device, electronic equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117544831B (en) | 2024-05-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11763193B2 (en) | Systems and method for performing contextual classification using supervised and unsupervised training | |
| CN107291723B (en) | Method and device for classifying webpage texts and method and device for identifying webpage texts | |
| CN113254574A (en) | Method, device and system for auxiliary generation of customs official documents | |
| CN111639484A (en) | Method for analyzing seat call content | |
| CN112434164B (en) | Network public opinion analysis method and system taking topic discovery and emotion analysis into consideration | |
| KR20200052412A (en) | Artificial intelligence employment system and employing method of thereof | |
| CN108536673B (en) | News event extraction method and device | |
| CN115203338A (en) | Label and label example recommendation method | |
| CN112380848B (en) | Text generation method, device, equipment and storage medium | |
| Heidari et al. | Financial footnote analysis: developing a text mining approach | |
| CN117544831B (en) | Automatic decomposing method and system for classroom teaching links | |
| CN115713085A (en) | Document theme content analysis method and device | |
| CN113094547B (en) | Method for searching specific action video clip in Japanese online video corpus | |
| CN114661900A (en) | Text annotation recommendation method, device, equipment and storage medium | |
| CN113901793A (en) | Event extraction method and device combining RPA and AI | |
| Gomes Jr et al. | Framework for knowledge discovery in educational video repositories | |
| CN112579666A (en) | Intelligent question-answering system and method and related equipment | |
| Pant et al. | Automatic Software Engineering Position Resume Screening using Natural Language Processing, Word Matching, Character Positioning, and Regex | |
| Ren et al. | An Automated Goal Labeling Method Based on User Reviews. | |
| Shekhar | Text Mining and Sentiment Analysis | |
| CN113821681B (en) | Video tag generation method, device and equipment | |
| CN111949781B (en) | Intelligent interaction method and device based on natural sentence syntactic analysis | |
| CN112100314B (en) | API course compilation generation method based on software development question-answering website | |
| Chand | Lecture video segmentation using speech content | |
| CN117520570A (en) | Intelligent text auxiliary processing system oriented to policy documents |
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 | ||
| GR01 | Patent grant |