CN109783647A - The construction method of intelligence learning model - Google Patents
The construction method of intelligence learning model Download PDFInfo
- Publication number
- CN109783647A CN109783647A CN201811574626.8A CN201811574626A CN109783647A CN 109783647 A CN109783647 A CN 109783647A CN 201811574626 A CN201811574626 A CN 201811574626A CN 109783647 A CN109783647 A CN 109783647A
- Authority
- CN
- China
- Prior art keywords
- knowledge point
- topic
- knowledge
- individualized medicine
- exercise
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Electrically Operated Instructional Devices (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The present invention provides a kind of construction methods of intelligence learning model, it is characterised in that the following steps are included: the first step, initial exercise push;Second step, acquisition practice data;Third step updates individualized medicine map;4th step pushes exercise resource.The purpose of the present invention is to the defects of the prior art, provide a kind of construction method of intelligence learning model, realize the individualized learning of student, meet the needs of adaptive learning.
Description
Technical field
The present invention relates to big data analysis and field of artificial intelligence, and in particular to a kind of building of intelligence learning model
Method.
Background technique
On-line study has become new trend of the enterprises and institutions on mode of learning, current mode of learning or dependence
Student selects the push or random, extensive push training packets of realization exercise resource after the catalogue of knowledge point, learns for effectively covering
Member's knowledge point defect meets the requirement of student's individualized learning, differentiation study, has invented a kind of building of intelligence learning model
Method.It realizes the accurate push of student's education resource, promotes study validity, specific aim, learning efficiency.
Summary of the invention
The purpose of the present invention is to the defects of the prior art, provide a kind of construction method of intelligence learning model, real
The individualized learning of existing student, meets the needs of adaptive learning.
The present invention provides a kind of construction methods of intelligence learning model, it is characterised in that the following steps are included:
The first step, initial exercise push;
Second step, acquisition practice data;
Third step updates individualized medicine map;
4th step pushes exercise resource.
In above-mentioned technical proposal, the first step the following steps are included:
If student learns for the first time, that is, individualized medicine map, without any data of this student, student voluntarily selects
It selects and is no more than N/2 knowledge point, it is assumed that the knowledge point quantity selected is s (0 < s≤N/2);
3) the volume distribution of each knowledge point:
tsi=round (N/s) (1 ..., s-1), round is the function that rounds up.
Due to the case where there are more 1,
4) each knowledge point topic selection:
With knowledge point, minimum degree-of-difficulty factor 1, corresponding knowledge point topic numerical digit ts is chosen from exam pooliTopic, complete just
The push of beginning exercise.
In above-mentioned technical proposal, second step the following steps are included:
Acquisition practice data, i.e. answer situation of the acquisition student's individual to push exercise.Including topic number, topic name
Whether title, knowledge point, answer correct, answer duration, whether time-out, constitute the basic data collection of student's learning behavior.
In above-mentioned technical proposal, third step the following steps are included:
Every time after practice, batch updating is carried out to individualized medicine spectrum library, existing knowledge point is needed to update difficulty
Mean value, wrong topic rate, delay rate, severity index, add for not depositing knowledge point.After the completion of index updates, to knowledge mapping
It resequences according to severity index is descending;
The computation rule of each index is as follows in individualized medicine spectrum library
2) knowledge point severity index ai
ai=ei×80+pi×20
2) knowledge point mistake topic rate ei。
3) the delay rate p of knowledge pointi
4) the difficulty mean value d of knowledge pointi
diValue round up, retain a decimal.
In above-mentioned technical proposal, the 4th step the following steps are included:
Determine topic distributed number n:
If 1) there is no data in individualized medicine picture library, initial exercise push is please referred to, if it exists data, setting knowledge points
It is N/2, if having data in individualized medicine picture library, based on the quantity N/2 of knowledge point, knowledge point is from individualized medicine spectrum library
Middle acquisition.If knowledge point quantity is less than N/2 in individualized medicine map, actual quantity is taken;If more than or be equal to N/2, then n=N/
2;
2) n severity index a before taking individualized medicine map to ranki, and it is inherently descending according to severity index
It is arranged, therefore it is highest at first to verify index.If severity index aiAll it is 0, then mean allocation is used, to each
The scope of one's knowledge corresponds to topic 2.The case where not being at least 0 in the presence of 1, then calculate the weight coefficient ra of severity indexi(i=1 ... n);
3) the weight coefficient ra of severity index is calculatedi(i=1 ... n):
It is assumed that weight coefficient raiNumber for 0 is k (1≤k < n), then each 1 topic of first distribution in this k knowledge point;
tsi=round [(N-k) × rai] (i=2 .., n-k)
Wherein round is the function that rounds up, it is contemplated that is easy to have more 1 in the function that rounds up, compensate having more
Give weight highest ts1;
Determine the source of topic:
The source of topic is with knowledge point, difficulty mean value, and topic number is primary condition, is provided by way of comparison from exercise
Source is extracted in library, the knowledge point that knowledge point is N/2 before taking individualized medicine map to rank, topic number tsiObtaining value method exist
It is clear in step 1, difficulty mean value dniIt is converted based on the difficulty mean value D (i) for corresponding to knowledge point in knowledge mapping;
Conversion method is as follows:
3) such as knowledge point z, to deserved severity index diIt is 0.
Then dni=1;
4) such as knowledge point z, to deserved severity index diIt is not 0.
Then dni=floor (di)
Floor is the function being rounded downwards;
When screening topic, need to filter out the exercise done in exercise packet.
The present invention on the basis of establishing question bank model, by acquiring the learning records of student, know by the individual for establishing student
Know map, then by building resource supplying engine, dynamic pushes education resource, realizes the individualized learning of student, meet adaptive
The needs that should learn.
Detailed description of the invention
Fig. 1 is for the invention patent structural schematic diagram
Specific embodiment
The following further describes the present invention in detail with reference to the accompanying drawings and specific embodiments, convenient for this hair is well understood
It is bright, but they limiting the invention.
As shown in Figure 1, the present invention provides a kind of construction methods of intelligence learning model, it is characterised in that including following step
It is rapid:
The first step, initial exercise push;
Second step, acquisition practice data;
Third step updates individualized medicine map;
4th step pushes exercise resource.
In above-mentioned technical proposal, the first step the following steps are included:
If student learns for the first time, that is, individualized medicine map, without any data of this student, student voluntarily selects
It selects and is no more than N/2 knowledge point, it is assumed that the knowledge point quantity selected is s (0 < s≤N/2);
5) the volume distribution of each knowledge point:
tsi=round (N/s) (1 ..., s-1), round is the function that rounds up.
Due to the case where there are more 1,
6) each knowledge point topic selection:
With knowledge point, minimum degree-of-difficulty factor 1, corresponding knowledge point topic numerical digit ts is chosen from exam pooliTopic, complete just
The push of beginning exercise.
In above-mentioned technical proposal, second step the following steps are included:
Acquisition practice data, i.e. answer situation of the acquisition student's individual to push exercise.Including topic number, topic name
Whether title, knowledge point, answer correct, answer duration, whether time-out, constitute the basic data collection of student's learning behavior.
In above-mentioned technical proposal, third step the following steps are included:
Every time after practice, batch updating is carried out to individualized medicine spectrum library based on table one, to existing knowledge point
It needs to update difficulty mean value, wrong topic rate, delay rate, severity index, is added for not depositing knowledge point.Index, which updates, to be completed
Afterwards, it resequences to knowledge mapping according to severity index is descending;
Table one
Topic | The scope of one's knowledge | Conventional duration | Practical duration | It is whether correct | It is whether overtime |
Topic 1 | Z1 | L1 | T1 | It is | It is no |
Topic 2 | Z2 | L2 | T2 | It is no | In vain |
Topic 3 | Z3 | L3 | T3 | It is | T2>L2 |
Topic 4 | Z4+Z5 | L4 | T4 | It is | T4<L4 |
...... | ..... | ...... | ..... | ...... | ..... |
The computation rule of each index is as follows in individualized medicine spectrum library
3) knowledge point severity index ai
ai=ei×80+pi×20
2) knowledge point mistake topic rate ei。
3) the delay rate p of knowledge pointi
4) the difficulty mean value d of knowledge pointi
diValue round up, retain a decimal.
In above-mentioned technical proposal, the 4th step the following steps are included:
Determine topic distributed number n:
If 1) there is no data in individualized medicine picture library, initial exercise push is please referred to, if it exists data, setting knowledge points
It is N/2, if having data in individualized medicine picture library, based on the quantity N/2 of knowledge point, knowledge point is from individualized medicine spectrum library
Middle acquisition.If knowledge point quantity is less than N/2 in individualized medicine map, actual quantity is taken;If more than or be equal to N/2, then n=N/
2;
2) n severity index a before taking individualized medicine map to ranki, and it is inherently descending according to severity index
It is arranged, therefore it is highest at first to verify index.If severity index aiAll it is 0, then mean allocation is used, to each
The scope of one's knowledge corresponds to topic 2.The case where not being at least 0 in the presence of 1, then calculate the weight coefficient ra of severity indexi(i=1 ... n);
3) the weight coefficient ra of severity index is calculatedi(i=1 ... n):
It is assumed that weight coefficient raiNumber for 0 is k (1≤k < n), then each 1 topic of first distribution in this k knowledge point;
tsi=round [(N-k) × rai] (i=2, n-k)
Wherein round is the function that rounds up, it is contemplated that is easy to have more 1 in the function that rounds up, compensate having more
Give weight highest ts1;
Determine the source of topic:
The source of topic is with knowledge point, difficulty mean value, and topic number is primary condition, is provided by way of comparison from exercise
Source is extracted in library, the knowledge point that knowledge point is N/2 before taking individualized medicine map to rank, topic number tsiObtaining value method exist
It is clear in step 1, difficulty mean value dniIt is converted based on the difficulty mean value D (i) for corresponding to knowledge point in knowledge mapping;
Conversion method is as follows:
5) such as knowledge point z, to deserved severity index diIt is 0.
Then dni=1;
6) such as knowledge point z, to deserved severity index diIt is not 0.
Then dni=floor (di)
Floor is the function being rounded downwards;
When screening topic, need to filter out the exercise done in exercise packet.
The content that this specification is not described in detail belongs to the prior art well known to professional and technical personnel in the field.
Claims (5)
1. a kind of construction method of intelligence learning model, it is characterised in that the following steps are included:
The first step, initial exercise push;
Second step, acquisition practice data;
Third step updates individualized medicine map;
4th step pushes exercise resource.
2. the construction method of intelligence learning model according to claim 1, it is characterised in that: the first step includes following step
It is rapid:
If student learns for the first time, that is, individualized medicine map, without any data of this student, student voluntarily selects not
More than N/2 knowledge point, it is assumed that the knowledge point quantity selected is s (0 < s≤N/2);
1) the volume distribution of each knowledge point:
tsi=round (N/s) (1 ..., s-1), round is the function that rounds up.
Due to the case where there are more 1,
2) each knowledge point topic selection:
With knowledge point, minimum degree-of-difficulty factor 1, corresponding knowledge point topic numerical digit ts is chosen from exam pooliTopic, complete initial practise
Topic push.
3. the construction method of intelligence learning model according to claim 2, it is characterised in that second step the following steps are included:
Acquisition practice data, i.e. answer situation of the acquisition student's individual to push exercise.Including topic number, topic title, know
Know whether correct point, answer, answer duration, whether time-out, constitute the basic data collection of student's learning behavior.
4. the construction method of intelligence learning model according to claim 3, it is characterised in that third step the following steps are included:
Every time practice after, to individualized medicine spectrum library carry out batch updating, to existing knowledge point need to update difficulty mean value,
Wrong topic rate, delay rate, severity index, add for not depositing knowledge point.After the completion of index updates, to knowledge mapping according to tight
Index is descending again resequences;
The computation rule of each index is as follows in individualized medicine spectrum library
1) knowledge point severity index ai
ai=ei×80+pi×20
2) knowledge point mistake topic rate ei。
3) the delay rate p of knowledge pointi
4) the difficulty mean value d of knowledge pointi
diValue round up, retain a decimal.
5. the construction method of intelligence learning model according to claim 4, it is characterised in that the 4th step the following steps are included:
Determine topic distributed number n:
If 1) there is no data in individualized medicine picture library, initial exercise push is please referred to, if it exists data, setting knowledge points are N/
2, if having data in individualized medicine picture library, based on the quantity N/2 of knowledge point, knowledge point is obtained from individualized medicine spectrum library
It takes.If knowledge point quantity is less than N/2 in individualized medicine map, actual quantity is taken;If more than or be equal to N/2, then n=N/2;
2) n severity index a before taking individualized medicine map to ranki, and arranged according to severity index is inherently descending
Column, therefore it is highest at first to verify index.If severity index aiAll it is 0, then mean allocation is used, to each the scope of one's knowledge
Corresponding topic 2.The case where not being at least 0 in the presence of 1, then calculate the weight coefficient ra of severity indexi(i=1 ... n);
3) the weight coefficient ra of severity index is calculatedi(i=1 ... n):
It is assumed that weight coefficient raiNumber for 0 is k (1≤k < n), then each 1 topic of first distribution in this k knowledge point;
tsi=round [(N-k) × rai] (i=2, n-k)
Wherein round is the function that rounds up, it is contemplated that is easy to have more 1 in the function that rounds up, compensate having more to power
The highest ts of weight1;
Determine the source of topic:
The source of topic is with knowledge point, difficulty mean value, and topic number is primary condition, from exercise resources bank by way of comparison
Middle extraction, the knowledge point that knowledge point is N/2 before taking individualized medicine map to rank, topic number tsiObtaining value method in step
It is clear in one, difficulty mean value dniIt is converted based on the difficulty mean value D (i) for corresponding to knowledge point in knowledge mapping;
Conversion method is as follows:
1) such as knowledge point z, to deserved severity index diIt is 0.
Then dni=1;
2) such as knowledge point z, to deserved severity index diIt is not 0.
Then dni=floor (di)
Floor is the function being rounded downwards;
When screening topic, need to filter out the exercise done in exercise packet.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811574626.8A CN109783647B (en) | 2018-12-21 | 2018-12-21 | Construction method of intelligent learning model |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811574626.8A CN109783647B (en) | 2018-12-21 | 2018-12-21 | Construction method of intelligent learning model |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109783647A true CN109783647A (en) | 2019-05-21 |
CN109783647B CN109783647B (en) | 2023-06-13 |
Family
ID=66497625
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811574626.8A Active CN109783647B (en) | 2018-12-21 | 2018-12-21 | Construction method of intelligent learning model |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109783647B (en) |
Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101477755A (en) * | 2009-01-06 | 2009-07-08 | 陈诗周 | Error problem database system for student |
US20130288222A1 (en) * | 2012-04-27 | 2013-10-31 | E. Webb Stacy | Systems and methods to customize student instruction |
CN104966427A (en) * | 2015-05-27 | 2015-10-07 | 北京创数教育科技发展有限公司 | Self-adaptation teaching interaction system and method |
CN106469169A (en) * | 2015-08-19 | 2017-03-01 | 阿里巴巴集团控股有限公司 | Information processing method and device |
CN106599999A (en) * | 2016-12-09 | 2017-04-26 | 北京爱论答科技有限公司 | Evaluation method and system for using small amount of questions to accurately detect segmented weak knowledge points of student |
CN107085803A (en) * | 2017-03-31 | 2017-08-22 | 弘成科技发展有限公司 | The individualized teaching resource recommendation system of knowledge based collection of illustrative plates and capability comparison |
US20180053098A1 (en) * | 2016-08-16 | 2018-02-22 | International Business Machines Corporation | Automatic evaluation of a knowledge canvassing application |
CN108053130A (en) * | 2017-12-25 | 2018-05-18 | 郑州威科姆科技股份有限公司 | A kind of multimedia teaching analysis system and analysis method |
CN108647363A (en) * | 2018-05-21 | 2018-10-12 | 安徽知学科技有限公司 | Map construction, display methods, device, equipment and storage medium |
-
2018
- 2018-12-21 CN CN201811574626.8A patent/CN109783647B/en active Active
Patent Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101477755A (en) * | 2009-01-06 | 2009-07-08 | 陈诗周 | Error problem database system for student |
US20130288222A1 (en) * | 2012-04-27 | 2013-10-31 | E. Webb Stacy | Systems and methods to customize student instruction |
CN104966427A (en) * | 2015-05-27 | 2015-10-07 | 北京创数教育科技发展有限公司 | Self-adaptation teaching interaction system and method |
CN106469169A (en) * | 2015-08-19 | 2017-03-01 | 阿里巴巴集团控股有限公司 | Information processing method and device |
US20180053098A1 (en) * | 2016-08-16 | 2018-02-22 | International Business Machines Corporation | Automatic evaluation of a knowledge canvassing application |
CN106599999A (en) * | 2016-12-09 | 2017-04-26 | 北京爱论答科技有限公司 | Evaluation method and system for using small amount of questions to accurately detect segmented weak knowledge points of student |
CN107085803A (en) * | 2017-03-31 | 2017-08-22 | 弘成科技发展有限公司 | The individualized teaching resource recommendation system of knowledge based collection of illustrative plates and capability comparison |
CN108053130A (en) * | 2017-12-25 | 2018-05-18 | 郑州威科姆科技股份有限公司 | A kind of multimedia teaching analysis system and analysis method |
CN108647363A (en) * | 2018-05-21 | 2018-10-12 | 安徽知学科技有限公司 | Map construction, display methods, device, equipment and storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN109783647B (en) | 2023-06-13 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107610012B (en) | Course selecting and arranging system and course selecting and arranging method thereof | |
CN104835087B (en) | The data processing method and device of education testing system | |
CN104504949A (en) | Remote education information platform system | |
CN107464461A (en) | One kind teaching cloud platform and its implementation | |
CN109816265B (en) | Knowledge characteristic mastery degree evaluation method, question recommendation method and electronic equipment | |
CN108920663B (en) | Method for automatically recommending title content | |
CN102222155A (en) | Method for high-efficiency class grouping and course scheduling in multi-restraint condition | |
CN106095812A (en) | Intelligent test paper generation method based on similarity measurement | |
CN109739829A (en) | A kind of student knowledge system anlysis system based on ongoing operations | |
CN104063759A (en) | B/S-based power business application operation examination system | |
CN110288878A (en) | Adaptive learning method and device | |
CN109948473A (en) | A kind of method neural network based promoting student's applied problem solution topic ability | |
CN106875028A (en) | A kind of method for aiding in selecting a school and system | |
CN107819810A (en) | Adaptive planning formulates system | |
CN113657723A (en) | Knowledge point mastering condition diagnosis method, test question recommendation method and device | |
CN108073603A (en) | Operation distribution method and device | |
CN112951022A (en) | Multimedia interactive education training system | |
CN109783647A (en) | The construction method of intelligence learning model | |
CN112396545A (en) | Smart campus management system based on cloud platform | |
CN116797071A (en) | Course evaluation method with entropy weight-Topsis combined weighting improvement | |
CN110503346A (en) | School-based training quality evaluation platform, system and method based on data depth analysis | |
JP6395159B2 (en) | Education plan creation support apparatus and education plan creation method | |
CN109993441B (en) | Course weight matrix quantitative calculation method based on equivalent schooltime | |
CN108596461B (en) | Intelligent system and method for training effect evaluation | |
CN111241456A (en) | Learning resource allocation method and system |
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 |