CN109543963A - A kind of big data analysis method and system based on student's study habit - Google Patents
A kind of big data analysis method and system based on student's study habit Download PDFInfo
- Publication number
- CN109543963A CN109543963A CN201811313696.8A CN201811313696A CN109543963A CN 109543963 A CN109543963 A CN 109543963A CN 201811313696 A CN201811313696 A CN 201811313696A CN 109543963 A CN109543963 A CN 109543963A
- Authority
- CN
- China
- Prior art keywords
- value
- attribute
- attributive character
- analysis
- matching
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 61
- 238000007405 data analysis Methods 0.000 title claims abstract description 18
- 238000004458 analytical method Methods 0.000 claims abstract description 107
- 230000006399 behavior Effects 0.000 claims abstract description 20
- 230000000694 effects Effects 0.000 claims description 19
- 230000036541 health Effects 0.000 claims description 10
- 238000013500 data storage Methods 0.000 claims description 2
- 238000004088 simulation Methods 0.000 abstract description 8
- 238000011156 evaluation Methods 0.000 description 4
- 238000002474 experimental method Methods 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 238000013139 quantization Methods 0.000 description 3
- 238000013459 approach Methods 0.000 description 2
- 230000005764 inhibitory process Effects 0.000 description 2
- 230000033001 locomotion Effects 0.000 description 2
- 238000007726 management method Methods 0.000 description 2
- 238000012552 review Methods 0.000 description 2
- 206010000117 Abnormal behaviour Diseases 0.000 description 1
- 241001269238 Data Species 0.000 description 1
- PEDCQBHIVMGVHV-UHFFFAOYSA-N Glycerine Chemical compound OCC(O)CO PEDCQBHIVMGVHV-UHFFFAOYSA-N 0.000 description 1
- 230000003542 behavioural effect Effects 0.000 description 1
- 238000002512 chemotherapy Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 235000005911 diet Nutrition 0.000 description 1
- 230000037213 diet Effects 0.000 description 1
- 230000004069 differentiation Effects 0.000 description 1
- 235000013399 edible fruits Nutrition 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 235000013305 food Nutrition 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
- 208000024891 symptom Diseases 0.000 description 1
- 238000012549 training Methods 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
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
-
- 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
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Educational Administration (AREA)
- Economics (AREA)
- Tourism & Hospitality (AREA)
- General Physics & Mathematics (AREA)
- Educational Technology (AREA)
- Marketing (AREA)
- Theoretical Computer Science (AREA)
- Development Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Game Theory and Decision Science (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The present invention discloses a kind of big data analysis method and system based on student's study habit, which includes: step 1: acquisition student's basic data is simultaneously stored onto computer cluster, is classified using distributed computing to mass data;Step 2: constructing students ' behavior analysis model using sorted data: according to the attributive character value of each classification data and analysis dimension EkAll properties characteristic value matching value set and EkAttribute Association weighted value obtains EkThe judgement schematics of attribute;According to EkThe judgement schematics of attribute obtain student's prediction model using the linear least square of curve matching.The acquisition of knowledge that the application can not only excavate each student is horizontal, knowledge uses ability, mode of learning preference, the personalized situation such as be good at field, but also it can get the simulation and forecast curve of the student by student's prediction model, by carrying out parameter adjustment to the simulation and forecast curve, reasonably to suggest making it to obtain optimum efficiency to student.
Description
Technical field
The present invention relates to big data analysis field, specifically a kind of big data analysis method based on student's study habit.
Background technique
All using Intelligent campus as the key content of its informatization, campus big data technology conduct is most heavy for numerous colleges and universities
The technical support wanted can effectively promote Informatization Construction of Campus level and construct effects of biology.
Student is the main body of education activities, and classroom instruction is a most important link in students'learning.Student exists
How is subjective role performance in education activities, directly affects the height of quality of instruction.And the learning behavior for monitoring student is point
Analyse most effective, the most straightforward approach of students'learning.Therefore, studied by collecting and recording the learning behavior of student and
The learning actuality for analyzing student, is targetedly solved, and is the key that improve Talents Quality.Existing human body behavioural characteristic is known
Other device and method are mainly used in medical domain, security fields, Sports Field etc..Such as in medical domain, by analyzing patient
Characteristic behavior understand its symptom;In security fields, by judging it in the abnormal behavior of high sensitive area analysis human body, whether there is or not danger
Danger;In the various pedometers of Sports Field, by the amount of exercise etc. for analyzing the counting of human body paces human body.But these are
Disclosed apparatus and method cannot be acquired and analyze for student's learning behavior.
Based on this, the application proposes a kind of students ' behavior analysis model and the big data analysis method for the model,
The campus data of magnanimity, isomery, multidimensional are cleaned, are integrated, excavated and applied, potential, valuable, pole is therefrom extracted
Have the information of potential using value, provides the data of science for work in every such as the teaching of school, scientific research, logistics, management, securities
Support, this is for realizing that educating power's dream has important practical significance.
Summary of the invention
The brief overview about the embodiment of the present invention is given below, in order to provide about certain aspects of the invention
Basic comprehension.It should be appreciated that outlined below is not about exhaustive general introduction of the invention.It is not intended to determine this hair
Bright key or pith, nor is it intended to limit the scope of the present invention.Its purpose only provides certain in simplified form
A little concepts, taking this as a prelude to a more detailed description discussed later.
According to the one aspect of the application, a kind of big data analysis method based on student's study habit is provided, comprising:
Step 1: acquisition student's basic data storage utilizes distributed computing to the NoSQL database on computer cluster
Classify to mass data: learning data class, economic consumption class, Internet data class, life track class, in addition, its classification is also
It can be extended, be not limited to above-mentioned four class;
Step 2: constructing students ' behavior analysis model using sorted student's basic data, the specific steps are as follows:
Step 21: the attribute set of learning data class is denoted as A, A={ A1,A2,…Ai,…,Am};M is learning data class
Attribute number, wherein AiAttributive character value be denoted as { ai1,ai2,ai3,…,aiti, list is expressed as follows:
Learning data generic attribute | Attributive character value |
A1 | a11,a12,a13,…,a1t1 |
A2 | a21,a22,a23,…,a2t2 |
Ai | ai1,ai2,ai3,…,aiti |
… | … |
Am | am1,am2,am3,…,amtm |
Wherein, aijIt is the attribute A of learning data classiAttributive character value, aij(i=1 ..., m;J=1 ..., ti) under
In mark, i refers to that i-th of attribute, j refer to AiJ-th of attributive character value of attribute, tiIt is to indicate AiAttributive character value
Number;
Step 22: defining the E, E={ E that dimension is analyzed in student's Analysis model of network behaviors1,E2,…Ei,…,Ez, z is analysis
The number of the attribute of dimension;The analysis dimension E is including but not limited to course learning, health analysis, the attributes such as activity analysis;Its
In, EiAttributive character value be denoted as { ei1,ei2,ei3,…,eipi, list is expressed as follows:
Analyze dimension E | Attributive character value |
E1 | e11,e12,e13,…,e1p1 |
E2 | e21,e22,e23,…,e2p2 |
Ei | ei1,ei2,ei3,…,eipi |
… | … |
Ez | ez1,ez2,ez3,…,ezpz |
Wherein, eijIt is analysis dimension EiThe attributive character value enumerated, eij(i=1 ..., z;J=1 ..., pi) subscript
In, i refers to that i-th of analysis dimensional attribute, j refer to EiJ-th of attributive character value of attribute, piIt is to indicate EiAttributive character
The number of value;
Step 23: analyzing dimension in the attributive character value and students ' behavior analysis model of building learning data generic attribute set
E each characteristic value associated weights value and matching value set;
Wherein, building learning data generic attribute set A and analysis dimension EkThe characteristic value of (k=1 ..., z) attribute is associated with power
Weight values and matching value aggregate list are as follows:
Wherein, characteristic value method of discrimination Δi(i=1 ..., m) it is discrimination properties group Ai(i=1 ..., m) each characteristic value and Ek
Attributive character value matching degree method of discrimination;AMijFor according to characteristic value method of discrimination ΔiObtained AiAttributive character value
aijWith analysis dimension EkAll properties characteristic value matching value { am1,…,ampkSet, amqFor attribute AiAijFeature
Value and analysis dimension EkThe matching value of q-th of characteristic value of attribute;awikIt is attribute AiWith analysis dimension EkAttribute is in characteristic matching
Weighted value when calculating meets
Step 24: according to the A of step 2iAttributive character value aijWith analysis dimension EkAll properties characteristic value matching
It is worth { am1,…,ampkSet AMijAnd EkAttribute Association weighted value awik, obtain EkThe judgement schematics fk (x) of attribute;
Step 25: according to EkThe judgement schematics of attribute obtain student's prediction using the linear least square of curve matching
ModelγkFor undetermined coefficient (k=1 ..., z).
Further, the step 21 further include:
The attribute set of economic consumption class is denoted as B, B is denoted as { B1,B2,…Bi,…,Bn};N is the category of economic consumption class
The number of property, wherein BiAttributive character value be denoted as { bi1,bi2,bi3,…,biri, list is expressed as follows:
Economic consumption generic attribute | Attributive character value |
B1 | b11,b12,b13,…,b1r1 |
B2 | b21,b22,b23,…,b2r2 |
Bi | bi1,bi2,bi3,…,biri |
… | … |
Bn | bn1,bn2,bn3,…,bnrn |
Wherein, bijIt is the attribute B of economic consumption classiAttributive character value, bij(i=1 ..., n;J=1 ..., ri) under
In mark, i refers to that i-th of economic consumption generic attribute, j refer to BiJ-th of attributive character value of attribute, riIt is to indicate BiAttribute
The number of characteristic value;
The step 23 further include:
Construct the attributive character value and analysis dimension E of economic consumption generic attribute set BkThe feature of (k=1 ..., z) attribute
It is worth associated weights value and matching value set, is listed as follows:
Wherein, characteristic value method of discrimination Δm+i(i=1 ..., n) it is discrimination properties group BiThe attribute of (i=1 ..., n) and Ek
The method of discrimination of the matching degree of characteristic value;BMijFor according to characteristic value method of discrimination Δm+iObtained BiAttributive character value bijWith
Analyze dimension EkAll properties characteristic value matching value { bm1,…,bmpkSet, bmq(q=1 ..., pk) it is characterized BiBelong to
The b of propertyijCharacteristic value and analysis dimension EkThe matching value of q-th of characteristic value of attribute.bwikIt is attribute biWith analysis dimension EkAttribute
The weighted value in personal feature matching primitives meets
Further, the step 21 further include: the attribute set of Internet data class is denoted as C, C={ C1,C2,…
Ci,…,Co};O is the number of the attribute of Internet data class, wherein CiAttributive character value be denoted as { ci1,ci2,ci3,…,ciri},
List is expressed as follows (table 3):
Internet data generic attribute | Attributive character value |
C1 | c11,c12,c13,…,c1r1 |
C2 | c21,c22,c23,…,c2r2 |
Ci | ci1,ci2,ci3,…,ciri |
… | … |
Co | co1,co2,co3,…,coro |
Wherein, cijIt is the attribute C of Internet data classiAttributive character value, cij(i=1 ..., o;J=1 ..., ri) under
In mark, i refers to that the attribute of i-th of Internet data class, j refer to CiJ-th of attributive character value of attribute, riIt is to indicate CiCategory
The number of property characteristic value;
The step 23 further include:
Construct the attributive character value and analysis dimension E of Internet data generic attribute set CkThe feature of (k=1 ..., z) attribute
It is worth associated weights value and matching value aggregate list is as follows:
Wherein, characteristic value method of discrimination Δm+n+i(i=1 ..., o) it is discrimination properties group Ci(i=1 ..., o) and EkCategory
The method of discrimination of the matching degree of property characteristic value;CMijFor according to characteristic value method of discrimination Δm+n+iObtained CiAttributive character value
cijWith analysis dimension EkAll properties characteristic value matching value { cm1,…,cmpkSet, cmq(q=1 ..., pk) it is CiBelong to
The c of propertyijCharacteristic value and analysis dimension EkThe matching value of q-th of characteristic value of attribute;cwikIt is attribute ciWith analysis dimension EkAttribute
The weighted value in personal feature matching primitives meets
Further, the step 21 further include:
The attribute set for track class of living is denoted as D, D={ D1,D2,…Di,…,Dp};P is the attribute of life track class
Number, wherein DiAttributive character value be denoted as { di1,di2,di3,…,diri, list is expressed as follows:
Life track generic attribute | Attributive character value |
D1 | d11,d12,d13,…,d1r1 |
D2 | d21,d22,d23,…,d2r2 |
Di | di1,di2,di3,…,diri |
… | … |
Dp | dp1,dp2,dp3,…,dprp |
Wherein, dijIt is the attribute D of life track classiAttributive character value, dij(i=1 ..., n;J=1 ..., ri) under
In mark, i refers to that i-th of life track generic attribute, j refer to DiJ-th of attributive character value of attribute, riIt is to indicate DiAttribute
The number of characteristic value;
The step 23 further include:
The attributive character value and analysis dimension E of building life track generic attribute set DkThe feature of (k=1 ..., z) attribute
It is worth associated weights value and matching value aggregate list is following (table 7):
Wherein, characteristic value method of discrimination Δm+n+o+i(i=1 ..., n) it is discrimination properties group Di(i=1 ..., n) and EkCategory
The method of discrimination of the matching degree of property characteristic value;DMijFor according to characteristic value method of discrimination Δm+n+o+iObtained DiAttributive character value
dijWith analysis dimension EkAll properties characteristic value matching value { dm1,…,dmpkSet, dmq(q=1 ..., pk) be characterized
BiThe d of attributeijCharacteristic value and analysis dimension EkThe matching value of q-th of characteristic value of attribute;dwikIt is attribute diWith analysis dimension Ek
Attribute weighted value in personal feature matching primitives meets
Preferably, in step 24, EkThe judgement schematics of attribute include: that course learning judgement schematics f1 (x), health analysis are commented
Valence formula f2 (x) and activity analysis judgement schematics f3 (x);Then student's prediction model is F (x)=γ1·ef1(x)+γ2·ef2(x)+
γ3·ef3(x);Wherein, γ1、γ2And γ3Respectively undetermined coefficient.
The application is by constructing students ' behavior analysis model, no based on individual psychology and study analysis theory etc.
Acquisition of knowledge level, the mode of learning preference, extracurricular interest, joy of each student are only excavated by the judgement schematics of step 24
The personalized situation such as happy life, to realize more personalized campus administration and service, but also further passes through step
25 student's prediction model can get the simulation and forecast curve of the student, be made by carrying out parameter adjustment to the simulation and forecast curve
It obtains best trend, needs to reinforce in which aspect to obtain optimal effect, it can be achieved that thing half to be back-calculated to obtain the student
The effect of function times provides the instruction of quantization for the teaching of teacher, receiving an education for student, before having boundless market
Scape.
Specific embodiment
The embodiment of the present invention addressed below.The elements and features described in one embodiment of the invention can be with
The elements and features shown in one or more embodiments combine.It should be noted that for purposes of clarity, in explanation
The expression and description of component unrelated to the invention, known to persons of ordinary skill in the art and processing is omitted.
Students ' behavior can the behavior of day student campus (such as: the rate of attendance, to class rate, achievement, experiment, programming, track, consumption, on
The big datas such as net) quantified, the data after quantization can daily schedule regularity to student, level of effort, learning skill, warp
The various dimensions such as Ji situation are analyzed, and are finally reached and are disclosed students'growth track, are carried out student's academic warning, are precisely subsidized, just
The purpose of industry is recommended, so as to for school to student carry out the personalized educational management with precision and guidance provide it is important according to
According to.The acquisition of knowledge that the application can not only excavate each student is horizontal, knowledge using ability, mode of learning preference, be good at
The personalization such as field situation, but also can get by student's prediction model the simulation and forecast curve of the student, by the mould
Quasi- prediction curve carries out parameter adjustment, reasonably to suggest making it to obtain optimum efficiency to student.
Specifically, a kind of big data analysis method based on student's study habit of the invention, comprising:
Step 1: acquisition student's basic data is stored to the NoSQL database on computer cluster, and to these magnanimity bases
Plinth data carry out preliminary classification: learning data class, economic consumption class, Internet data class, life track class, in addition, its classification is also
It can be extended, such as artistic accomplishment class etc., subsequent prediction model increases and artistic accomplishment behavior continuous item;Acquire data tool
Body is to collect students ' behavior analysis model related data by terminal device, and terminal device includes but is not limited to that student passes through movement
The data of terminal upload, campus card consumption data, each access control system in campus (school gate gate inhibition, library's entrance guard, dormitory door
Taboo, laboratory gate inhibition etc.) acquisition data and campus in camera acquisition data.
Step 2: based on individual psychology and study analysis theory etc., sorted student's basic data being configured to
Students ' behavior analysis model, the specific steps are as follows:
Step 21: defining the E, E={ E that dimension is analyzed in student's Analysis model of network behaviors1,E2,…Ei,…,Ez, z is analysis
The number of the attribute of dimension;The analysis dimension E is including but not limited to course learning (each section learns situation), health analysis (diet
Habit, motion conditions, work and rest rule), (community activity goes out school situation, competition, experimental conditions, programming energy for activity analysis
Power) etc. attributes;Wherein, EiAttributive character value be denoted as { ei1,ei2,ei3,…,eipi, list is expressed as follows (table 1):
Analyze dimension E | Attributive character value |
E1 | e11,e12,e13,…,e1p1 |
E2 | e21,e22,e23,…,e2p2 |
Ei | ei1,ei2,ei3,…,eipi |
… | … |
Ez | ez1,ez2,ez3,…,ezpz |
Table 1
Wherein, eijIt is analysis dimension EiThe attributive character value enumerated, eij(i=1 ..., z;J=1 ..., pi) subscript
In, i refers to that i-th of analysis dimensional attribute, j refer to EiJ-th of attributive character value of attribute, piIt is to indicate EiAttributive character
The number of value.Such as EiFor activity analysis, eijIt can be community activity, go out school situation, competition, experimental conditions, programming energy
Power etc..
Step 22: by the attribute set of learning data class (such as the rate of attendance, punctuality rate, coursework performance, examination
Achievement reviews one's lessons duration, read a book duration, experimental activity etc.) it is denoted as A, A={ A1,A2,…Ai,…,Am};M is learning data class
The number of attribute, wherein AiAttributive character value be denoted as { ai1,ai2,ai3,…,aiti, list is expressed as follows (table 2):
Table 2
Wherein, aijIt is the attribute A of learning data classiAttributive character value, aij(i=1 ..., m;J=1 ..., ti) under
In mark, i refers to that i-th of attribute, j refer to AiJ-th of attributive character value of attribute, tiIt is to indicate AiAttributive character value
Number;Characteristic value method of discrimination Δi(i=1 ..., m) it is discrimination properties group Ai(i=1 ..., m) and EkAttributive character value matching
The method of discrimination of degree;Associated data set is method of discrimination ΔiUsed in related data set;Characteristic value discrimination standard σijIt is
Corresponding attribute AiConfirm each characteristic value aij(i=1 ..., m;J=1 ..., ti) judgment criteria value.For example, learning data class
Attribute includes each subject rate of attendance (such as physics lesson, computer, high number etc.), each subject punctuality rate (be late, leave early), each
Section's operation performance, each subject related experiment start data, each subject achievement, review one's lessons duration, library's reading duration etc.,
If AiFor the physics lesson rate of attendance, AiAttribute include that physics lesson be late number of days, physics lesson of number of days, physics lesson of turning out for work is left early number of days etc.
Deng if aijIt is that physics lesson is turned out for work number of days, ΔiIt is to judge the physics lesson rate of attendance and corresponding EkAttributive character value matching degree
Judgment method, be usually a threshold interval (threshold interval under students ' behavior analysis model initial situation is preset value,
Later period students ' behavior analysis module can adaptively be adjusted according to clustering algorithm, can also manual modification), such as work as EkAttribute
When characteristic value is Physics Course achievement, then ΔiIt can be set as [0.6,0.9], work as EkAttributive character value be chemo achievement when,
Then ΔiIt can be set as [0.02,0.3] etc.;Associated data set RiIt is method of discrimination ΔiUsed in dependent thresholds section data
Set, proposes associated data set here, is subsequent data training for convenience, so that model is optimized;Characteristic value differentiates
Standard σijIt is corresponding attribute AiConfirm the judgment criteria value of each characteristic value.
The attribute set (such as economic situation, moon spending limit, consumption number of times, consumer applications etc.) of economic consumption class is denoted as
B, B are denoted as { B1,B2,…Bi,…,Bn};N is the number of the attribute of economic consumption class, wherein BiAttributive character value be denoted as
{bi1,bi2,bi3,…,biri, list is expressed as follows (table 3):
Table 3
Wherein, bijIt is the attribute B of economic consumption classiAttributive character value, bij(i=1 ..., n;J=1 ..., ri) under
In mark, i refers to that i-th of economic consumption generic attribute, j refer to BiJ-th of attributive character value of attribute, riIt is to indicate BiAttribute
The number of characteristic value.Characteristic value method of discrimination Δm+i(i=1 ..., n) it is discrimination properties group BiThe attribute of (i=1 ..., n) and Ek
The method of discrimination of the matching degree of characteristic value;Associated data set is method of discrimination Δm+iUsed in related data set;Characteristic value
Discrimination standard ηijIt is corresponding BiAttribute confirms each characteristic value bijDiscrimination standard value.
Attribute (such as online duration, the game, video etc.) set of Internet data class is denoted as C, C={ C1,C2,…Ci,…,
Co};O is the number of the attribute of Internet data class, wherein CiAttributive character value be denoted as { ci1,ci2,ci3,…,ciri, list table
Show following (table 4):
Table 4
Wherein, cijIt is the attribute C of Internet data classiAttributive character value, cij(i=1 ..., o;J=1 ..., ri) under
In mark, i refers to that the attribute of i-th of Internet data class, j refer to CiJ-th of attributive character value of attribute, riIt is to indicate CiCategory
The number of property characteristic value.Characteristic value method of discrimination Δm+i(i=1 ..., o) it is discrimination properties group Ci(i=1 ..., o) and EkCategory
The method of discrimination of the matching degree of property characteristic value;Associated data set is method of discrimination Δm+n+iUsed in related data set;Feature
It is worth discrimination standard λijIt is corresponding BiAttribute confirms each characteristic value bijDiscrimination standard value.
Attribute (such as food and drink, the work and rest, online duration etc.) set of life track class is denoted as D, D={ D1,D2,…Di,…,
Dp};P is the number of the attribute of life track class, wherein DiAttributive character value be denoted as { di1,di2,di3,…,diri, list table
Show following (table 5):
Table 5
Wherein, dijIt is the attribute D of life track classiAttributive character value, dij(i=1 ..., n;J=1 ..., ri) under
In mark, i refers to that i-th of life track generic attribute, j refer to DiJ-th of attributive character value of attribute, riIt is to indicate DiAttribute
The number of characteristic value.Characteristic value method of discrimination Δm+n+o+i(i=1 ..., n) it is discrimination properties group DiThe category of (i=1 ..., n) and Ek
The method of discrimination of the matching degree of property characteristic value;Associated data set is method of discrimination Δm+iUsed in related data set;Feature
It is worth discrimination standard μijIt is corresponding DiAttribute confirms each characteristic value dijDiscrimination standard value.
Step 23: according to individual psychology and study analysis theory, constructing learning data generic attribute set, economy respectively
Consumer attribute set, Internet data generic attribute set, the attributive character value for track generic attribute set of living and students ' behavior point
Analyse each characteristic value associated weights value that the E of dimension is analyzed in model and matching value set;
Wherein, building learning data generic attribute set A and analysis dimension EkThe characteristic value of (k=1 ..., z) attribute is associated with power
Weight values and matching value aggregate list are following (table 6):
Table 6
Wherein AMijFor AiAttributive character value aijWith analysis dimension EkAll properties characteristic value matching value { am1,…,
ampkSet, amqFor attribute AiAijCharacteristic value and analysis dimension EkThe matching value of q-th of characteristic value of attribute;awikIt is to belong to
Property AiWith analysis dimension EkAttribute weighted value when characteristic matching calculates, meets
Construct the attributive character value and analysis dimension E of economic consumption generic attribute set BkThe feature of (k=1 ..., z) attribute
It is worth associated weights value and matching value set, is listed as follows (table 7):
Table 7
Wherein BMijFor BiAttributive character value bijWith analysis dimension EkAll properties characteristic value matching value { bm1,…,
bmpkSet, bmq(q=1 ..., pk) it is characterized BiThe b of attributeijCharacteristic value and analysis dimension EkQ-th of characteristic value of attribute
Matching value.bwikIt is attribute biWith analysis dimension EkAttribute weighted value in personal feature matching primitives meets
Construct the attributive character value and analysis dimension E of Internet data generic attribute set CkThe feature of (k=1 ..., z) attribute
It is worth associated weights value and matching value aggregate list is following (table 8):
Table 8
Wherein CMijFor CiAttributive character value cijWith analysis dimension EkAll properties characteristic value matching value { cm1,…,
cmpkSet, cmq(q=1 ..., pk) it is CiThe c of attributeijCharacteristic value and analysis dimension EkOf q-th of characteristic value of attribute
With value.cwikIt is attribute ciWith analysis dimension EkAttribute weighted value in personal feature matching primitives meets
The attributive character value and analysis dimension E of building life track generic attribute set DkThe feature of (k=1 ..., z) attribute
It is worth associated weights value and matching value aggregate list is following (table 9):
Table 9
Wherein DMijFor DiAttributive character value dijWith analysis dimension EkAll properties characteristic value matching value { dm1,…,
dmpkSet, dmq(q=1 ..., pk) it is characterized BiThe d of attributeijCharacteristic value and analysis dimension EkQ-th of characteristic value of attribute
Matching value.dwikIt is attribute diWith analysis dimension EkAttribute weighted value in personal feature matching primitives meets
Step 24: according to the A of step 2iAttributive character value aijWith analysis dimension EkAll properties characteristic value matching
It is worth { am1,…,ampkSet AMijAnd EkAttribute Association weighted value awik, obtain EkThe judgement schematics fk (x) of attribute;
Step 25: according to EkThe judgement schematics of attribute obtain student's prediction using the linear least square of curve matching
ModelγkFor undetermined coefficient (k=1 ..., z).Student's prediction model therein is by advance most
Small square law carries out data fitting to realize, the determination of specific coefficient is to approach experiment by repeatedly simulating the limit to obtain
, and will not be described here in detail for specific process.
Wherein fk (x) is the judgement schematics of different analysis dimensions, for example, EkAttribute include course learning, health analysis,
Activity analysis, then EkThe judgement schematics of attribute include: course learning judgement schematics f1 (x), health analysis judgement schematics f2 (x) and
Activity analysis judgement schematics f3 (x);So student's prediction model is F (x)=γ1·ef1(x)+γ2·ef2(x)+γ3·ef3(x);
Wherein, γ1、γ2And γ3Respectively undetermined coefficient can be preset.
In the present invention, fk (x) is realized using the weighting of the attribute value of respective weights:
Course learning judgement schematics Wherein ω
11, ω 12, ω 13 and ω 14 are respectively undetermined coefficient, can be preset.F1 (x) can be normalized or not normalize, can
It is set according to practical calculating.For example, wherein course learning f1 (x) include because being known as study habit, physical fitness (is
No health), work and rest habit etc., first according to characteristic value method of discrimination Δ obtain with each EkAttributive character value matching value collection
It closes, namely relevant data acquisition system is obtained by screening, then according to EkAttribute Association weighted value is weighted summation;Secondly,
Different EkAttribute incidence coefficient it is also different.
Health analysis judgement schematics Wherein ω 21, ω 22, ω 23 and ω 24 are respectively undetermined coefficient, can be preset.
Activity analysis judgement schematics Wherein ω 31, ω 32, ω 33 and ω 34 are respectively undetermined coefficient, can be preset.
Each of through the above steps, students ' behavior is subjected to differentiation attribute description respectively, can facilitate with analysis model
Analysis dimension is evaluated respectively, in use, corresponding evaluation index can be arranged to each analysis dimension in advance, then according to not
The evaluation of estimate of the student is calculated with the different evaluation formula of analysis dimension, then is compared with corresponding evaluation index
Obtain superiority and inferiority situation of the student under the analysis dimension.In addition, can also student's prediction model can get the student simulation it is pre-
Curve is surveyed, such as determines that student's prediction model is the simulation and forecast curve of final grade by undetermined coefficient, by the simulation
Prediction curve, which carries out parameter adjustment, makes it obtain optimum efficiency, needs to reinforce in which aspect to obtain to be back-calculated to obtain the student
Optimal final grade provides optimal striving direction and accurately instruction to student, it can be achieved that the effect got twice the result with half the effort
Fruit provides the instruction of quantization for the teaching of teacher, receiving an education for student, has boundless market prospects.
Method of the invention be not limited to specifications described in time sequencing execute, when can also according to others
Between sequentially, in parallel or independently execute.Therefore, the execution sequence of method described in this specification is not to skill of the invention
Art range is construed as limiting.
Although being had been disclosed above by the description to specific embodiments of the present invention to the present invention, it answers
The understanding, above-mentioned all embodiments and example are exemplary, and not restrictive.Those skilled in the art can be in institute
Design is to various modifications of the invention, improvement or equivalent in attached spirit and scope of the claims.These modification, improve or
Person's equivalent should also be as being to be considered as included in protection scope of the present invention.
Claims (9)
1. a kind of big data analysis method based on student's study habit, comprising:
Step 1: acquisition student's basic data storage to the NoSQL database on computer cluster, and to magnanimity basic data into
Row preliminary classification: learning data class, economic consumption class, Internet data class, life track class;
Step 2: constructing students ' behavior analysis model using sorted student's basic data, the specific steps are as follows:
Step 21: the attribute set of learning data class is denoted as A, A={ A1,A2,…Ai,…,Am};M is the category of learning data class
The number of property, wherein AiAttributive character value be denoted as { ai1,ai2,ai3,…,aiti, list is expressed as follows:
Wherein, aijIt is the attribute A of learning data classiAttributive character value, aij(i=1 ..., m;J=1 ..., ti) subscript in,
I refers to that i-th of attribute, j refer to AiJ-th of attributive character value of attribute, tiIt is to indicate AiAttributive character value number;
Step 22: defining the E, E={ E that dimension is analyzed in student's Analysis model of network behaviors1,E2,…Ei,…,Ez, z is analysis dimension
Attribute number;The analysis dimension E is including but not limited to course learning, health analysis, the attributes such as activity analysis;Wherein, Ei
Attributive character value be denoted as { ei1,ei2,ei3,…,eipi, list is expressed as follows:
Wherein, eijIt is analysis dimension EiThe attributive character value enumerated, eij(i=1 ..., z;J=1 ..., pi) subscript in, i
Refer to that i-th of analysis dimensional attribute, j refer to EiJ-th of attributive character value of attribute, piIt is to indicate EiAttributive character value
Number;
Step 23: the E of dimension is analyzed in the attributive character value and students ' behavior analysis model of building learning data generic attribute set
Each characteristic value associated weights value and matching value set;
Wherein, building learning data generic attribute set A and analysis dimension EkThe characteristic value associated weights value of (k=1 ..., z) attribute
It is as follows with matching value aggregate list:
Wherein, characteristic value method of discrimination Δi(i=1 ..., m) it is discrimination properties group Ai(i=1 ..., m) each characteristic value and EkCategory
The method of discrimination of the matching degree of property characteristic value;AMijFor according to characteristic value method of discrimination ΔiObtained AiAttributive character value aijWith
Analyze dimension EkAll properties characteristic value matching value { am1,…,ampkSet, amqFor attribute AiAijCharacteristic value with point
Analyse dimension EkThe matching value of q-th of characteristic value of attribute;awikIt is attribute AiWith analysis dimension EkAttribute is when characteristic matching calculates
Weighted value meets
Step 24: according to the A of step 2iAttributive character value aijWith analysis dimension EkAll properties characteristic value matching value
{am1,…,ampkSet AMijAnd EkAttribute Association weighted value awik, obtain EkThe judgement schematics fk (x) of attribute;
Step 25: according to EkThe judgement schematics of attribute obtain student's prediction model using the linear least square of curve matchingγkFor undetermined coefficient (k=1 ..., z).
2. big data analysis method according to claim 1, it is characterised in that: the step 21 further include: economy disappears
The attribute set of expense class is denoted as B, and B is denoted as { B1,B2,…Bi,…,Bn};N is the number of the attribute of economic consumption class, wherein Bi's
Attributive character value is denoted as { bi1,bi2,bi3,…,biri, list is expressed as follows:
Wherein, bijIt is the attribute B of economic consumption classiAttributive character value, bij(i=1 ..., n;J=1 ..., ri) subscript in,
I refers to that i-th of economic consumption generic attribute, j refer to BiJ-th of attributive character value of attribute, riIt is to indicate BiAttributive character value
Number;
The step 23 further include:
Construct the attributive character value and analysis dimension E of economic consumption generic attribute set BkThe characteristic value association of (k=1 ..., z) attribute
Weighted value and matching value set, are listed as follows:
Wherein, characteristic value method of discrimination Δm+i(i=1 ..., n) it is discrimination properties group BiThe attributive character of (i=1 ..., n) and Ek
The method of discrimination of the matching degree of value;BMijFor according to characteristic value method of discrimination Δm+iObtained BiAttributive character value bijWith analysis
Dimension EkAll properties characteristic value matching value { bm1,…,bmpkSet, bmq(q=1 ..., pk) it is characterized BiAttribute
bijCharacteristic value and analysis dimension EkThe matching value of q-th of characteristic value of attribute;bwikIt is attribute biWith analysis dimension EkAttribute is a
Weighted value when body characteristics matching primitives meets
3. big data analysis method according to claim 2, it is characterised in that: the step 21 further include: by upper netting index
C, C={ C are denoted as according to the attribute set of class1,C2,…Ci,…,Co};O is the number of the attribute of Internet data class, wherein CiCategory
Property characteristic value is denoted as { ci1,ci2,ci3,…,ciri, list is expressed as follows (table 3):
Wherein, cijIt is the attribute C of Internet data classiAttributive character value, cij(i=1 ..., o;J=1 ..., ri) subscript in,
I refers to that the attribute of i-th of Internet data class, j refer to CiJ-th of attributive character value of attribute, riIt is to indicate CiAttributive character
The number of value;
The step 23 further include:
Construct the attributive character value and analysis dimension E of Internet data generic attribute set CkThe characteristic value association of (k=1 ..., z) attribute
Weighted value and matching value aggregate list are as follows:
Wherein, characteristic value method of discrimination Δm+n+i(i=1 ..., o) it is discrimination properties group Ci(i=1 ..., o) and EkAttributive character
The method of discrimination of the matching degree of value;CMijFor according to characteristic value method of discrimination Δm+n+iObtained CiAttributive character value cijWith point
Analyse dimension EkAll properties characteristic value matching value { cm1,…,cmpkSet, cmq(q=1 ..., pk) it is CiThe c of attributeij
Characteristic value and analysis dimension EkThe matching value of q-th of characteristic value of attribute;cwikIt is attribute ciWith analysis dimension EkAttribute is in individual
Weighted value when characteristic matching calculates meets
4. big data analysis method according to claim 3, it is characterised in that: the step 21 further include: by life rail
The attribute set of mark class is denoted as D, D={ D1,D2,…Di,…,Dp};P is the number of the attribute of life track class, wherein DiCategory
Property characteristic value is denoted as { di1,di2,di3,…,diri, list is expressed as follows:
Wherein, dijIt is the attribute D of life track classiAttributive character value, dij(i=1 ..., n;J=1 ..., ri) subscript in,
I refers to that i-th of life track generic attribute, j refer to DiJ-th of attributive character value of attribute, riIt is to indicate DiAttributive character value
Number;
The step 23 further include:
The attributive character value and analysis dimension E of building life track generic attribute set DkThe characteristic value association of (k=1 ..., z) attribute
Weighted value and matching value aggregate list are following (table 7):
Wherein, characteristic value method of discrimination Δm+n+o+i(i=1 ..., n) it is discrimination properties group Di(i=1 ..., n) and EkAttribute it is special
The method of discrimination of the matching degree of value indicative;DMijFor according to characteristic value method of discrimination Δm+n+o+iObtained DiAttributive character value dijWith
Analyze dimension EkAll properties characteristic value matching value { dm1,…,dmpkSet, dmq(q=1 ..., pk) it is characterized BiBelong to
The d of propertyijCharacteristic value and analysis dimension EkThe matching value of q-th of characteristic value of attribute;dwikIt is attribute diWith analysis dimension EkAttribute
The weighted value in personal feature matching primitives meets
5. big data analysis method according to claim 4, it is characterised in that: in step 24, EkThe judgement schematics packet of attribute
It includes: course learning judgement schematics f1 (x), health analysis judgement schematics f2 (x) and activity analysis judgement schematics f3 (x);
In step 25, student's prediction model is F (x)=γ1·ef1(x)+γ2·ef2(x)+γ3·ef3(x);Wherein, γ1、γ2With
γ3Respectively undetermined coefficient.
6. big data analysis method according to claim 5, it is characterised in that:
Course learning judgement schematics Wherein ω 11, ω 12, ω 13 and ω 14 are respectively undetermined coefficient.
7. big data analysis method according to claim 5, it is characterised in that:
Health analysis judgement schematics Wherein ω 21, ω 22, ω 23 and ω 24 are respectively undetermined coefficient.
8. big data analysis method according to claim 5, it is characterised in that:
Activity analysis judgement schematics f3 (x)=; Wherein ω 31, ω 32, ω 33 and ω 34 are respectively undetermined coefficient.
9. a kind of big data analysis system based on student's study habit, it is characterised in that: the analysis system uses claim
The big data analysis method of 1-8 is realized.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811313696.8A CN109543963B (en) | 2018-11-06 | 2018-11-06 | Big data analysis method and system based on student learning habits |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811313696.8A CN109543963B (en) | 2018-11-06 | 2018-11-06 | Big data analysis method and system based on student learning habits |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109543963A true CN109543963A (en) | 2019-03-29 |
CN109543963B CN109543963B (en) | 2022-01-11 |
Family
ID=65845074
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811313696.8A Active CN109543963B (en) | 2018-11-06 | 2018-11-06 | Big data analysis method and system based on student learning habits |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109543963B (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110060538A (en) * | 2019-04-08 | 2019-07-26 | 上海云之驾科技股份有限公司 | Personalized artificial based on historical data modeling intelligently drives training and practices system and method |
CN110072191A (en) * | 2019-04-23 | 2019-07-30 | 安徽致远慧联电子科技有限公司 | Track analysis system and analysis method in school based on wireless technology |
CN110083641A (en) * | 2019-04-26 | 2019-08-02 | 广州大学 | Intelligence analysis method and device based on goal behavior |
CN111584020A (en) * | 2020-04-30 | 2020-08-25 | 张建春 | Personal healthy life big data acquisition and analysis system |
CN111863266A (en) * | 2020-01-16 | 2020-10-30 | 上海海事大学 | Method for screening risk factors of sporadic colorectal adenomas based on directional weighted association rule model |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20020087551A1 (en) * | 2000-11-01 | 2002-07-04 | Hickey Matthew W. | Automatic data transmission in response to content of electronic forms satisfying criteria |
CN103886026B (en) * | 2014-02-25 | 2017-09-05 | 厦门客来点信息科技有限公司 | Clothing matching method based on personal feature |
CN107180284A (en) * | 2017-07-07 | 2017-09-19 | 北京航空航天大学 | A kind of SPOC student based on learning behavior feature shows weekly Forecasting Methodology and device |
CN107967572A (en) * | 2017-12-15 | 2018-04-27 | 华中师范大学 | A kind of intelligent server based on education big data |
CN108121785A (en) * | 2017-12-15 | 2018-06-05 | 华中师范大学 | A kind of analysis method based on education big data |
-
2018
- 2018-11-06 CN CN201811313696.8A patent/CN109543963B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20020087551A1 (en) * | 2000-11-01 | 2002-07-04 | Hickey Matthew W. | Automatic data transmission in response to content of electronic forms satisfying criteria |
CN103886026B (en) * | 2014-02-25 | 2017-09-05 | 厦门客来点信息科技有限公司 | Clothing matching method based on personal feature |
CN107180284A (en) * | 2017-07-07 | 2017-09-19 | 北京航空航天大学 | A kind of SPOC student based on learning behavior feature shows weekly Forecasting Methodology and device |
CN107967572A (en) * | 2017-12-15 | 2018-04-27 | 华中师范大学 | A kind of intelligent server based on education big data |
CN108121785A (en) * | 2017-12-15 | 2018-06-05 | 华中师范大学 | A kind of analysis method based on education big data |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110060538A (en) * | 2019-04-08 | 2019-07-26 | 上海云之驾科技股份有限公司 | Personalized artificial based on historical data modeling intelligently drives training and practices system and method |
CN110072191A (en) * | 2019-04-23 | 2019-07-30 | 安徽致远慧联电子科技有限公司 | Track analysis system and analysis method in school based on wireless technology |
CN110072191B (en) * | 2019-04-23 | 2021-01-12 | 安徽致远慧联电子科技有限公司 | Student in-school trajectory analysis system and method based on wireless technology |
CN110083641A (en) * | 2019-04-26 | 2019-08-02 | 广州大学 | Intelligence analysis method and device based on goal behavior |
CN111863266A (en) * | 2020-01-16 | 2020-10-30 | 上海海事大学 | Method for screening risk factors of sporadic colorectal adenomas based on directional weighted association rule model |
CN111863266B (en) * | 2020-01-16 | 2023-09-19 | 上海海事大学 | Dangerous factor screening method for sporadic colorectal adenoma based on directional weighted association rule model |
CN111584020A (en) * | 2020-04-30 | 2020-08-25 | 张建春 | Personal healthy life big data acquisition and analysis system |
Also Published As
Publication number | Publication date |
---|---|
CN109543963B (en) | 2022-01-11 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109543963A (en) | A kind of big data analysis method and system based on student's study habit | |
Konstantopoulos | Trends of school effects on student achievement: Evidence from NLS: 72, HSB: 82, and NELS: 92 | |
CN108596420A (en) | A kind of talent assessment system and method for Behavior-based control | |
Abrams | Building criminal capital vs. specific deterrence: The effect of incarceration length on recidivism | |
Nzewi et al. | Time management and academic performance of postgraduate students in Nigerian Universities | |
CN106373057A (en) | Network education-orientated poor learner identification method | |
CN110111894A (en) | The mental health method for early warning of behavior is serialized based on deep learning student | |
Ahmad et al. | Principal component analysis and self-organizing map clustering for student browsing behaviour analysis | |
CN109523088A (en) | The abnormal behaviour forecasting system of forced quarantine addict received treatment based on big data | |
Fachinelli et al. | Measuring the capital systems categories: the perspective of an integrated value system of social life as perceived by young citizens | |
Oreški et al. | Estimating profile of successful IT student: data mining approach | |
CN108881157A (en) | A kind of individual information security capabilities evaluation method and system based on PC terminal behavior | |
Corr et al. | Programmatic Accreditation in Sport Management and the Absence of Perceived Quality. | |
Sirotin et al. | Regional structure of the country on costs and results of innovative activity: the case of the Russian Federation | |
Feng et al. | Construction of Students' Comprehensive Quality Evaluation Model Based on Improved AHP | |
CN110414769A (en) | Student's career evaluation system | |
Ali S et al. | Multi-class LDA classifier and CNN feature extraction for student performance analysis during Covid-19 pandemic | |
Sasani et al. | Designing and Explaining the Biorhythm Management Model of Higher Education Managers based on Fuzzy Approach | |
Wang | Construction of Assessment Index System Model of Higher Art Teaching Quality Based on Fuzzy Clustering Algorithm | |
Zhang et al. | Machine learning based teaching quality evaluation | |
Zhou et al. | The Construction of Teaching Ability Evaluation System of Badminton in Physical Education Major | |
Ghalenovy et al. | Multiple Relationships between Successful Intelligence and Self-Regulated Learning Dimensions: Comparing Gifted and Ordinary Students in Mashhad | |
Wang et al. | Research on influencing factors of college students' academic security based on campus data | |
Luan et al. | Measuring the Higher Education System | |
Oza et al. | Student Learning Behavior: an Artificial Neural Network approach |
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 |