CN116523225B - Data mining-based overturning classroom hybrid teaching method - Google Patents
Data mining-based overturning classroom hybrid teaching method Download PDFInfo
- Publication number
- CN116523225B CN116523225B CN202310411785.0A CN202310411785A CN116523225B CN 116523225 B CN116523225 B CN 116523225B CN 202310411785 A CN202310411785 A CN 202310411785A CN 116523225 B CN116523225 B CN 116523225B
- Authority
- CN
- China
- Prior art keywords
- teaching
- knowledge
- skill
- students
- enterprise
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 73
- 238000007418 data mining Methods 0.000 title claims abstract description 9
- 230000001149 cognitive effect Effects 0.000 claims abstract description 37
- 238000009826 distribution Methods 0.000 claims abstract description 36
- 230000007306 turnover Effects 0.000 claims abstract description 21
- 238000013210 evaluation model Methods 0.000 claims abstract description 19
- 230000008859 change Effects 0.000 claims abstract description 11
- 230000007115 recruitment Effects 0.000 claims description 71
- 238000011156 evaluation Methods 0.000 claims description 68
- 238000012549 training Methods 0.000 claims description 25
- 238000005516 engineering process Methods 0.000 claims description 23
- 239000013598 vector Substances 0.000 claims description 21
- 238000004458 analytical method Methods 0.000 claims description 19
- 238000012360 testing method Methods 0.000 claims description 18
- 230000000694 effects Effects 0.000 claims description 16
- 238000005065 mining Methods 0.000 claims description 16
- 230000008569 process Effects 0.000 claims description 15
- 238000012216 screening Methods 0.000 claims description 12
- 238000004422 calculation algorithm Methods 0.000 claims description 9
- 238000013461 design Methods 0.000 claims description 9
- 238000002372 labelling Methods 0.000 claims description 9
- 238000012545 processing Methods 0.000 claims description 9
- 238000010801 machine learning Methods 0.000 claims description 8
- 230000019771 cognition Effects 0.000 claims description 7
- 238000013136 deep learning model Methods 0.000 claims description 7
- 239000011159 matrix material Substances 0.000 claims description 7
- 230000009466 transformation Effects 0.000 claims description 7
- 238000011426 transformation method Methods 0.000 claims description 7
- 238000004140 cleaning Methods 0.000 claims description 6
- 238000007781 pre-processing Methods 0.000 claims description 6
- 238000007670 refining Methods 0.000 claims description 6
- 230000011218 segmentation Effects 0.000 claims description 6
- 238000010206 sensitivity analysis Methods 0.000 claims description 6
- 230000003930 cognitive ability Effects 0.000 claims description 4
- 238000010586 diagram Methods 0.000 claims description 4
- 238000013527 convolutional neural network Methods 0.000 claims description 3
- 238000012937 correction Methods 0.000 claims description 3
- 230000006870 function Effects 0.000 claims description 3
- 230000006872 improvement Effects 0.000 claims description 3
- 238000013178 mathematical model Methods 0.000 claims description 3
- 238000012544 monitoring process Methods 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 3
- 238000013139 quantization Methods 0.000 claims description 3
- 238000011867 re-evaluation Methods 0.000 claims description 3
- 238000012163 sequencing technique Methods 0.000 claims description 3
- 230000008030 elimination Effects 0.000 claims description 2
- 238000003379 elimination reaction Methods 0.000 claims description 2
- 238000009472 formulation Methods 0.000 claims description 2
- 239000000203 mixture Substances 0.000 claims description 2
- 238000003058 natural language processing Methods 0.000 description 5
- 238000004891 communication Methods 0.000 description 3
- 238000011161 development Methods 0.000 description 3
- 230000018109 developmental process Effects 0.000 description 3
- 230000004927 fusion Effects 0.000 description 3
- 238000007726 management method Methods 0.000 description 3
- 238000010276 construction Methods 0.000 description 2
- 238000010411 cooking Methods 0.000 description 2
- 238000005192 partition Methods 0.000 description 2
- 238000013519 translation Methods 0.000 description 2
- 238000012795 verification Methods 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000001174 ascending effect Effects 0.000 description 1
- 238000012550 audit Methods 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000001575 pathological effect Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000012502 risk assessment Methods 0.000 description 1
- 238000012954 risk control Methods 0.000 description 1
- 230000009182 swimming Effects 0.000 description 1
Classifications
-
- 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
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
- G06Q10/06315—Needs-based resource requirements planning or analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/21—Design, administration or maintenance of databases
- G06F16/215—Improving data quality; Data cleansing, e.g. de-duplication, removing invalid entries or correcting typographical errors
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/25—Integrating or interfacing systems involving database management systems
- G06F16/254—Extract, transform and load [ETL] procedures, e.g. ETL data flows in data warehouses
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
- G06N20/10—Machine learning using kernel methods, e.g. support vector machines [SVM]
-
- 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
- G06Q10/00—Administration; Management
- G06Q10/10—Office automation; Time management
- G06Q10/105—Human resources
- G06Q10/1053—Employment or hiring
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/20—Education
- G06Q50/205—Education administration or guidance
-
- 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
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Theoretical Computer Science (AREA)
- Strategic Management (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Tourism & Hospitality (AREA)
- Entrepreneurship & Innovation (AREA)
- Data Mining & Analysis (AREA)
- Economics (AREA)
- Databases & Information Systems (AREA)
- Quality & Reliability (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- General Engineering & Computer Science (AREA)
- Educational Administration (AREA)
- Operations Research (AREA)
- Software Systems (AREA)
- Educational Technology (AREA)
- Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- General Health & Medical Sciences (AREA)
- Development Economics (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Game Theory and Decision Science (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application provides a data mining-based overturning classroom hybrid teaching method, which comprises the following steps: the method comprises the steps of excavating the change of the requirements of different posts and skills of an enterprise; converting the enterprise skill demand variation into a time-series knowledge demand distribution; digging overturning classroom teaching content information according to the cognitive demands and the operation demands respectively; marking knowledge point classification and teaching classification, and judging whether each skill can be obtained through a turnover classroom with the mixing proportion; constructing a skill ability evaluation model; calculating the capability difference before and after teaching based on the capability evaluation model; the method specifically comprises the steps of formulating a standard based on the distribution of calculated scores, analyzing whether to continue using the overturning classroom teaching of the mixing proportion according to the standard, and adjusting the mixing teaching proportion by combining with the forecasting condition of the enterprise capacity requirement.
Description
Technical Field
The invention relates to the technical field of information, in particular to a data mining-based overturning classroom hybrid teaching method.
Background
The overturning classroom is a mode that students can actively digest knowledge after learning knowledge autonomously, and teach out to give other classmates and teachers to give lessons, and can enable the students to know knowledge more deeply. The overturning classroom can promote the teaching effect, but the participants are required to have deep knowledge so as to explain the teaching to others. Students are not trained in lectures and may understand knowledge of the students better, but for listeners, effects may be pathological, knowledge, content and the like, and the explanation effect is difficult to compare with the explanation effect of a formal teacher after multiple training and teaching culture. Therefore, what courses can become a turning classroom, what turning subjects are actually unsuitable, and what courses can only be explained by adopting normal courses to improve the student capacity, so that how to discriminate course differences is an important problem. The ability of students can be improved by turning over the classroom, and only the improved ability is dynamically evaluated, enterprises have the ability requirement, and only the ability adapting to the enterprise requirement can finally give play to skills, so that the final value is realized. However, the demands of enterprises are variable, the demands on talent ability are also variable, and the talent ability is constantly variable along with the learning and practice of courses. Therefore, how to pass the learning course through the evaluation of enterprises, the evaluation of a fusion school or a training mechanism is finally converted into the design content of the overturning class, wherein the fusion and calculation from the requirement to the relation of teaching are a set of methods which need big data to carry out mining analysis so as to obtain accurate design.
Disclosure of Invention
The invention provides a data mining-based overturning classroom hybrid teaching method, which mainly comprises the following steps:
the method comprises the steps of excavating the change of the requirements of different posts and skills of an enterprise; converting the enterprise skill requirement variation into a time-series knowledge requirement distribution, wherein the converting the enterprise skill requirement variation into the time-series knowledge requirement distribution specifically comprises: excavating enterprise recruitment operation type and cognition type requirements, and predicting enterprise operation type and cognition type requirements; digging overturning classroom teaching content information according to the cognitive demands and the operation demands respectively; the method comprises the steps of marking knowledge point classification and teaching classification, judging whether each skill can be obtained through the overturning class with the mixing proportion, marking knowledge point classification and teaching classification, and judging whether each skill can be obtained through the overturning class with the mixing proportion, wherein the method specifically comprises the following steps: word labeling is carried out, a domain knowledge structure is obtained, and whether knowledge requirements are matched with teaching contents or not is judged; building a skill ability evaluation model, wherein the building of the skill ability evaluation model specifically comprises the following steps: screening keywords of knowledge demand dimensionality required for recruitment, evaluating the cognitive type skill ability of students according to skill ability evaluation index design questions, evaluating the operation type skill ability of the students by combining an image recognition technology and a voice recognition technology, and comprehensively evaluating the skill ability of the students by adopting a fuzzy transformation method; calculating the capability difference before and after teaching based on the capability evaluation model; according to upset classroom teaching effect adjustment teaching type, specifically include: and (3) formulating a standard based on the distribution of the calculated scores, and adjusting the mixed teaching proportion according to whether the overturning classroom teaching of the mixed proportion is continuously used or not according to standard analysis and combining with the enterprise capacity demand prediction condition.
Further optionally, the mining the enterprise for different post demand quantity and skill demand variation includes:
the enterprise recruitment basic data comprise post names, quantity, post contents and text data required by posts, and data contained in each piece of recruitment information are obtained through recruitment websites and APP; aiming at operation type requirements and cognitive type requirements, collecting enterprise recruitment basic data by taking month as a unit to form data of requirement variation; performing data preprocessing by using an ETL method, and reserving text data containing post names, quantity, post contents and post requirements; the text data of the post requirements and post contents are subjected to word segmentation processing by utilizing the ETL method and the processed data and utilizing a Jieba library; data cleaning is carried out on text data after word segmentation, stop words and the like are removed, a part-of-speech list is distributed to each word through a POS (point of sale) tag by using a dictionary, and the list is classified into the part-of-speech of each word by using an ambiguity elimination rule; based on the part-of-speech classification of each word, the skill requirement feature words of each post are extracted by using an NLP method, and the feature words are textual into vectors of N dimensions.
Further optionally, the converting the enterprise skill requirement variation into a time-series knowledge requirement profile includes:
Acquiring a knowledge system of recruitment information of each specific occupation post by utilizing a Web text mining technology, wherein the occupation capability requirement, the curriculum occupation capability target and the knowledge target of each specific occupation post need a knowledge system with corresponding modules to match the curriculum teaching content; aiming at different enterprises recruited each year, counting keyword frequency by taking years as a unit, and drawing a variation demand distribution diagram of a knowledge time sequence; comprising the following steps: excavating the recruitment operation type and the cognitive type requirements of enterprises; predicting enterprise operation class and cognitive class requirements;
the mining of enterprise recruitment operation class and cognitive class requirements specifically comprises:
the method comprises the steps of dividing N-dimensional vectors textified by feature words in the scheme into a training set and a test set by using an SVM algorithm and using post class names as labels, taking 70% as the training set and 30% as the test set, and then adopting an SVM based on a Gaussian radial basis function to carry out machine learning classification. And (3) for the results of machine learning classification through the SVM, a Web text mining technology is applied to acquire the knowledge systems of each specialty one by one, classification comprehensive analysis among posts is carried out according to the processes, and a distribution system of specific posts under the large posts is constructed.
The predicting the operation type and the cognitive type demands of the enterprise specifically comprises the following steps:
the collected feature words are text into quantitative data converted from N-dimensional vectors, and the quantitative data is input as training data. And establishing a prediction model by using a lightGBM algorithm, traversing training data, calculating the accumulated statistic of each discrete value in the histogram, and traversing and searching the optimal partition point according to the discrete value of the histogram when the feature selection is carried out, so as to obtain a final prediction result. And determining main parameters of the prediction model according to the training data, inputting the data into the prediction model for comprehensive operation, and obtaining a prediction result by adjusting the parameters.
Further optionally, the mining the information of the teaching content of the overturning class according to the cognitive requirement and the operation requirement includes:
from two angles of operation class requirements and cognition class requirements, using 'turning class application', 'turning class practice' and the like as keywords, and collecting information of teaching in which knowledge fields the turning class is currently applied to; based on the collected information of teaching of each subject knowledge field, collecting course names and knowledge point names of the subject knowledge field, and establishing a professional classification word stock so as to expand the established course word stock; data cleaning is carried out on the collected information of the overturning class, text analysis is carried out on the screened information through NLP, and a word stock of overturning class teaching content is built; based on the established word stock of the overturning class contents, the teaching content characteristic words of each overturning class are extracted and used as specific teaching contents.
Further optionally, the marking the classification of knowledge points and the classification of teaching, and the judging whether each skill can be obtained through the turning class with the mixing ratio comprises:
the N-dimensional word vector which is formed by the text of the feature words comprises feature words of recruitment required knowledge points, the feature words of the recruitment required knowledge points are matched in a word stock of teaching contents one by one, and feature words which can be matched accurately and feature words which cannot be matched accurately are obtained; scanning a word library of the teaching content by using an Apriori algorithm for the feature words which cannot be matched accurately, counting the occurrence times of a first-level club option set, eliminating candidate sets which do not meet the conditions, generating a second-level candidate set on a first-level frequent set, scanning the data rate, counting the occurrence times of the second-level candidate set, sequentially circulating, and finally mining the association degree of the required knowledge points and the teaching content in the field; if the knowledge characteristic words corresponding to the required skills can find the matched words in the word stock, a proper overturning classroom can be found for teaching, and the matched words are found in the overturning classroom, otherwise, the overturning classroom teaching mode needs to be optimized for re-evaluation; comprising the following steps: word labeling is carried out, and a domain knowledge structure is obtained; judging whether the knowledge requirement is matched with the teaching content or not;
The word labeling is carried out, and the domain knowledge structure is obtained, which specifically comprises the following steps:
and searching data of professions related to the overturning classroom, and collecting professional knowledge text information. And processing the collected professional knowledge text information by using text analysis, marking the terms of different professional knowledge, establishing a professional knowledge system word stock, and obtaining a domain knowledge structure.
Judging whether the knowledge requirement is matched with the teaching content or not specifically comprises the following steps:
and analyzing the matching degree between the overturning classroom teaching content and the enterprise recruitment knowledge demand, and taking the similarity between the capacity demand word vector and the word vector of the teaching content as the matching degree between texts, so as to establish a matching model. If the overturning classroom teaching content is successfully matched with the enterprise recruitment knowledge requirement, the specific numerical value is used for marking. If the overturn classroom teaching content is not successfully matched with the enterprise recruitment knowledge requirement, the overturn classroom teaching content is marked by another specific numerical value.
Further optionally, the constructing a skill ability evaluation model includes:
determining skill capability to be evaluated and corresponding knowledge requirements according to recruitment requirements, and dividing the knowledge requirements corresponding to the skill capability required by recruitment into three dimensions of general basic knowledge, business domain knowledge and professional front domain knowledge; the method comprises the steps of formulating a student skill ability evaluation index by screening keywords in three dimensions of general basic knowledge, business field knowledge and professional front field knowledge; the cognitive skills are evaluated by designing different evaluation questions, objectively and accurately reflecting the cognitive ability and knowledge level of the recruiter; collecting learning data of students in a learning process through images and voice forms, analyzing the learning data of the students, and evaluating the operation type skill ability of the students; aiming at the characteristic of ambiguity of student ability level evaluation, a fuzzy transformation evaluation method is adopted, various evaluation indexes are comprehensively considered, a student skill ability evaluation model is constructed according to the data characteristics of students, and comprehensive scores of the students are obtained by inputting cognitive skill ability scores and operation skill ability scores of the students corresponding to three dimensions of innovation, application and basic knowledge memory; comprising the following steps: screening keywords of knowledge requirement dimension required by recruitment; according to the skill ability evaluation index design questions, evaluating the cognitive type skill ability of the students; performing student operation type skill capability evaluation by combining an image recognition technology and a voice recognition technology; comprehensively evaluating the skills of students by adopting a fuzzy transformation method;
The screening of keywords of the knowledge requirement dimension required for recruitment specifically comprises the following steps:
screening keywords related to student skill ability according to the enterprise recruitment requirement list; classifying and integrating the screened keywords to form a comprehensive keyword list covering three dimensions of general basic knowledge, business domain knowledge and professional front domain knowledge; and selecting keywords as skill ability evaluation indexes, and defining weights.
The method for evaluating the cognitive type skill ability of the students according to the skill ability evaluation index design questions specifically comprises the following steps:
keywords of enterprise recruitment knowledge demands are collected from three dimensions respectively, the keywords are used as index words, and qualitative evaluation of questions is planned by combining teaching experience and machine recommendation; normalizing based on the enterprise recruitment knowledge demand frequency distribution; after defining the weight for each index, refining the evaluation standard and the grading rule, and recursively judging and setting the standard score for the corresponding topic of each keyword from 0 to 10 by using an on-line or off-line test. And after scoring the operations corresponding to all the keywords, calculating an ensemble average score, and finally obtaining the cognitive type skill ability score of each dimension of the student.
The method for evaluating the skill capability of the student operation type by combining the image recognition technology and the voice recognition technology specifically comprises the following steps:
after the approval of the school side and the teacher and student is obtained, the classroom image and the classroom voice of the student are collected through a camera or other equipment; keywords of enterprise recruitment knowledge demands are collected from three dimensions respectively, classroom images and classroom voices corresponding to the keywords are collected, image and voice data of actual operation processes of students in different scenes are obtained, and the students are labeled. Feature information about the student's course of operation is extracted from the original image using preprocessing techniques. The deep learning model convolutional neural network is used to automatically extract specific features of specific operations and skills. And extracting the characteristic information of the student voice from the voice signal by using the Mel frequency cepstrum coefficient. Training the image and voice data of the students by using the deep learning model to train out a model for classifying the operation types and skills of the students. In monitoring and recording the student's operations and voices, models are used to identify and evaluate the student's skills and types of operations. After defining the weight for each operation type, refining the evaluation standard and the grading rule, and recursively judging and setting the standard score for the operation corresponding to each keyword from 0 to 10 by using an online or offline test. And after scoring the operations corresponding to all the keywords, calculating an ensemble average score, and finally obtaining the operation type skill ability scores of the students in each dimension.
The comprehensive evaluation of student skill ability is carried out by adopting a fuzzy transformation method, which comprises the following steps:
setting a comprehensive evaluation dimension set R and a grading set Z; integrating the grading set of each student according to the columns to obtain a fuzzy matrix R with three dimensions of each student; multiplying the weight value by a fuzzy matrix, and performing fuzzy transformation to obtain a final comprehensive evaluation value; verifying and using newly collected data to prove talent demand prediction results of enterprises, wherein the judgment content mainly comprises whether the prediction is based on sufficiency, whether a prediction method is proper and scientific, whether the prediction results are built reasonably by reliable post academic specifications, and whether the prediction results are reliable; if the prediction result does not accord with the actual result, repeating the prediction process, and adjusting the mathematical model to finally obtain an accurate and reliable prediction result. And for the enterprise skill prediction part, collecting the existing data as a test set to carry out fitting prediction, comparing the result of the prediction by using the method, and judging whether the result meets the requirements or not, otherwise, carrying out prediction again.
Further optionally, the calculating the capability difference before and after teaching based on the capability evaluation model includes:
the method comprises the steps that before the overturning classroom teaching is used, the students are respectively subjected to capability evaluation by texts, images and voices after the overturning classroom teaching is finished; calculating the score of each student before the use of the turnover classroom teaching in three dimensions of innovation, application and basic knowledge memorization and the score after the completion of the turnover classroom teaching, and recording evaluation score data; calculating the comprehensive score of the capability evaluation of each student before using the overturning classroom teaching and the comprehensive score of the capability evaluation after finishing the overturning classroom teaching by using the constructed skill capability evaluation model; and calculating the comprehensive score difference value of the capability evaluation of the students before and after the overturning classroom teaching.
Further optionally, the adjusting the teaching type according to the overturning classroom teaching effect includes:
based on the calculated capability difference values of students before and after teaching, carrying out longitudinal and transverse comparison analysis of probability and score values; comparing and analyzing from the angles of combination of the weight values of the three dimensions of innovation, application and knowledge memory of the capability requirement and the frequency distribution conditions of the three dimensions before and after the teaching of students, and determining the teaching effect; combining the fitting condition of frequency distribution of talent demand variation of enterprises, if the deviation degree exceeds a preset threshold value, predicting again, and adjusting the content of classroom learning according to a new prediction result; comprising the following steps: formulating a criterion based on the distribution of the calculated scores; according to standard analysis, whether to continue using the overturning classroom teaching of the mixing proportion; adjusting the mixed teaching proportion by combining the capability demand prediction condition of the enterprise;
the distribution formulation standard based on the calculated score specifically comprises the following steps:
and converting the calculated scores into probability matrixes of different dimension scores of each overturning class. Multiplying the weight value of each dimension by the probability vector, and carrying out normalization operation to obtain the score distribution condition of different dimensions of students in the overturning classroom. And dividing the standard of improving the capacity according to the score distribution condition, and taking the fact that the students with the preset threshold number progress by more than a preset threshold score as the standard of improving the capacity of the students in the class.
Whether the overturning classroom teaching of the mixing proportion is continuously used or not is analyzed according to the standard, and the overturning classroom teaching specifically comprises the following steps:
if the probability of the achievement is greater than or equal to the teaching target, the teaching through the turnover classroom can be considered to meet the skill requirement of recruiting enterprises in the corresponding knowledge field. If the probability of the achievement is smaller than the teaching target, the skill required by the recruitment enterprise is not successfully obtained through the study of the turnover class, and correction can be performed by referring to the turnover class teaching mode which reaches the standard in the same field or other teaching modes can be adopted. And carrying out sensitivity analysis on the proportion of each mode in the mixed teaching, sequencing the final result of the sensitivity analysis, and selecting the optimal result for adjustment.
The method for adjusting the mixed teaching proportion by combining the capability demand prediction condition of the enterprise specifically comprises the following steps:
and obtaining a teaching strategy recommendation result according to the test student performance and the post-class operation condition. Fitting is carried out on the enterprise capacity requirement and the student capacity improvement condition respectively, the whole student capacity index is used as an input layer, quantization and grading are carried out, the whole student capacity index is used as an offline model for training, the whole student capacity index is adjusted according to the enterprise capacity requirement, an online model is updated, and finally the adjustment of the teaching strategy is changed. Finally, the proportion of each learning mode in the mixed teaching is adjusted, and the overall teaching strategy recommendation result with the proportion determined in each teaching mode is obtained.
The technical scheme provided by the embodiment of the invention can have the following beneficial effects:
the invention can predict talent demands of enterprise recruitment through a machine learning method, and adjust the knowledge type of the turnover class to be taught according to the demands of enterprises on knowledge by combining with the existing course system. And dynamically adjusting the ability evaluation state according to talent learning results, including examination achievements and classroom performances, and finally feeding back to the enterprise. Further, according to the change of talent demands of enterprises, the teaching content and the overturning class form are changed.
Drawings
Fig. 1 is a flow chart of a data mining-based overturning classroom hybrid teaching method.
Fig. 2 is a schematic diagram of a data mining-based overturning classroom hybrid teaching method.
Detailed Description
For a further understanding of the present invention, the present invention will be described in detail with reference to the drawings and examples. The present application is described in further detail below with reference to the drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the invention and are not limiting of the invention. It should be noted that, for convenience of description, only the portions related to the invention are shown in the drawings.
The embodiment of the overturning classroom mixing teaching method based on data mining specifically comprises the following steps:
step 101, the change of the different post demands and the skill demands of the enterprise is excavated.
The enterprise recruitment basic data comprise post names, quantity, post contents and text data required by posts, and data contained in each piece of recruitment information are obtained through recruitment websites and APP; aiming at operation type requirements and cognitive type requirements, collecting enterprise recruitment basic data by taking month as a unit to form data of requirement variation; performing data preprocessing by using an ETL method, and reserving text data containing post names, quantity, post contents and post requirements; the text data of the post requirements and post contents are subjected to word segmentation processing by utilizing the ETL method and the processed data and utilizing a Jieba library; and (3) data cleaning is carried out on the text data after word segmentation, stop words and the like are removed, a part-of-speech list is allocated to each word by using a dictionary through POS marking, and the list is classified into the part-of-speech of each word by using a disambiguation rule. Based on the part-of-speech classification of each word, the skill requirement feature words of each post are extracted by using an NLP method, and the feature words are textual into vectors of N dimensions. For example, the post name is "accounting" post, then count the post of the same name to get the post quantity, draw the literal description paragraph of the post content, the literal paragraph that the post requires separately at the same time. The operation class needs swimming, and the cognitive class needs such as etiquette learning. Deleting the data of the text description paragraph with no post content or post requirement as missing data. The text paragraph content, such as 'classmates with English communication and translation capability' is decomposed into words with 'English', 'communication', and the like, and finally keywords 'English', 'communication' and 'translation' are extracted.
Step 102, converting the enterprise skill requirement variation into a time-series knowledge requirement distribution.
And acquiring a knowledge system of recruitment information of each specific occupation post by utilizing a Web text mining technology, wherein the occupation capability requirement, the curriculum occupation capability target and the knowledge target of each specific occupation post need a knowledge system of a corresponding module to match the curriculum teaching content. And counting keyword frequency by taking years as a unit for different enterprises recruiting each year, and drawing a variation demand distribution diagram of a knowledge time sequence. The market demand change and the specific requirements on talents are known, professional capacity demand analysis is carried out, professional capacity demands of specific professional posts are analyzed and confirmed, a talent cultivation scheme is adjusted according to enterprise demands, professional cultivation targets are adjusted to enterprise demands, requirements of local economic construction and social development can be met, and a large number of qualified talents are cultivated for industries and enterprises. The education in the turnover classroom mode needs to establish a professional education cultivation target and talent ability cultivation scheme which are suitable for professional posts, take the ability as a home position, and aim to cultivate practical and skill talents, and cultivate high quality and skill innovation talents facing the first line of production.
And excavating the recruitment operation type and the cognitive type requirements of the enterprises.
The method comprises the steps of dividing N-dimensional vectors textified by feature words in the scheme into a training set and a test set by using an SVM algorithm and using post class names as labels, taking 70% as the training set and 30% as the test set, and then adopting an SVM based on a Gaussian radial basis function to carry out machine learning classification. And (3) for the results of machine learning classification through the SVM, a Web text mining technology is applied to acquire the knowledge systems of each specialty one by one, classification comprehensive analysis among posts is carried out according to the processes, and a distribution system of specific posts under the large posts is constructed. And (3) performing dimension reduction processing, namely selecting the features, and leaving the feature items with strong classification capability and removing the feature items with weak capability. In the text classification problem, there are samples which are not linearly separable, a decision surface cannot be directly obtained, and the SVM maps a sample set which is linearly separable in a low-dimensional space to enable the samples to obtain the characteristic of being linearly separable in a high-dimensional feature space, and an optimal hyperplane is obtained in the high-dimensional feature space, so that the classification of the samples is finally realized. For example, the collected post information includes "accounting" and "audit" both belonging to the group of financial posts, so that the specific posts are classified in a large direction. For example, the knowledge system of "accounting" is obtained for economy, management and financial winter management.
And predicting the operation type and cognitive type requirements of the enterprise.
The collected feature words are text into quantitative data converted from N-dimensional vectors, and the quantitative data is input as training data. And establishing a prediction model by using a lightGBM algorithm, traversing training data, calculating the accumulated statistic of each discrete value in the histogram, and traversing and searching the optimal partition point according to the discrete value of the histogram when the feature selection is carried out, so as to obtain a final prediction result. And determining main parameters of the prediction model according to the training data, inputting the data into the prediction model for comprehensive operation, and obtaining a prediction result by adjusting the parameters. For example, a lightGBM model is built with the debugged parameter values, the prediction results are output, the ROC value of the output results is 0.833386, the first result of the privateloaderboard rank is exceeded (0.829072), and the final prediction results are output.
And step 103, excavating overturning classroom teaching content information according to the cognitive demands and the operation demands respectively.
From two angles of operation class requirements and cognition class requirements, the information of which knowledge fields the turning class is currently applied to for teaching is collected by taking turning class application, turning class practice and the like as keywords. Based on the collected information of teaching of each subject knowledge field, collecting course names and knowledge point names of the subject knowledge field, and establishing a professional classification word stock so as to expand the established course word stock. And cleaning the data of the collected information of the overturning class, performing text analysis on the screened information through NLP, and establishing a word stock of the overturning class teaching content. Based on the established word stock of the overturning class contents, the teaching content characteristic words of each overturning class are extracted and used as specific teaching contents. For example, in the existing overturn class of accounting profession, course introduction is "basic theory mainly introducing financial accounting and method for creating financial report", and after analysis, the word library "basic theory", "financial report creation" of the overturn class teaching content is created.
And 104, marking knowledge point classification and teaching classification, and judging whether each skill can be obtained through the overturning class with the mixing proportion.
The N-dimensional word vector which is formed by the text of the feature words comprises feature words of recruitment required knowledge points, the feature words of the recruitment required knowledge points are matched in a word stock of teaching contents one by one, and feature words which can be matched accurately and feature words which cannot be matched accurately are obtained; and scanning a word library of the teaching content by using an Apriori algorithm for the feature words which cannot be matched accurately, counting the occurrence times of the first-level club option sets, eliminating the candidate sets which do not meet the conditions, generating a second-level candidate set on the first-level frequent set, scanning the data rate, counting the occurrence times of the second-level candidate set, sequentially circulating, and finally mining the association degree of the required knowledge points and the teaching content in the field. If the knowledge characteristic words corresponding to the required skills can find the matched words in the word stock, a proper overturning classroom can be found for teaching, and the matched words are found in the overturning classroom, otherwise, the overturning classroom teaching mode needs to be optimized for re-evaluation. For example, after recruitment information is decomposed, including "make financial report", the words are stored in the teaching content word stock of the accounting course, and then it can be judged that the skill can be obtained through learning in the turnover class. If the matching of multiple courses is successful, a course can be selected for verification, and if the course can not cultivate the skill ability corresponding to the course of the student, the next course is selected for verification. For example, teaching mode optimization includes highlighting professional guidance, improving the proportion of each part of the course, and the like.
And carrying out word labeling and obtaining a domain knowledge structure.
And searching data of professions related to the overturning classroom, and collecting professional knowledge text information. And processing the collected professional knowledge text information by using text analysis, marking the terms of different professional knowledge, establishing a professional knowledge system word stock, and obtaining a domain knowledge structure. The knowledge structure refers to the constitution and combination mode of a knowledge system owned by a person after special learning and training, and the reasonable knowledge structure, namely, the knowledge structure has not only profound special knowledge, but also a broad knowledge surface, has the most reasonable and optimized knowledge system actually required by the development of enterprises, establishes the reasonable knowledge structure, cultures the scientific thinking mode, and the reasonable knowledge structure is a necessary condition for serving as the professional post of the modern society. The practical skill of the user is improved to meet the requirements of the professional post in the future society, the professional post of the modern society is required to have a reasonable knowledge structure, various learned knowledge can be scientifically combined according to the specific requirements of the current social development and the professional, and the knowledge structure becomes talents adapting to the social requirements, so that the construction of the domain knowledge structure is very important for meeting the recruitment requirements of enterprises. For example, the technical knowledge terms of accounting professions include "cost management", "business forecast", and the audited professional knowledge terms include "risk assessment", so as to construct a knowledge structure of financial accounting fields.
And judging whether the knowledge requirement is matched with the teaching content.
And analyzing the matching degree between the overturning classroom teaching content and the enterprise recruitment knowledge demand, and taking the similarity between the capacity demand word vector and the word vector of the teaching content as the matching degree between texts, so as to establish a matching model. If the overturning classroom teaching content is successfully matched with the enterprise recruitment knowledge requirement, the specific numerical value is used for marking. If the overturn classroom teaching content is not successfully matched with the enterprise recruitment knowledge requirement, the overturn classroom teaching content is marked by another specific numerical value. For example, if the feature word of the enterprise recruitment information is overlapped with the word in the word stock of the teaching content, each piece of overturning class and recruitment information containing the feature word of the recruitment information is marked with the same number (1, 1), (2, 2) larger than 0, and the like, and is used as a matched array, and if the matching is unsuccessful, the recruitment information is marked as 0.
And 105, constructing a skill ability evaluation model.
Determining skill capability to be evaluated and corresponding knowledge requirements according to recruitment requirements, and dividing the knowledge requirements corresponding to the skill capability required by recruitment into three dimensions of general basic knowledge, business domain knowledge and professional front domain knowledge; the method comprises the steps of formulating a student skill ability evaluation index by screening keywords in three dimensions of general basic knowledge, business field knowledge and professional front field knowledge; the cognitive skills are evaluated by designing different evaluation questions, objectively and accurately reflecting the cognitive ability and knowledge level of the recruiter; collecting learning data of students in a learning process through images and voice forms, analyzing the learning data of the students, and evaluating the operation type skill ability of the students; aiming at the characteristic of ambiguity of student ability level evaluation, a fuzzy transformation evaluation method is adopted, various evaluation indexes are comprehensively considered, a student skill ability evaluation model is constructed according to the data characteristics of students, and comprehensive scores of the students are obtained by inputting cognitive skill ability scores and operation skill ability scores of the students corresponding to three dimensions of innovation, application and basic knowledge memory; general basic knowledge includes language and logic, mathematics and statistics, computer foundations, and the like; business domain knowledge refers to business related industries and business knowledge such as finance, medical treatment, entertainment, etc.; the technical frontier domain knowledge refers to the latest technologies and trends such as artificial intelligence, blockchain, etc.
Keywords of knowledge requirement dimensions required for recruitment are screened.
Screening keywords related to student skill ability according to the enterprise recruitment requirement list; classifying and integrating the screened keywords to form a comprehensive keyword list covering three dimensions of general basic knowledge, business domain knowledge and professional front domain knowledge; selecting keywords as skill ability evaluation indexes, and defining weights; assessment of underlying knowledge requirements, such as that an enterprise may guide a job seeker through an associated examination or interview to demonstrate the level of mastery of such knowledge. For example, the financial domain may screen keywords such as "risk control", "investment policy", etc. In the field of machine learning and data science, new technologies including deep learning models, natural language processing, image processing and the like are currently used for evaluating professional front-end knowledge requirements, such as taking a data analyst as an example.
And evaluating the cognitive type skill ability of the students according to the skill ability evaluation index design questions.
Keywords of enterprise recruitment knowledge demands are collected from three dimensions respectively, the keywords are used as index words, and qualitative evaluation of questions is planned by combining teaching experience and machine recommendation; normalizing based on the enterprise recruitment knowledge demand frequency distribution; after defining the weight for each index, refining the evaluation standard and the grading rule, and recursively judging and setting the standard score for the corresponding topic of each keyword from 0 to 10 by using an on-line or off-line test. And after scoring the operations corresponding to all the keywords, calculating an ensemble average score, and finally obtaining the cognitive type skill ability score of each dimension of the student. When keywords are selected as indexes, more attention to business field knowledge or professional front knowledge may be required according to enterprise requirements, and some enterprises may pay more attention to the mastery degree of basic knowledge, so actual requirements and post attributes need to be considered when keywords are selected, and different indexes may be available for different positions. For example, the experience of the SQL database required by a company can be evaluated according to factors such as the volume of data used by a recruiter, whether the data is reasonable and standard, the query speed and the like, and the score can be recursively judged and scored from 0 score to 10 score respectively.
And carrying out student operation type skill capability evaluation by combining an image recognition technology and a voice recognition technology.
After the approval of the school side and the teacher and student is obtained, the classroom image and the classroom voice of the student are collected through a camera or other equipment; keywords of enterprise recruitment knowledge demands are collected from three dimensions respectively, classroom images and classroom voices corresponding to the keywords are collected, image and voice data of actual operation processes of students in different scenes are obtained, and the students are labeled. Feature information about the student's course of operation is extracted from the original image using preprocessing techniques. The deep learning model convolutional neural network is used to automatically extract specific features of specific operations and skills. And extracting the characteristic information of the student voice from the voice signal by using the Mel frequency cepstrum coefficient. Training the image and voice data of the students by using the deep learning model to train out a model for classifying the operation types and skills of the students. In monitoring and recording the student's operations and voices, models are used to identify and evaluate the student's skills and types of operations. After defining the weight for each operation type, refining the evaluation standard and the grading rule, and recursively judging and setting the standard score for the operation corresponding to each keyword from 0 to 10 by using an online or offline test. And after scoring the operations corresponding to all the keywords, calculating an ensemble average score, and finally obtaining the operation type skill ability scores of the students in each dimension. For example, in a cooking operation, skill level may be determined using a knife skill, a cooking time, a material cut, etc. as an index. For example, students operate in a virtual learning environment or an actual scene, and feedback of models is used as a reference basis for student skills and operation types.
And comprehensively evaluating the skills of students by adopting a fuzzy transformation method.
Setting a comprehensive evaluation dimension set R and a grading set Z; integrating the grading set of each student according to the columns to obtain a fuzzy matrix R with three dimensions of each student; multiplying the weight value by a fuzzy matrix, and performing fuzzy transformation to obtain a final comprehensive evaluation value; verifying and using newly collected data to prove talent demand prediction results of enterprises, wherein the judgment content mainly comprises whether the prediction is based on sufficiency, whether a prediction method is proper and scientific, whether the prediction results are built reasonably by reliable post academic specifications, and whether the prediction results are reliable; if the prediction result does not accord with the actual result, repeating the prediction process, and adjusting the mathematical model to finally obtain an accurate and reliable prediction result. And for the enterprise skill prediction part, collecting the existing data as a test set to carry out fitting prediction, comparing the result of the prediction by using the method, and judging whether the result meets the requirements or not, otherwise, carrying out prediction again. For example, the weight values of the three dimensions are (0.3,0.5,0.2), respectively, and the score of student A is (5, 4, 3), and the score of student A is 4.1. In the teaching process, the change of the internal activities of the students cannot be displayed, and the cognitive characteristics of the students cannot be accurately evaluated only through the presumption of external behavior, so that the degree of coincidence between a model established by a traditional method for accurately processing knowledge and the actual situation is not high, and the cognitive ability of the students is determined by using a comprehensive evaluation method based on fuzzy transformation, so that the problems of fuzzy evaluation boundary and uncertain properties of the learning level of the students can be solved.
And 106, calculating the capability difference before and after teaching based on the capability evaluation model.
The method comprises the steps that before the overturning classroom teaching is used, the students are respectively subjected to capability evaluation by texts, images and voices after the overturning classroom teaching is finished; calculating the score of each student before the use of the turnover classroom teaching in three dimensions of innovation, application and basic knowledge memorization and the score after the completion of the turnover classroom teaching, and recording evaluation score data; calculating the comprehensive score of the capability evaluation of each student before using the overturning classroom teaching and the comprehensive score of the capability evaluation after finishing the overturning classroom teaching by using the constructed skill capability evaluation model; calculating a comprehensive score difference value of capability evaluation of students before and after the overturning classroom teaching; assuming that, based on the constructed skill ability evaluation model, the comprehensive score of the skill evaluation before the students use the overturning classroom teaching is calculated to be 80, and the comprehensive score of the ability evaluation after the overturning classroom teaching is finished is calculated to be 90, then the comprehensive score difference of the ability evaluation of the students before and after the overturning classroom teaching is calculated to be 10.
And step 107, adjusting the teaching type according to the overturning classroom teaching effect.
And carrying out longitudinal and transverse comparison analysis of the probability and the score based on the calculated capability difference values of students before and after teaching. Comparing and analyzing from the angles of combination of the weight values of the three dimensions of innovation, application and knowledge memory of the capability requirement and the frequency distribution conditions of the three dimensions before and after the teaching of students, and determining the teaching effect; and (3) combining the fitting condition of the frequency distribution of the talent demand variation of the enterprise, and if the deviation degree exceeds a preset threshold value, re-predicting, and adjusting the content of classroom learning according to the new prediction result. For example, if the weights of the a-ability in three dimensions are (0.3,0.6,0.1), the score of the second dimension is improved, and the influence ratio on teaching effect is the largest. If the average score corresponding to each dimension is increased, the number of students in high sections is increased or the number of students in low sections is decreased, if the average score is changed from 60 to 80, the number of students in the score section of 100-80 is changed from 3 to 10, and the number of students in the score section of 80-60 is changed from 5 to 3, the turnover classroom is considered to obtain the effect.
Criteria are formulated based on the distribution of the calculated scores.
And converting the calculated scores into probability matrixes of different dimension scores of each overturning class. Multiplying the weight value of each dimension by the probability vector, and carrying out normalization operation to obtain the score distribution condition of different dimensions of students in the overturning classroom. And dividing the standard of improving the capacity according to the score distribution condition, and taking the fact that the students with the preset threshold number progress by more than a preset threshold score as the standard of improving the capacity of the students in the class. For example, the score change of each student is assigned, each dimension is assigned a score of 1 per ascending score segment, the score change is obtained by multiplying the score change by the weight value of each dimension, if each dimension ascends by one score segment, the score is 1 after the score change is multiplied by the weight value from (2, 3, 4) to (3, 4, 5). Students with success standard of 60% promote more than 1 minute, for example, 30 students participate in the learning of the course, and then more than one minute is required to be promoted by 18 students.
And analyzing whether to continue using the overturning class teaching of the mixing proportion according to the standard.
If the probability of the achievement is greater than or equal to the teaching target, the teaching through the turnover classroom can be considered to meet the skill requirement of recruiting enterprises in the corresponding knowledge field. If the probability of the achievement is smaller than the teaching target, the skill required by the recruitment enterprise is not successfully obtained through the study of the turnover class, and correction can be performed by referring to the turnover class teaching mode which reaches the standard in the same field or other teaching modes can be adopted. And carrying out sensitivity analysis on the proportion of each mode in the mixed teaching, sequencing the final result of the sensitivity analysis, and selecting the optimal result for adjustment. For example, for each overturning class, students with success standard of 60% can raise more than 1 minute, raise the probability of "4 minutes" and the probability of "3 minutes" and the probability of "2 minutes" and the probability of "1 minute" =0.6, and the students can be considered to successfully acquire the required skills of the enterprise corresponding to the class through the overturning class.
And adjusting the mixed teaching proportion by combining the enterprise capacity demand prediction condition.
And obtaining a teaching strategy recommendation result according to the test student performance and the post-class operation condition. Fitting is carried out on the enterprise capacity requirement and the student capacity improvement condition respectively, the whole student capacity index is used as an input layer, quantization and grading are carried out, the whole student capacity index is used as an offline model for training, the whole student capacity index is adjusted according to the enterprise capacity requirement, an online model is updated, and finally the adjustment of the teaching strategy is changed. Finally, the proportion of each learning mode in the mixed teaching is adjusted, and the overall teaching strategy recommendation result with the proportion determined in each teaching mode is obtained. The teaching strategy recommendation is an intelligent recommendation module based on a DNN and XGBoost fusion model, and can quantify the work level of students and the influence of teacher strategies and establish an inference relationship. The teacher can improve teaching content, revise course standard, formulate teaching plan and talent training scheme etc. according to the relevant information, raise the teaching level, different teachers can produce different effects even if adopt the same tactics, therefore this model is the exclusive teaching model that builds for every teacher.
The foregoing description is only of the preferred embodiments of the present invention, and is not intended to limit the scope of the invention, but rather is intended to cover any equivalents of the structures or equivalent processes disclosed herein or in the alternative, which may be employed directly or indirectly in other related arts.
Claims (8)
1. The data mining-based overturning classroom hybrid teaching method is characterized by comprising the following steps of:
the method comprises the steps of excavating the change of the requirements of different posts and skills of an enterprise; converting the enterprise skill requirement variation into a time-series knowledge requirement distribution, wherein the converting the enterprise skill requirement variation into the time-series knowledge requirement distribution specifically comprises: excavating enterprise recruitment operation type and cognition type requirements, and predicting enterprise operation type and cognition type requirements; digging overturning classroom teaching content information according to the cognitive demands and the operation demands respectively; the method comprises the steps of marking knowledge point classification and teaching classification, judging whether each skill can be obtained through the overturning class with the mixing proportion, marking knowledge point classification and teaching classification, and judging whether each skill can be obtained through the overturning class with the mixing proportion, wherein the method specifically comprises the following steps: word labeling is carried out, a domain knowledge structure is obtained, and whether knowledge requirements are matched with teaching contents or not is judged; building a skill ability evaluation model, wherein the building of the skill ability evaluation model specifically comprises the following steps: screening keywords of knowledge demand dimensionality required for recruitment, evaluating the cognitive type skill ability of students according to skill ability evaluation index design questions, evaluating the operation type skill ability of the students by combining an image recognition technology and a voice recognition technology, and comprehensively evaluating the skill ability of the students by adopting a fuzzy transformation method;
The method for evaluating the skill capability of the student operation type by combining the image recognition technology and the voice recognition technology specifically comprises the following steps:
after the approval of the school side and the teacher and student is obtained, the classroom image and the classroom voice of the student are collected through a camera or other equipment; collecting keywords of enterprise recruitment knowledge demands from three dimensions respectively, collecting classroom images and classroom voices corresponding to the keywords, acquiring image and voice data of students in actual operation processes under different scenes, and labeling the image and voice data; extracting characteristic information about the student operation process from the original image by using a preprocessing technology; automatically extracting specific characteristics of specific operations and skills by using a deep learning model convolutional neural network; extracting characteristic information of student voices from voice signals by using Mel frequency cepstrum coefficients; training the image and voice data of the students by using a deep learning model to train a model for classifying the operation types and skills of the students; identifying and evaluating skills and operation types of students using the model while monitoring and recording operations and voices of the students; after defining the weight for each operation type, refining the evaluation standard and the grading rule, and recursively judging and setting a standard score for the operation corresponding to each keyword from 0 to 10 by using an online or offline test; after scoring the operations corresponding to all the keywords, calculating an overall average score, and finally obtaining the operation type skill ability scores of all the dimensions of the students;
The comprehensive evaluation of student skill ability is carried out by adopting a fuzzy transformation method, which comprises the following steps:
setting a comprehensive evaluation dimension set R and a grading set Z; integrating the grading set of each student according to the columns to obtain a fuzzy matrix R with three dimensions of each student; multiplying the weight value by a fuzzy matrix, and performing fuzzy transformation to obtain a final comprehensive evaluation value; verifying and using newly collected data to prove talent demand prediction results of enterprises, wherein the judgment content mainly comprises whether the prediction is based on sufficiency, whether a prediction method is proper and scientific, whether the prediction results are built reasonably by reliable post academic specifications, and whether the prediction results are reliable; if the prediction result does not accord with the actual result, repeating the prediction process, and adjusting the mathematical model to finally obtain an accurate and reliable prediction result; for the enterprise skill prediction part, collecting the existing data as a test set for fitting prediction, comparing the result of the prediction with the result of the prediction by using the method, judging whether the result meets the requirement, and otherwise, re-predicting;
calculating the capability difference before and after teaching based on the capability evaluation model; according to upset classroom teaching effect adjustment teaching type, specifically include: and (3) formulating a standard based on the distribution of the calculated scores, and adjusting the mixed teaching proportion according to whether the overturning classroom teaching of the mixed proportion is continuously used or not according to standard analysis and combining with the enterprise capacity demand prediction condition.
2. The method of claim 1, wherein the mining enterprise for different post demand quantity and skill demand variations comprises:
the enterprise recruitment basic data comprise post names, quantity, post contents and text data required by posts, and data contained in each piece of recruitment information are obtained through recruitment websites and APP; aiming at operation type requirements and cognitive type requirements, collecting enterprise recruitment basic data by taking month as a unit to form data of requirement variation; performing data preprocessing by using an ETL method, and reserving text data containing post names, quantity, post contents and post requirements; the text data of the post requirements and post contents are subjected to word segmentation processing by utilizing the ETL method and the processed data and utilizing a Jieba library; data cleaning is carried out on text data after word segmentation, stop words and the like are removed, a part-of-speech list is distributed to each word through a POS (point of sale) tag by using a dictionary, and the list is classified into the part-of-speech of each word by using an ambiguity elimination rule; based on the part-of-speech classification of each word, the skill requirement feature words of each post are extracted by using an NLP method, and the feature words are textual into vectors of N dimensions.
3. The method of claim 1, wherein the converting enterprise skill need variation into a time-series knowledge need distribution comprises:
Acquiring a knowledge system of recruitment information of each specific occupation post by utilizing a Web text mining technology, wherein the occupation capability requirement, the curriculum occupation capability target and the knowledge target of each specific occupation post need a knowledge system with corresponding modules to match the curriculum teaching content; aiming at different enterprises recruited each year, counting keyword frequency by taking years as a unit, and drawing a variation demand distribution diagram of a knowledge time sequence; comprising the following steps: excavating the recruitment operation type and the cognitive type requirements of enterprises; predicting enterprise operation class and cognitive class requirements;
the mining of enterprise recruitment operation class and cognitive class requirements specifically comprises:
using an SVM algorithm, using the post class name as a label, dividing the text of the feature words in the scheme into N-dimensional vectors into a training set and a test set, taking 70% as the training set and 30% as the test set, and then adopting an SVM based on a Gaussian radial basis function to carry out machine learning classification; for the results of machine learning classification through SVM, web text mining technology is applied to acquire the knowledge systems of each specialty one by one, classification comprehensive analysis among posts is carried out according to the processes, and a distribution system of specific posts under the large posts is constructed;
The predicting the operation type and the cognitive type demands of the enterprise specifically comprises the following steps:
the collected feature words are text into quantitative data converted from N-dimensional vectors, and the quantitative data are input as training data; building a prediction model by using a lightGBM algorithm, traversing training data, calculating accumulated statistics of each discrete value in the histogram, and traversing and searching optimal division points according to the discrete values of the histogram when performing feature selection to obtain a final prediction result; and determining main parameters of the prediction model according to the training data, inputting the data into the prediction model for comprehensive operation, and obtaining a prediction result by adjusting the parameters.
4. The method of claim 1, wherein mining the flipping class instructional content information according to the cognitive class requirements and the operational class requirements, respectively, comprises:
from two angles of operation class requirements and cognition class requirements, using 'turning class application', 'turning class practice' and the like as keywords, and collecting information of teaching in which knowledge fields the turning class is currently applied to; based on the collected information of teaching of each subject knowledge field, collecting course names and knowledge point names of the subject knowledge field, and establishing a professional classification word stock so as to expand the established course word stock; data cleaning is carried out on the collected information of the overturning class, text analysis is carried out on the screened information through NLP, and a word stock of overturning class teaching content is built; based on the established word stock of the overturning class contents, the teaching content characteristic words of each overturning class are extracted and used as specific teaching contents.
5. The method of claim 1, wherein the marking knowledge point classification and teaching classification to determine whether each skill can be obtained through a class of such mixed proportion turnovers comprises:
the feature words are text into N-dimensional word vectors, the feature words of the recruitment required knowledge points are included, the feature words of the recruitment required knowledge points are matched in a word stock of teaching content one by one, and feature words which can be matched accurately and feature words which cannot be matched accurately are obtained; scanning a word library of the teaching content by using an Apriori algorithm for the feature words which cannot be matched accurately, counting the occurrence times of a first-level candidate set, eliminating the candidate set which does not meet the condition, generating a second-level candidate set on a first-level frequent set, scanning the data rate, counting the occurrence times of the second-level candidate set, sequentially circulating, and finally mining the association degree of the required knowledge points and the teaching content in the field; if the knowledge characteristic words corresponding to the required skills can find the matched words in the word stock, a proper overturning classroom can be found for teaching, and the matched words are found in the overturning classroom, otherwise, the overturning classroom teaching mode needs to be optimized for re-evaluation; comprising the following steps: word labeling is carried out, and a domain knowledge structure is obtained; judging whether the knowledge requirement is matched with the teaching content or not;
The word labeling is carried out, and the domain knowledge structure is obtained, which specifically comprises the following steps:
performing data search on professions related to the overturning classroom, and collecting professional knowledge text information; processing the collected professional knowledge text information by utilizing text analysis, marking the terms of different professional knowledge, establishing a professional knowledge system word stock, and obtaining a domain knowledge structure;
judging whether the knowledge requirement is matched with the teaching content or not specifically comprises the following steps:
analyzing the matching degree between the overturning classroom teaching content and the enterprise recruitment knowledge demand, and taking the similarity between the capacity demand word vector and the word vector of the teaching content as the matching degree between texts, so as to establish a matching model; if the overturning classroom teaching content is successfully matched with the enterprise recruitment knowledge requirement, labeling by using a specific numerical value; if the overturn classroom teaching content is not successfully matched with the enterprise recruitment knowledge requirement, the overturn classroom teaching content is marked by another specific numerical value.
6. The method of claim 1, wherein the constructing a skill ability assessment model comprises:
determining skill capability to be evaluated and corresponding knowledge requirements according to recruitment requirements, and dividing the knowledge requirements corresponding to the skill capability required by recruitment into three dimensions of general basic knowledge, business domain knowledge and professional front domain knowledge; the method comprises the steps of formulating a student skill ability evaluation index by screening keywords in three dimensions of general basic knowledge, business field knowledge and professional front field knowledge; the cognitive skills are evaluated by designing different evaluation questions, objectively and accurately reflecting the cognitive ability and knowledge level of the recruiter; collecting learning data of students in a learning process through images and voice forms, analyzing the learning data of the students, and evaluating the operation type skill ability of the students; aiming at the characteristic of ambiguity of student ability level evaluation, a fuzzy transformation evaluation method is adopted, various evaluation indexes are comprehensively considered, a student skill ability evaluation model is constructed according to the data characteristics of students, and comprehensive scores of the students are obtained by inputting cognitive skill ability scores and operation skill ability scores of the students corresponding to three dimensions of innovation, application and basic knowledge memory; comprising the following steps: screening keywords of knowledge requirement dimension required by recruitment; according to the skill ability evaluation index design questions, evaluating the cognitive type skill ability of the students; performing student operation type skill capability evaluation by combining an image recognition technology and a voice recognition technology; comprehensively evaluating the skills of students by adopting a fuzzy transformation method;
The screening of keywords of the knowledge requirement dimension required for recruitment specifically comprises the following steps:
screening keywords related to student skill ability according to the enterprise recruitment requirement list; classifying and integrating the screened keywords to form a comprehensive keyword list covering three dimensions of general basic knowledge, business domain knowledge and professional front domain knowledge; selecting keywords as skill ability evaluation indexes, and defining weights;
the method for evaluating the cognitive type skill ability of the students according to the skill ability evaluation index design questions specifically comprises the following steps:
keywords of enterprise recruitment knowledge demands are collected from three dimensions respectively, the keywords are used as index words, and qualitative evaluation of questions is planned by combining teaching experience and machine recommendation; normalizing based on the enterprise recruitment knowledge demand frequency distribution; after defining the weight of each index, refining the evaluation standard and the grading rule, and recursively judging and setting standard scores for the topics corresponding to each keyword from 0 to 10 by using online or offline test; and after scoring the operations corresponding to all the keywords, calculating an ensemble average score, and finally obtaining the cognitive type skill ability score of each dimension of the student.
7. The method of claim 1, wherein the computing the teaching front-to-back capability difference based on the capability assessment model comprises:
the method comprises the steps that before the overturning classroom teaching is used, the students are respectively subjected to capability evaluation by texts, images and voices after the overturning classroom teaching is finished; calculating the score of each student before the use of the turnover classroom teaching in three dimensions of innovation, application and basic knowledge memorization and the score after the completion of the turnover classroom teaching, and recording evaluation score data; calculating the comprehensive score of the capability evaluation of each student before using the overturning classroom teaching and the comprehensive score of the capability evaluation after finishing the overturning classroom teaching by using the constructed skill capability evaluation model; and calculating the comprehensive score difference value of the capability evaluation of the students before and after the overturning classroom teaching.
8. The method of claim 1, wherein said adjusting the teaching type according to the flip class teaching effect comprises:
based on the calculated capability difference values of students before and after teaching, carrying out longitudinal and transverse comparison analysis of probability and score values; comparing and analyzing from the angles of combination of the weight values of the three dimensions of innovation, application and knowledge memory of the capability requirement and the frequency distribution conditions of the three dimensions before and after the teaching of students, and determining the teaching effect; combining the fitting condition of frequency distribution of talent demand variation of enterprises, if the deviation degree exceeds a preset threshold value, predicting again, and adjusting the content of classroom learning according to a new prediction result; comprising the following steps: formulating a criterion based on the distribution of the calculated scores; according to standard analysis, whether to continue using the overturning classroom teaching of the mixing proportion; adjusting the mixed teaching proportion by combining the capability demand prediction condition of the enterprise;
The distribution formulation standard based on the calculated score specifically comprises the following steps:
converting the calculated score into a probability matrix of different dimension scores of each overturning class; multiplying the weight value of each dimension by the probability vector, and carrying out normalization operation to obtain the score distribution conditions of different dimensions of students in the overturning classroom; dividing the standard of improving the capacity according to the score distribution condition, and taking the fact that the students with the preset threshold number progress by more than a preset threshold score as the standard of improving the capacity of the students in the class;
whether the overturning classroom teaching of the mixing proportion is continuously used or not is analyzed according to the standard, and the overturning classroom teaching specifically comprises the following steps:
if the probability of the achievement is greater than or equal to the teaching target, the skill requirement of recruiting enterprises in the corresponding knowledge field can be met through the teaching of the turnover class; if the probability of the achievement is smaller than the teaching target, the skill required by recruiting enterprises is not successfully obtained through the study of the turnover class, and correction can be performed by referring to the turnover class teaching mode which reaches the standard in the same field or other teaching modes can be adopted; performing sensitivity analysis on the proportion of each mode in mixed teaching, sequencing the final result of the sensitivity analysis, and selecting an optimal result for adjustment;
The method for adjusting the mixed teaching proportion by combining the capability demand prediction condition of the enterprise specifically comprises the following steps:
obtaining a teaching strategy recommendation result according to the student performance and the post-class operation condition in the test; fitting the enterprise capacity requirement and the student capacity improvement condition respectively, taking the whole student capacity index as an input layer, carrying out quantization and grading, training as an offline model, adjusting the enterprise capacity requirement, updating an online model, and finally changing the adjustment of the teaching strategy; finally, the proportion of each learning mode in the mixed teaching is adjusted, and the overall teaching strategy recommendation result with the proportion determined in each teaching mode is obtained.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310411785.0A CN116523225B (en) | 2023-04-18 | 2023-04-18 | Data mining-based overturning classroom hybrid teaching method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310411785.0A CN116523225B (en) | 2023-04-18 | 2023-04-18 | Data mining-based overturning classroom hybrid teaching method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116523225A CN116523225A (en) | 2023-08-01 |
| CN116523225B true CN116523225B (en) | 2024-01-23 |
Family
ID=87402274
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310411785.0A Active CN116523225B (en) | 2023-04-18 | 2023-04-18 | Data mining-based overturning classroom hybrid teaching method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116523225B (en) |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105512864A (en) * | 2016-01-28 | 2016-04-20 | 丁沂 | Method for automatically acquiring post professional ability requirements based on internet |
| CN107590133A (en) * | 2017-10-24 | 2018-01-16 | 武汉理工大学 | The method and system that position vacant based on semanteme matches with job seeker resume |
| CN108121785A (en) * | 2017-12-15 | 2018-06-05 | 华中师范大学 | A kind of analysis method based on education big data |
| CN108280583A (en) * | 2018-01-26 | 2018-07-13 | 重庆工商大学 | Post skill requirement analysis method based on big data |
| CN112132536A (en) * | 2020-08-31 | 2020-12-25 | 三盟科技股份有限公司 | Post recommendation method, system, computer equipment and storage medium |
| CN112686789A (en) * | 2021-01-11 | 2021-04-20 | 重庆电子工程职业学院 | Intelligent evaluation method for classroom teaching effect of colleges and universities |
| CN112861530A (en) * | 2021-03-17 | 2021-05-28 | 华南农业大学 | Course setting analysis method based on text mining |
| CN113886588A (en) * | 2021-10-20 | 2022-01-04 | 华南农业大学 | Major professional employment direction identification method based on recruitment text mining |
| CN114282056A (en) * | 2021-12-08 | 2022-04-05 | 华特数字科技有限公司 | Interactive turnover classroom teaching method and system |
| CN115330568A (en) * | 2022-10-14 | 2022-11-11 | 北京师范大学 | Intelligent classroom teaching quality evaluation system and method |
-
2023
- 2023-04-18 CN CN202310411785.0A patent/CN116523225B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105512864A (en) * | 2016-01-28 | 2016-04-20 | 丁沂 | Method for automatically acquiring post professional ability requirements based on internet |
| CN107590133A (en) * | 2017-10-24 | 2018-01-16 | 武汉理工大学 | The method and system that position vacant based on semanteme matches with job seeker resume |
| CN108121785A (en) * | 2017-12-15 | 2018-06-05 | 华中师范大学 | A kind of analysis method based on education big data |
| CN108280583A (en) * | 2018-01-26 | 2018-07-13 | 重庆工商大学 | Post skill requirement analysis method based on big data |
| CN112132536A (en) * | 2020-08-31 | 2020-12-25 | 三盟科技股份有限公司 | Post recommendation method, system, computer equipment and storage medium |
| CN112686789A (en) * | 2021-01-11 | 2021-04-20 | 重庆电子工程职业学院 | Intelligent evaluation method for classroom teaching effect of colleges and universities |
| CN112861530A (en) * | 2021-03-17 | 2021-05-28 | 华南农业大学 | Course setting analysis method based on text mining |
| CN113886588A (en) * | 2021-10-20 | 2022-01-04 | 华南农业大学 | Major professional employment direction identification method based on recruitment text mining |
| CN114282056A (en) * | 2021-12-08 | 2022-04-05 | 华特数字科技有限公司 | Interactive turnover classroom teaching method and system |
| CN115330568A (en) * | 2022-10-14 | 2022-11-11 | 北京师范大学 | Intelligent classroom teaching quality evaluation system and method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116523225A (en) | 2023-08-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Nieto et al. | Supporting academic decision making at higher educational institutions using machine learning-based algorithms | |
| CN107230174B (en) | Online interactive learning system and method based on network | |
| CN109858797B (en) | Multi-dimensional informatics analysis method based on knowledge network accurate online education system | |
| CN111325529A (en) | Talent evaluation-based personnel and post matching method and system | |
| CN114913729B (en) | Question selecting method, device, computer equipment and storage medium | |
| Wilson et al. | Classification of open-ended responses to a research-based assessment using natural language processing | |
| Liu | Data Analysis of Educational Evaluation Using K‐Means Clustering Method | |
| PIAD | DETERMINING THE DOMINANT ATTRIBUTES OF INFORMATION TECHNOLOGY GRADUATES EMPLOYABILITY PREDICTION USING DATA MINING CLASSIFICATION TECHNIQUES. | |
| CN117726485B (en) | Intelligent adaptation education learning method and system based on big data | |
| Alowaigl et al. | A multi-criteria assessment of decision support systems in educational environments | |
| CN118195560A (en) | Multi-dimensional data matching method and system based on professional talents | |
| Wang | Analysis and prediction of CET4 scores based on data mining algorithm | |
| CN117313947A (en) | Learning early warning method based on random forest and tuna group optimization algorithm | |
| CN116523225B (en) | Data mining-based overturning classroom hybrid teaching method | |
| Li et al. | A comprehensive evaluation model for university classroom teaching quality based on machine vision and Fermatean fuzzy sets | |
| CN109272233A (en) | A kind of employee's competency appraisal procedure closed based on type-2 fuzzy sets | |
| Muraina et al. | Forecasting Students’ Job Placement using Data Science Paradigms | |
| Koyuncu et al. | Comparison of data mining classification algorithms on educational data under different conditions | |
| CN113822774A (en) | Financial teaching system, using method thereof and computer readable storage medium | |
| Mahboob et al. | A comparative study of engineering students pedagogical progress | |
| Sanmorino et al. | Building Research Productivity Framework in Higher Education Institution | |
| Zhu | Reform and Practice of Public English Examination Mode in Colleges and Universities Using Big Data Analysis and Speech Recognition | |
| CN116109456B (en) | Comprehensive evaluation method and system for intelligent education, electronic equipment and storage medium | |
| Sungar et al. | Predicting student’s performance using machine learning | |
| Khadragy et al. | Adaptive Network Based Fuzzy Inference System and the Future of Employability |
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 | ||
| GR01 | Patent grant |