CN117151948B - College student outgoing check-in comprehensive management method and system - Google Patents
College student outgoing check-in comprehensive management method and system Download PDFInfo
- Publication number
- CN117151948B CN117151948B CN202311422091.3A CN202311422091A CN117151948B CN 117151948 B CN117151948 B CN 117151948B CN 202311422091 A CN202311422091 A CN 202311422091A CN 117151948 B CN117151948 B CN 117151948B
- Authority
- CN
- China
- Prior art keywords
- student
- data
- check
- time scale
- optimal
- 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
- 238000007726 management method Methods 0.000 title claims abstract description 24
- 230000007613 environmental effect Effects 0.000 claims abstract description 71
- 238000000034 method Methods 0.000 claims abstract description 50
- 238000007906 compression Methods 0.000 claims abstract description 27
- 230000006835 compression Effects 0.000 claims abstract description 25
- 238000004364 calculation method Methods 0.000 claims description 26
- 238000005457 optimization Methods 0.000 claims description 9
- YHXISWVBGDMDLQ-UHFFFAOYSA-N moclobemide Chemical compound C1=CC(Cl)=CC=C1C(=O)NCCN1CCOCC1 YHXISWVBGDMDLQ-UHFFFAOYSA-N 0.000 claims description 8
- 238000010606 normalization Methods 0.000 claims description 8
- 230000003044 adaptive effect Effects 0.000 claims description 6
- 238000004458 analytical method Methods 0.000 claims description 6
- 238000000354 decomposition reaction Methods 0.000 claims description 5
- 238000013144 data compression Methods 0.000 claims description 4
- 238000012545 processing Methods 0.000 claims description 4
- 238000013523 data management Methods 0.000 abstract description 2
- 230000008569 process Effects 0.000 description 9
- 230000008859 change Effects 0.000 description 4
- 230000000694 effects Effects 0.000 description 4
- 238000012512 characterization method Methods 0.000 description 3
- 238000005070 sampling Methods 0.000 description 3
- 238000010276 construction Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000011002 quantification Methods 0.000 description 2
- 238000006467 substitution reaction Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 230000011218 segmentation 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
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Educational Administration (AREA)
- Tourism & Hospitality (AREA)
- Strategic Management (AREA)
- Educational Technology (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Business, Economics & Management (AREA)
- Economics (AREA)
- Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Marketing (AREA)
- Human Resources & Organizations (AREA)
- General Health & Medical Sciences (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to the technical field of data management, and provides a comprehensive management method and system for attendance of college students, wherein the method comprises the following steps: collecting a plurality of check-in data of a student going out and environmental data of multiple dimensions; clustering to obtain a plurality of student categories according to the check-in data of different students; acquiring the optimal time scale of each student category according to the check-in data of the same student category and different time scales; obtaining the optimal period length for each student according to the data segments obtained by the check-in data under different period lengths and the environmental data of multiple dimensions under the optimal time scale of the class of the student to which each student belongs; and acquiring the self-adaptive self-correlation coefficient of each student according to the optimal cycle length, constructing a prediction model, and performing prediction coding compression on check-in data of each student. The invention aims to solve the problem that the comprehensive management efficiency of data is affected due to data redundancy caused by a large number of student check-in data.
Description
Technical Field
The invention relates to the technical field of data management, in particular to a comprehensive management method and system for attendance of college students.
Background
Under the current educational environment, student safety and management are very concerned by universities, and many schools are looking for more effective and more accurate methods for tracking and managing the extrados activities of students; traditional student management modes, such as manual roll-call, check-in list and the like, can meet the demands to a certain extent, but have obvious limitations, such as time consumption, easy error, incapability of real-time tracking and the like, and are difficult to more closely monitor the health condition and travel of students; by constructing the student out check-in system of the university, a large amount of check-in data occupy a large amount of storage space, so that the storage space is tense, and the data processing and inquiring speed is reduced; meanwhile, long-term storage of raw check-in data without sorting or compression may cause problems of data redundancy and low information retrieval efficiency.
Since the student check-in data is only time-varying and the time format has strong regularity, a large redundancy of storage space can occur in the compression process; meanwhile, as the format change of time is regular, the student outgoing check-in data is compressed by adopting a predictive coding method in the prior art; the traditional predictive coding compression method uses an ARIMA predictive model for compression, however, in the construction process of the predictive model, the autocorrelation coefficient determines whether the predictive effect of the predictive model is good or bad, if the autocorrelation coefficient is unreasonably set, the compression efficiency is greatly reduced, and the management efficiency of the student outgoing sign-in data is further affected.
Disclosure of Invention
The invention provides a comprehensive management method and system for attendance of students in colleges and universities, which aims to solve the problem that the efficiency of comprehensive management of data is affected due to data redundancy caused by the attendance data of a large number of students in the prior art, and adopts the following technical scheme:
in a first aspect, an embodiment of the present invention provides a method for comprehensively managing attendance of students in colleges and universities, including the steps of:
collecting a plurality of check-in data of a plurality of students going out in different time scales and environment data in a plurality of dimensions; the check-in data comprise check-in times counted in each unit time length under different time scales, and the place of each check-in is recorded;
according to the check-in data of different students, obtaining similar distances among the students, and clustering to obtain a plurality of student categories; acquiring the optimal time scale of each student category according to the check-in data of the same student category and different time scales;
acquiring the optimization degree of each cycle length of each student and acquiring the optimal cycle length according to the data segments obtained by the check-in data under different cycle lengths and the environmental data of multiple dimensions under the optimal time scale of the class of the student to which each student belongs;
And acquiring the self-adaptive self-correlation coefficient of each student according to the optimal cycle length, constructing a prediction model, and performing prediction coding compression on check-in data of each student.
Further, the method for obtaining the similar distance between the students specifically comprises the following steps:
according to the check-in data and the check-in places of each student in different time scales, the check-in frequency and the check-in interval factor of each student in each check-in place of each student in each time scale are obtained; first, theStudents and the firstSimilar distance of individual studentsThe calculation method of (1) is as follows:
;
wherein,represent the firstStudents and the firstThe number of places to which the individual students have checked in;representing the number of time scales;represent the firstUnder the individual time scalesThe first student isThe frequency of check-ins for the individual sites,represent the firstUnder the individual time scalesThe first student isCheck-in frequency of the individual places;represent the firstUnder the individual time scalesThe first student isA check-in interval factor for the individual sites;represent the firstUnder the individual time scalesThe first student isCheck-in interval factor for individual places.
Furthermore, the check-in frequency and the check-in interval factor of each student at each checked-in place under each time scale are specifically obtained by the following steps:
For any place where any student and any student check-in, acquiring a plurality of check-ins of the student at the place, acquiring corresponding moments of each check-in at different time scales, and for any time scale, calculating the ratio of the check-in times of the student at the place to the unit time length of the time scale, and recording the ratio as the check-in frequency of the student at the place under the time scale;
calculating time intervals for two adjacent check-ins of the place under the time scale to obtain a plurality of time intervals, obtaining the ratio of each time interval to the unit time length of the time scale, marking the ratio as a plurality of check-in intervals of the student at the place under the time scale, calculating variances for all the check-in intervals, and marking the variances as check-in interval factors of the student at the place under the time scale.
Further, the optimal time scale of each student category is specifically obtained by the following steps:
according to the check-in data of each student in different time scales, a plurality of ICA components and check-in rule degrees of each student in each time scale are obtained; for the firstClass of students, the firstOptimum degree of each time scale to the student classThe calculation method of (1) is as follows:
;
Wherein,represent the firstThe number of students in a single student class,represent the firstThe first student categoryPersonal studyIs produced at the firstThe degree of regularity of check-in on a time scale,the representation includes the firstThe number of time scale combinations of the individual time scales,the representation includes the firstThe first time scaleThe absolute value of the difference between the two time scale lengths in the time scale combinations,representing the maximum value of the length of the time scale,the representation includes the firstThe first time scaleIn the time scale combination, the firstThe first student categoryThe maximum DTW distance between the ICA components of the individual students at two time scales;an exponential function based on a natural constant; the time scale combination is obtained by taking any two time scales as a combination;
and obtaining the optimal degree of each time scale on the student class, and taking the time scale corresponding to the maximum value of the optimal degree as the optimal time scale of the student class.
Further, the specific acquisition method of the plurality of ICA components and the degree of check-in regularity of each student at each time scale is as follows:
for any student in any student category, respectively obtaining a check-in data sequence according to time sequence for check-in data of the student under each time scale, and performing ICA decomposition on the check-in data sequence to obtain a plurality of ICA components under each time scale;
For a check-in data sequence of the student under any time scale, dividing the check-in data sequence by the unit time length of the next adjacent time scale to obtain a plurality of sequence segments under the time scale, calculating pearson correlation coefficients for any two sequence segments, and recording the average value of absolute values of all pearson correlation coefficients as the check-in rule degree of the student in the time scale.
Further, the method for obtaining the preference degree and obtaining the optimal cycle length for each cycle length of each student comprises the following specific steps:
for any student, acquiring an optimal time scale of the class of the student to which the student belongs, and acquiring a plurality of period lengths of the student according to the unit time length of the optimal time scale; acquiring a check-in data sequence of the student in an optimal time scale, dividing the check-in data sequence into a plurality of sections according to the cycle length for any cycle length, and recording the sections as a plurality of data sections of the cycle length; acquiring a plurality of data segments of each cycle length of the student under an optimal time scale;
acquiring the preference degree of each period length according to the data segments with different period lengths and the environmental data with multiple dimensions under the optimal time scale; and taking the cycle length corresponding to the maximum value of the preference degree as the optimal cycle length of the student.
Further, the specific obtaining method of the preference degree of each cycle length comprises the following steps:
acquiring the environmental data corresponding to the check-in data of each student in each dimension and the association coefficient of each dimension according to the check-in data of each student in the optimal time scale of the class of the student and the environmental data of a plurality of dimensions;
for any student, calculating the range of the signed data in each data segment for a plurality of data segments of the student under any period length of the optimal time scale, and recording the range of each data segment; acquiring an environmental data segment in each dimension for each data segment, and calculating a pearson correlation coefficient for any two environmental data segments in the same dimension; for the firstData segment, except forOutside the data segmentData segment numberReference coefficients of individual data segmentsThe calculation method of (1) is as follows:
;
wherein,representing the number of dimensions of the environmental data,representing the first studentThe correlation coefficient of the individual dimensions,representation divide byOutside the data segmentThe variation coefficients of the pearson correlation coefficients of any two environmental data segments in the plurality of environmental data segments corresponding to the data segments and all the previous data segments; Representation divide byOutside the data segmentThe maximum value of the extremely bad data segment in the data segments and all the previous data segments;is a super parameter;
obtain and divide the firstEach data segment except the data segments and the first data segmentReference coefficients of the data segments, the minimum value of all the reference coefficients is taken as the firstThe regular coefficients of the individual data segments; and acquiring the regular coefficient of each data segment of the student under the period length, and taking the average value of all the regular coefficients as the preference degree of the period length.
Furthermore, the environment data corresponding to the check-in data of each student in each dimension and the association coefficient of each dimension are calculated according to the following specific calculation method:
for any student, for the environmental data of any dimension of the student, acquiring the environmental number corresponding to each check-in data of the student under the optimal time scaleAccording to the above; the first studentCorrelation of environment data and check-in data of individual dimensionsThe calculation method of (1) is as follows:
;
wherein,representing the number of check-in data for the student at the optimal time scale,representing the first studentThe slope of each check-in data and the immediately preceding check-in data,representing the first studentThe check-in data is at the firstThe environment data corresponding to each dimension and the adjacent previous check-in data are in the first place The slope of the environmental data corresponding to the individual dimensions,the representation is to take the absolute value,an exponential function based on a natural constant;
and acquiring the relevance of the environmental data and the check-in data of each dimension of the student, and carrying out softmax normalization on the relevance, wherein the obtained result is used as the relevance coefficient of each dimension.
Further, the method for obtaining the self-adaptive autocorrelation coefficient of each student and constructing the prediction model according to the optimal period length comprises the following specific steps:
for any student, adaptive autocorrelation coefficients for predictive coding compression of check-in data for that studentThe calculation method of (1) is as follows:
;
wherein,indicating the optimal cycle length for the student,indicating the total length of time the student has collected check-in data,super-parameters which are autocorrelation coefficients; and constructing an ARIMA prediction model for all check-in data of the student according to the autocorrelation coefficients of the student.
In a second aspect, another embodiment of the present invention provides an integrated management system for attendance of students in colleges and universities, the system comprising:
the student check-in data acquisition module is used for acquiring a plurality of check-in data of a plurality of students in different time scales and environment data in a plurality of dimensions;
The check-in data processing analysis module is used for acquiring similar distances among students according to check-in data of different students and clustering to obtain a plurality of student categories; acquiring the optimal time scale of each student category according to the check-in data of the same student category and different time scales;
acquiring the optimization degree of each cycle length of each student and acquiring the optimal cycle length according to the data segments obtained by the check-in data under different cycle lengths and the environmental data of multiple dimensions under the optimal time scale of the class of the student to which each student belongs;
and the check-in data compression management module is used for acquiring the self-adaptive self-correlation coefficient of each student according to the optimal cycle length, constructing a prediction model and carrying out prediction coding compression on the check-in data of each student.
The beneficial effects of the invention are as follows: according to the invention, the self-adaptive autocorrelation coefficient is obtained through the check-in data of the students in colleges and universities, an accurate ARIMA prediction model is constructed, and prediction coding compression is carried out; the method comprises the steps of clustering all check-in data of students, and obtaining optimal time scales of each student category according to optimal degrees of different time scales; quantifying the similar distance between students through the student check-in data to obtain student categories, and obtaining an optimal time scale according to the similar change characteristics between the student check-in data in the same student category and the capability of substituting the characterization by different time scales, and obtaining the subsequent optimal cycle length through the optimal time scale with stronger regularity; the method comprises the steps of obtaining an optimal period length by analyzing environmental data changes of different dimensions in an optimal time scale and combining the changes of original sign-in data, and finally obtaining an autocorrelation coefficient; the method and the device have the advantages that the defect that the traditional ARIMA prediction model is irregular in data storage deviation due to unreasonable setting of autocorrelation coefficients in the encoding compression process and the prediction effect in the data compression process is not ideal is avoided, so that the compression efficiency is low is overcome, the compression effect is improved, and compression management of sign-in data of students in colleges and universities is realized.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions of the prior art, the drawings which are used in the description of the embodiments or the prior art will be briefly described, it being obvious that the drawings in the description below are only some embodiments of the invention, and that other drawings can be obtained according to these drawings without inventive faculty for a person skilled in the art.
Fig. 1 is a schematic flow chart of a general management method for attendance of students in colleges and universities according to an embodiment of the present invention;
fig. 2 is a block diagram of a general management system for attendance of students in colleges and universities according to another embodiment of the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Referring to fig. 1, a flowchart of a method for comprehensively managing attendance of students in colleges and universities according to an embodiment of the present invention is shown, and the method includes the following steps:
step S001, collecting a plurality of check-in data of a plurality of students in different time scales and environment data in a plurality of dimensions.
The purpose of this embodiment is to comprehensively manage the check-in data of the students in colleges and universities, so that the check-in data of the students need to be collected, in this embodiment, a plurality of time scales are set to analyze the check-in data, the smallest time scale is one day, for any one student, the check-in times of each minute in one day of the student are counted, and as the check-in data of each minute, the check-in data of each minute is obtained for the student on the day scale; for other time scales, counting the number of check-in times per hour by a week scale as check-in data, counting the number of check-in times per day by a month scale as check-in data, counting the number of check-in times per week by a quarter scale as check-in data, acquiring a plurality of check-in data of different time scales for the student based on the number of check-in times per minute, and recording the place of each check-in while counting the number of check-in times; several check-in data of different time scales are acquired for all students of the college.
Further, environmental data of multiple dimensions are obtained simultaneously, in this embodiment, curriculum dimensions and weather dimensions are obtained to be used as environmental data for analysis, for the curriculum dimensions, day is used as sampling frequency, and the number of curriculum of each student per day is counted to be used as the environmental data of each student in the curriculum dimensions; for the weather dimension, the weather is also taken as the sampling frequency, sunny days and cloudiness are taken as the same weather data and recorded as 1, the weather suitable for going out is ensured, other weather is taken as the second weather data and recorded as 0, the weather unsuitable for going out is represented, the weather data of each day is obtained, and the environmental data of the weather dimension is obtained.
Thus, a plurality of check-in data of different time scales of each student of the colleges and universities and environmental data of the curriculum dimension and environmental data of the weather dimension are collected.
Step S002, obtaining similar distances among students according to the check-in data of different students, and clustering to obtain a plurality of student categories; and acquiring the optimal time scale of each student class according to the check-in data of the same student class and different time scales.
It should be noted that, in the process of performing predictive coding compression on the check-in data of the student, the autocorrelation coefficient has a dependency relationship with the check-in data to be compressed, the periodicity of the change of the check-in data needs to be obtained, the change of the check-in data in one period is stable, and the larger the corresponding autocorrelation coefficient is, so that the accurate periodicity characteristic of the check-in data of the student needs to be obtained by analysis.
It should be further noted that, different students have different outgoing habits, and the students need to be classified according to the check-in data of the students, the students with relatively similar outgoing habits are classified into one student category, and the similar distance between the students is quantified through the check-in time and place, so that the clustering acquisition of the student categories is completed; after the student category is obtained, the capability of replacing the characterization of each other in different time scales is quantified according to sign-in data in different time scales in the student category, so that the optimal time scale is obtained, and a basis is provided for subsequent periodic analysis.
In particular, the method comprises the steps of,for any student and any place where the student checked in, a plurality of check-ins of the student at the place are obtained, the corresponding time of each check-in at different time scales is obtained, for any time scale, the ratio of the check-in times of the student at the place to the unit time length of the time scale is calculated and is recorded as the check-in frequency of the student at the place under the time scale, the unit time length is a sampling time interval, the day scale is a minute, the week scale is an hour, the month scale is a day, and the quarter scale is a week, and in the calculation process, the hour (60 minutes), the day (1440 minutes) and the week (10080 minutes) are converted into a minute unit to be calculated so as to unify the dimension; calculating time intervals for two adjacent check-in of the place under the time scale to obtain a plurality of time intervals, obtaining the ratio of each time interval to the unit time length of the time scale, marking the ratio as a plurality of check-in intervals of the student at the place under the time scale, calculating variances for all the check-in intervals, and marking the variances as check-in interval factors of the student at the place under the time scale; obtaining the check-in frequency and the check-in interval factor of each student at each checked-in place under each time scale according to the method Students and the firstSimilar distance of individual studentsThe calculation method of (1) is as follows:
;
wherein,represent the firstStudents and the firstThe number of places in which each student checked in, it should be noted that the places in which both students checked in are counted only once, that is, the same place appears only once in the number statistics;representing the number of time scales, the present embodiment employsDescription is made;represent the firstUnder the individual time scalesThe first student isThe frequency of check-ins for the individual sites,represent the firstUnder the individual time scalesThe first student isCheck-in frequency of the individual places;represent the firstUnder the individual time scalesThe first student isA check-in interval factor for the individual sites;represent the firstUnder the individual time scalesThe first student isA check-in interval factor for the individual sites; it should be noted that, in the calculation process, if a student does not check in at a certain place, the corresponding check-in frequency and check-in interval factor are set to 0 for calculation; the similarity distance is quantified through the difference of the sign-in frequencies of different students at the same place and the difference of the sign-in interval factors under the same time scale, the smaller the difference is, the smaller the similarity distance is, the larger the similarity is, and the average value is obtained for the different time scales and the different places to obtain the similarity distance between the two students finally; according to the method, similar distances are obtained for any two students according to the check-in data.
Further, DBSCAN clustering is carried out on all students, the distance measurement adopts the similar distance among the students, a plurality of clusters are obtained through clustering, each cluster is marked as one student category, and then a plurality of student categories are obtained.
Further, for any student in any student category, the check-in data of the student under each time scale is respectively obtained according to the time sequence, and the check-in data sequence is subjected to ICA decomposition to obtain a plurality of ICA components under each time scale, wherein the ICA decomposition is a known technology, and the embodiment is not repeated; meanwhile, for a check-in data sequence of the student under any time scale, dividing the check-in data sequence by the unit time length of the next adjacent time scale to obtain a plurality of sequence segments under the time scale, calculating pearson correlation coefficients for any two sequence segments, and recording the average value of absolute values of all pearson correlation coefficients as the check-in rule degree of the student in the time scale; it should be noted that, the next adjacent time scale of the day scale is a week scale, and the corresponding unit time length is hour, then the sign-in data sequence of the day scale is segmented in each hour, that is, a sequence segment is obtained every 60 minutes; the week scale is segmented by day, and a sequence segment is obtained every 24 hours; segmenting the month scale in weeks, and obtaining a sequence segment every 7 days; while the quarter scale has no next time scale, so the segmentation is performed at 4 weeks, i.e. a sequence segment is obtained every 4 weeks; meanwhile, if the quantity of the check-in data contained in the last sequence segment is insufficient, discarding the check-in data, and not participating in calculation of the check-in rule degree; taking any two time scales as a time scale combination, in the embodiment, as 4 time scales are combined, 6 time scale combinations are obtained; according to the method, according to the check-in data of each student in each student category, a plurality of ICA components and check-in rule degree of each student in each time scale are obtained.
Further, for the firstClass of students, the firstOptimum degree of each time scale to the student classThe calculation method of (1) is as follows:
;
wherein,represent the firstThe number of students in a single student class,represent the firstThe first student categoryThe first student isThe degree of regularity of check-in on a time scale,the representation includes the firstThe number of time scale combinations of the individual time scales,the representation includes the firstThe first time scaleThe absolute value of the difference between the two time scale lengths in the individual time scale combinations, for example, the absolute value of the difference between the day scale and the week scale is 6 days, namely, the difference between 1 day and 7 days, the month scale is calculated according to 30 days, and the quarter scale is calculated according to 90 days;represents the maximum of the time scale length, i.e., 90 days in the quarter scale;the representation includes the firstThe first time scaleIn the time scale combination, the firstThe first student categoryThe maximum value of the DTW distance between ICA components of the students at two time scales is needed to be explained, namely, only the DTW distance between ICA components of different time scales is calculated;representing an exponential function based on natural constants, the present embodiment employsModel to present inverse proportional relationship and normalization process, whereinFor model input, the implementer can set an inverse proportion function and a normalization function according to actual conditions.
At this time, the smaller the difference of the two time scale lengths under the time scale combination is, the smaller the maximum value of the DTW distance between the ICA components of the two different time scales is, the smaller the difference of the time scale lengths is, the stronger the characterization capability of mutual substitution is, and the smaller the DTW distance of the ICA components is, the greater the overall similarity of the check-in data sequences of the different time scales is; the similarity of all time scales obtained by combining the time scales is quantified and averaged, and the bigger the degree of the check-in rule is, the more obvious the periodicity is, the easier the optimal period length is found in the follow-up, the bigger the similarity is, the stronger the capability of representing other time scales is, and the more obvious the periodicity is, and the bigger the optimal degree of the corresponding time scales is; and calculating the average value of all students in the student category to obtain the final optimal degree.
Further, according to the method, the optimal degree of each time scale to the student category is obtained, and the time scale corresponding to the maximum value of the optimal degree is used as the optimal time scale of the student category; and obtaining the optimal time scale of each student class according to the method.
So far, students are divided into a plurality of student categories, and an optimal time scale is obtained for each student category.
Step S003, according to the data segment obtained by the check-in data under different cycle lengths and the environmental data of multiple dimensions under the optimal time scale of the class of each student, obtaining the optimal degree of each cycle length of each student and obtaining the optimal cycle length.
After the optimal time scale is obtained, analyzing the optimal period length of the student under the corresponding optimal time scale, and obtaining the optimal period length by iterating the period length and obtaining the optimal degree of the period length; in the process of calculating the preference degree, the similarity among the data segments divided by the period length needs to be quantized, the preference degree is obtained through the similarity, the greater the similarity is, the greater the preference degree is, and the optimal period length is obtained; in the similarity quantification process, the correlation influence of the environmental data of other dimensions on the check-in data is required to be combined, the similarity of other data segments and the data segments is analyzed according to the correlation influence, and finally the acquisition of the preference degree is completed.
Specifically, for any student, acquiring an optimal time scale of a student class to which the student belongs, taking a unit time length of the optimal time scale as an initial period length, taking a step length as the unit time length, setting a maximum period length as half of the total time length of check-in data acquired by the student, and obtaining a plurality of period lengths of the student through iteration; in step S002, the check-in data sequence of the student at each time scale is obtained, the check-in data sequence of the student at the optimal time scale is obtained, and by taking any period length as an example, the check-in data sequence is divided into a plurality of segments by the period length and is recorded as a plurality of data segments of the period length, and it is to be noted that if the number of check-in data in the last data segment does not meet the period length, discarding is performed; and acquiring a plurality of data segments of each cycle length of the student under the optimal time scale according to the method.
Further, for the environmental data of any dimension of the student, the optimal time of the student is obtainedThe unit of the environmental data is the day, the unit time length of the optimal time scale is different, the unit time length of the day scale and the week scale is the minute and the hour, and the environmental data corresponding to the day of the minute or the hour is taken as the corresponding environmental data; the unit time length of the quarter scale is week, more than day, and the average value of the environmental data corresponding to all days in the week is taken as the environmental data corresponding to the week, then the student is the firstCorrelation of environment data and check-in data of individual dimensionsThe calculation method of (1) is as follows:
;
wherein,representing the number of check-in data for the student at the optimal time scale,representing the first studentThe slope of each sign-in data and the adjacent previous sign-in data, namely, the abscissa is time, and the ordinate is the slope obtained by the sign-in times;representing the first studentThe check-in data is at the firstThe environment data corresponding to each dimension and the adjacent previous check-in data are in the first placeSlope of the environmental data corresponding to each dimension, namely, the abscissa is time, and the ordinate is slope obtained by two environmental data; The representation is to take the absolute value,representing an exponential function based on natural constants, the present embodiment employsModel to present inverse proportional relationship and normalization process, whereinFor inputting the model, an implementer can set an inverse proportion function and a normalization function according to actual conditions; the slope between adjacent check-in data and the slope between corresponding environmental data, the closer the ratio is to 1, the greater the correlation between the environmental data and the check-in data of the dimension; according to the method, the relevance of the environmental data and the check-in data of each dimension of the student is obtained, the relevance is subjected to softmax normalization, and the obtained result is used as the relevance coefficient of each dimension.
Further, for a plurality of data segments of the student under any period length of the optimal time scale, calculating the extremely bad sign-on data in each data segment, namely, the difference value between the maximum value and the minimum value of sign-on data in the data segment is recorded as the extremely bad of each data segment; acquiring an environmental data segment in each dimension for each data segment, calculating pearson correlation coefficients for any two environmental data segments in the same dimension, namely, corresponding environmental data of the sign-in data in each dimension, and acquiring corresponding environmental data segments according to the data segments of the sign-in data; then in the first place By way of example, except for the first data segmentOutside the data segmentData segment numberReference coefficients of individual data segmentsThe calculation method of (1) is as follows:
;
wherein,representing the number of dimensions of the environmental data,representing the first studentThe correlation coefficient of the individual dimensions,representation divide byOutside the data segmentThe coefficient of variation of the pearson correlation coefficient of any two environmental data segments among the plurality of environmental data segments corresponding to the data segments and all the previous data segments is calculated as a known technology, and the embodiment is not repeated;representation divide byOutside the data segmentThe data segment is extremely bad in the data segments and all the previous data segmentsIs the maximum value of (2);is super-parametric, the embodiment adoptsDescription is made; by correlating coefficients as weights, a larger coefficient of variation indicates a larger periodicity, while the allowable fluctuation range is quantified by a very bad.
Further, the dividing step is obtained according to the methodEach data segment except the data segments and the first data segmentReference coefficients of the data segments, the minimum value of all the reference coefficients is taken as the firstThe regular coefficients of the individual data segments; acquiring the rule coefficient of each data segment of the student under the period length according to the method, and taking the average value of all the rule coefficients as the preference degree of the period length; according to the optimization degree of each cycle length of the student under the optimal time scale, taking the cycle length corresponding to the maximum value of the optimization degree as the optimal cycle length of the student; and obtaining the optimal cycle length of each student according to the method.
So far, through the quantification of the relevance of the environmental data of other dimensions and the check-in data of students under the optimal time scale, the optimal degree is obtained for different period lengths under the optimal time scale, and then the optimal period length of each student is obtained.
Step S004, obtaining the self-adaptive self-correlation coefficient of each student according to the optimal period length, constructing a prediction model, and performing prediction coding compression on check-in data of each student.
After the optimal period length of each student is obtained, for any student, the adaptive autocorrelation system for performing predictive coding compression on check-in data of the studentNumber of digitsThe calculation method of (1) is as follows:
;
wherein,indicating the optimal cycle length for the student,representing the total length of time of check-in data (check-in data at all time scales) collected by the student,as the super parameter of the autocorrelation coefficient, the present embodiment adoptsDescription is made; the adaptive autocorrelation coefficients are obtained based on the optimal period length and the total time length at the maximum time scale.
Further, according to the autocorrelation coefficient of the student, an ARIMA prediction model is constructed for all check-in data of the student, the differential order of this embodiment is set to 3, so that a data error of each check-in data can be obtained, prediction coding compression is implemented according to the data error and the check-in data, compression management is implemented on the outgoing check-in data of the student through the prediction coding compression, wherein the ARIMA prediction model construction and the prediction coding are both known techniques, and this embodiment is not repeated; according to the method, the self-adaptive self-correlation coefficient is obtained for each student, a prediction model is built, prediction coding compression is carried out, and comprehensive management of the outgoing check-in data of the students is completed.
Therefore, the self-adaptive predictive coding compression is carried out on the outgoing check-in data of the students in colleges and universities, so that the management efficiency of the outgoing check-in data of the students is improved, and the comprehensive management of the outgoing check-in data of the students is realized.
Referring to fig. 2, a block diagram of an integrated management system for attendance of students in colleges and universities according to another embodiment of the present invention is shown, where the system includes:
the student check-in data acquisition module 101 is used for acquiring a plurality of check-in data of a plurality of students in different time scales and environment data in a plurality of dimensions.
Check-in data processing analysis module 102:
(1) According to the check-in data of different students, obtaining similar distances among the students, and clustering to obtain a plurality of student categories; acquiring the optimal time scale of each student category according to the check-in data of the same student category and different time scales;
(2) And acquiring the optimization degree of each cycle length of each student and acquiring the optimal cycle length according to the data segments obtained by the check-in data under different cycle lengths and the environmental data of multiple dimensions under the optimal time scale of the class of the student to which each student belongs.
The check-in data compression management module 103 is configured to obtain an adaptive autocorrelation coefficient of each student according to an optimal cycle length, construct a prediction model, and perform predictive coding compression on check-in data of each student.
The above description is only of the preferred embodiments of the present invention and is not intended to limit the invention, but any modifications, equivalent substitutions, improvements, etc. within the principles of the present invention should be included in the scope of the present invention.
Claims (4)
1. The comprehensive management method for the outgoing check-in of the college students is characterized by comprising the following steps of:
collecting a plurality of check-in data of a plurality of students going out in different time scales and environment data in a plurality of dimensions; the check-in data comprise check-in times counted in each unit time length under different time scales, and the place of each check-in is recorded;
according to the check-in data of different students, obtaining similar distances among the students, and clustering to obtain a plurality of student categories; acquiring the optimal time scale of each student category according to the check-in data of the same student category and different time scales;
acquiring the optimization degree of each cycle length of each student and acquiring the optimal cycle length according to the data segments obtained by the check-in data under different cycle lengths and the environmental data of multiple dimensions under the optimal time scale of the class of the student to which each student belongs;
acquiring self-adaptive self-correlation coefficients of each student according to the optimal cycle length, constructing a prediction model, and performing prediction coding compression on check-in data of each student;
The optimal time scale of each student category is obtained by the following specific method:
according to the check-in data of each student in different time scales, a plurality of ICA components and check-in rule degrees of each student in each time scale are obtained; for the firstCategory of individual students, th->Optimum degree of each time scale to the student classThe calculation method of (1) is as follows:
wherein,indicate->Number of students in each student category,/->Indicate->First->The student is at->Degree of check-in regularity on a time scale, +.>The representation includes->The number of time scale combinations of the individual time scales,the representation includes->First of the time scale>The absolute value of the difference between the two time scale lengths in the time scale combinations,represents the maximum value of the time scale length, +.>The representation includes->First of the time scale>In the individual time scale combinations +.>First->The maximum DTW distance between the ICA components of the individual students at two time scales; />An exponential function based on a natural constant; the time scale combination is obtained by taking any two time scales as a combination;
obtaining the optimal degree of each time scale on the student class, and taking the time scale corresponding to the maximum value of the optimal degree as the optimal time scale of the student class;
The specific acquisition method comprises the following steps of:
for any student in any student category, respectively obtaining a check-in data sequence according to time sequence for check-in data of the student under each time scale, and performing ICA decomposition on the check-in data sequence to obtain a plurality of ICA components under each time scale;
dividing the check-in data sequence of the student under any time scale by using the unit time length of the next adjacent time scale to obtain a plurality of sequence segments under the time scale, calculating pearson correlation coefficients for any two sequence segments, and recording the average value of absolute values of all pearson correlation coefficients as the check-in rule degree of the student in the time scale;
the method for obtaining the optimal degree of preference and obtaining the optimal cycle length for each cycle length of each student comprises the following specific steps:
for any student, acquiring an optimal time scale of the class of the student to which the student belongs, and acquiring a plurality of period lengths of the student according to the unit time length of the optimal time scale; acquiring a check-in data sequence of the student in an optimal time scale, dividing the check-in data sequence into a plurality of sections according to the cycle length for any cycle length, and recording the sections as a plurality of data sections of the cycle length; acquiring a plurality of data segments of each cycle length of the student under an optimal time scale;
Acquiring the preference degree of each period length according to the data segments with different period lengths and the environmental data with multiple dimensions under the optimal time scale; taking the cycle length corresponding to the maximum value of the preference degree as the optimal cycle length of the student;
the optimization degree of each period length is obtained by the following specific method:
acquiring the environmental data corresponding to the check-in data of each student in each dimension and the association coefficient of each dimension according to the check-in data of each student in the optimal time scale of the class of the student and the environmental data of a plurality of dimensions;
for any student, calculating the range of the signed data in each data segment for a plurality of data segments of the student under any period length of the optimal time scale, and recording the range of each data segment; acquiring an environmental data segment in each dimension for each data segment, and calculating a pearson correlation coefficient for any two environmental data segments in the same dimension; for the firstData segments except->Out of the data section->Data segment and->Reference coefficient of individual data segments->The calculation method of (1) is as follows:
wherein,number of dimensions representing environmental data, +.>Indicating the student's- >Correlation coefficient of individual dimensions->Indicate except->Out of the data section->The variation coefficients of the pearson correlation coefficients of any two environmental data segments in the plurality of environmental data segments corresponding to the data segments and all the previous data segments; />Indicate except->Out of the data section->The maximum value of the extremely bad data segment in the data segments and all the previous data segments; />Is a super parameter;
obtain and divide the firstEach data segment except the data segments is +.>Reference system for individual data segmentsNumber, the minimum value of all reference coefficients is taken as the +.>The regular coefficients of the individual data segments; acquiring a regular coefficient of each data segment of the student under the period length, and taking the average value of all the regular coefficients as the preference degree of the period length;
the specific calculation method includes the steps of:
for any student, for the environmental data of any dimension of the student, acquiring the environmental data corresponding to each check-in data of the student under the optimal time scale; the first studentCorrelation of environment data and check-in data of individual dimensionsThe calculation method of (1) is as follows:
Wherein,representing the number of check-in data of the student at an optimal time scale,/for example>Indicating the student's->Slope of individual check-in data and immediately preceding check-in data, +.>Indicating the student's->The personal check-in data is at->Environmental data corresponding to each dimension and the adjacent previous check-in data are at +.>Slope of the environment data corresponding to the respective dimension, +.>Representing absolute value>An exponential function based on a natural constant;
acquiring the relevance of the environmental data and the check-in data of each dimension of the student, and carrying out softmax normalization on the relevance, wherein the obtained result is used as a relevance coefficient of each dimension;
the method for obtaining the self-adaptive self-correlation coefficient of each student and constructing the prediction model according to the optimal cycle length comprises the following specific steps:
for any student, adaptive autocorrelation coefficients for predictive coding compression of check-in data for that studentThe calculation method of (1) is as follows:
wherein,indicating the optimal cycle length of the student, +.>Representing the total time of check-in data acquired by the studentLength of the room (I)>Super-parameters which are autocorrelation coefficients; and constructing an ARIMA prediction model for all check-in data of the student according to the autocorrelation coefficients of the student.
2. The method for comprehensively managing outgoing check-in of college students according to claim 1, wherein the similar distance between the students is obtained by the following steps:
according to the check-in data and the check-in places of each student in different time scales, the check-in frequency and the check-in interval factor of each student in each check-in place of each student in each time scale are obtained; first, theStudent and->Similar distance of individual studentsThe calculation method of (1) is as follows:
wherein,indicate->Student and->The number of places to which the individual students have checked in; />Representing the time scaleNumber of pieces;indicate->The>The student is at->Check-in frequency of individual places, < >>Indicate->The>The student is at->Check-in frequency of the individual places; />Indicate->The>The student is at->A check-in interval factor for the individual sites; />Indicate->The>The student is at->Check-in interval factor for individual places.
3. The method for comprehensively managing the attendance of students in colleges and universities according to claim 2, wherein the attendance frequency and the attendance interval factor of each student at each place where the students check in on each time scale are obtained by the following specific methods:
For any place where any student and any student check-in, acquiring a plurality of check-ins of the student at the place, acquiring corresponding moments of each check-in at different time scales, and for any time scale, calculating the ratio of the check-in times of the student at the place to the unit time length of the time scale, and recording the ratio as the check-in frequency of the student at the place under the time scale;
calculating time intervals for two adjacent check-ins of the place under the time scale to obtain a plurality of time intervals, obtaining the ratio of each time interval to the unit time length of the time scale, marking the ratio as a plurality of check-in intervals of the student at the place under the time scale, calculating variances for all the check-in intervals, and marking the variances as check-in interval factors of the student at the place under the time scale.
4. The utility model provides a college student goes out integrated management system of signing in which, this system includes:
the student check-in data acquisition module is used for acquiring a plurality of check-in data of a plurality of students in different time scales and environment data in a plurality of dimensions;
the check-in data processing analysis module is used for acquiring similar distances among students according to check-in data of different students and clustering to obtain a plurality of student categories; acquiring the optimal time scale of each student category according to the check-in data of the same student category and different time scales;
Acquiring the optimization degree of each cycle length of each student and acquiring the optimal cycle length according to the data segments obtained by the check-in data under different cycle lengths and the environmental data of multiple dimensions under the optimal time scale of the class of the student to which each student belongs;
the check-in data compression management module is used for acquiring the self-adaptive self-correlation coefficient of each student according to the optimal cycle length, constructing a prediction model and carrying out prediction coding compression on the check-in data of each student;
the optimal time scale of each student category is obtained by the following specific method:
according to the check-in data of each student in different time scales, a plurality of ICA components and check-in rule degrees of each student in each time scale are obtained; for the firstCategory of individual students, th->Optimum degree of each time scale to the student classThe calculation method of (1) is as follows:
wherein,indicate->Number of students in each student category,/->Representation ofFirst->First->The student is at->Degree of check-in regularity on a time scale, +.>The representation includes->The number of time scale combinations of the individual time scales,the representation includes->First of the time scale>The absolute value of the difference between the two time scale lengths in the time scale combinations, Represents the maximum value of the time scale length, +.>The representation includes->First of the time scale>In the individual time scale combinations +.>First->The maximum DTW distance between the ICA components of the individual students at two time scales; />An exponential function based on a natural constant; the time scale combination is obtained by taking any two time scales as a combination;
obtaining the optimal degree of each time scale on the student class, and taking the time scale corresponding to the maximum value of the optimal degree as the optimal time scale of the student class;
the specific acquisition method comprises the following steps of:
for any student in any student category, respectively obtaining a check-in data sequence according to time sequence for check-in data of the student under each time scale, and performing ICA decomposition on the check-in data sequence to obtain a plurality of ICA components under each time scale;
dividing the check-in data sequence of the student under any time scale by using the unit time length of the next adjacent time scale to obtain a plurality of sequence segments under the time scale, calculating pearson correlation coefficients for any two sequence segments, and recording the average value of absolute values of all pearson correlation coefficients as the check-in rule degree of the student in the time scale;
The method for obtaining the optimal degree of preference and obtaining the optimal cycle length for each cycle length of each student comprises the following specific steps:
for any student, acquiring an optimal time scale of the class of the student to which the student belongs, and acquiring a plurality of period lengths of the student according to the unit time length of the optimal time scale; acquiring a check-in data sequence of the student in an optimal time scale, dividing the check-in data sequence into a plurality of sections according to the cycle length for any cycle length, and recording the sections as a plurality of data sections of the cycle length; acquiring a plurality of data segments of each cycle length of the student under an optimal time scale;
acquiring the preference degree of each period length according to the data segments with different period lengths and the environmental data with multiple dimensions under the optimal time scale; taking the cycle length corresponding to the maximum value of the preference degree as the optimal cycle length of the student;
the optimization degree of each period length is obtained by the following specific method:
acquiring the environmental data corresponding to the check-in data of each student in each dimension and the association coefficient of each dimension according to the check-in data of each student in the optimal time scale of the class of the student and the environmental data of a plurality of dimensions;
For any student, calculating the range of the signed data in each data segment for a plurality of data segments of the student under any period length of the optimal time scale, and recording the range of each data segment; acquiring an environmental data segment in each dimension for each data segment, and calculating a pearson correlation coefficient for any two environmental data segments in the same dimension; for the firstData segments except->Out of the data section->Data segment and->Reference coefficient of individual data segments->The calculation method of (1) is as follows:
wherein,number of dimensions representing environmental data, +.>Indicating the student's->Correlation coefficient of individual dimensions->Indicate except->Out of the data section->The variation coefficients of the pearson correlation coefficients of any two environmental data segments in the plurality of environmental data segments corresponding to the data segments and all the previous data segments; />Indicate except->Out of the data section->The maximum value of the extremely bad data segment in the data segments and all the previous data segments; />Is a super parameter;
obtain and divide the firstEach data segment except the data segments is +.>Reference coefficients of the data segments, the minimum value of all reference coefficients is taken as the +.>The regular coefficients of the individual data segments; acquiring a regular coefficient of each data segment of the student under the period length, and taking the average value of all the regular coefficients as the preference degree of the period length;
The specific calculation method includes the steps of:
for any student, for the environmental data of any dimension of the student, acquiring the environmental data corresponding to each check-in data of the student under the optimal time scale; the first studentCorrelation of environment data and check-in data of individual dimensionsThe calculation method of (1) is as follows:
wherein,representing the number of check-in data of the student at an optimal time scale,/for example>Indicating the student's->Slope of individual check-in data and immediately preceding check-in data, +.>Indicating the student's->The personal check-in data is at->Environmental data corresponding to each dimension and the adjacent previous check-in data are at +.>Slope of the environment data corresponding to the respective dimension, +.>Representing absolute value>An exponential function based on a natural constant;
acquiring the relevance of the environmental data and the check-in data of each dimension of the student, and carrying out softmax normalization on the relevance, wherein the obtained result is used as a relevance coefficient of each dimension;
the method for obtaining the self-adaptive self-correlation coefficient of each student and constructing the prediction model according to the optimal cycle length comprises the following specific steps:
For any student, adaptive autocorrelation coefficients for predictive coding compression of check-in data for that studentThe calculation method of (1) is as follows:
wherein,indicating the optimal cycle length of the student, +.>Representing the total length of time of check-in data acquired by the student, < >>Super-parameters which are autocorrelation coefficients; and constructing an ARIMA prediction model for all check-in data of the student according to the autocorrelation coefficients of the student.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311422091.3A CN117151948B (en) | 2023-10-31 | 2023-10-31 | College student outgoing check-in comprehensive management method and system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311422091.3A CN117151948B (en) | 2023-10-31 | 2023-10-31 | College student outgoing check-in comprehensive management method and system |
Publications (2)
Publication Number | Publication Date |
---|---|
CN117151948A CN117151948A (en) | 2023-12-01 |
CN117151948B true CN117151948B (en) | 2024-02-27 |
Family
ID=88899157
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202311422091.3A Active CN117151948B (en) | 2023-10-31 | 2023-10-31 | College student outgoing check-in comprehensive management method and system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN117151948B (en) |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN118133048B (en) * | 2024-05-07 | 2024-07-09 | 临沂大学 | College student physique test data acquisition method and system |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US10122379B1 (en) * | 2016-03-23 | 2018-11-06 | EMC IP Holding Company LLC | Content-aware compression of data with reduced number of class codes to be encoded |
CN115733498A (en) * | 2023-01-10 | 2023-03-03 | 北京四维纵横数据技术有限公司 | Compression method and device of time sequence data, computer equipment and medium |
CN116383450A (en) * | 2023-06-05 | 2023-07-04 | 沧州中铁装备制造材料有限公司 | Railway and highway logistics transportation information comprehensive management system |
CN116664014A (en) * | 2023-07-25 | 2023-08-29 | 临沂大学 | Comprehensive evaluation system and method for college student management |
Family Cites Families (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8090709B2 (en) * | 2007-06-28 | 2012-01-03 | Microsoft Corporation | Representing queries and determining similarity based on an ARIMA model |
US9858311B2 (en) * | 2014-03-31 | 2018-01-02 | EMC IP Holding Company LLC | Heteroscedastic data compression using ARIMA-GARCH model estimation |
-
2023
- 2023-10-31 CN CN202311422091.3A patent/CN117151948B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US10122379B1 (en) * | 2016-03-23 | 2018-11-06 | EMC IP Holding Company LLC | Content-aware compression of data with reduced number of class codes to be encoded |
CN115733498A (en) * | 2023-01-10 | 2023-03-03 | 北京四维纵横数据技术有限公司 | Compression method and device of time sequence data, computer equipment and medium |
CN116383450A (en) * | 2023-06-05 | 2023-07-04 | 沧州中铁装备制造材料有限公司 | Railway and highway logistics transportation information comprehensive management system |
CN116664014A (en) * | 2023-07-25 | 2023-08-29 | 临沂大学 | Comprehensive evaluation system and method for college student management |
Non-Patent Citations (2)
Title |
---|
尚涛.《SAS数据挖掘与分析项目实战》.中国铁道出版社,2020,131. * |
时间序列三维荧光光谱数据的压缩预测;于绍慧;肖雪;丁虹;徐格;;计量学报;20171122(06);119-122 * |
Also Published As
Publication number | Publication date |
---|---|
CN117151948A (en) | 2023-12-01 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Arguez et al. | The definition of the standard WMO climate normal: The key to deriving alternative climate normals | |
Wang et al. | Inter-comparison of extra-tropical cyclone activity in nine reanalysis datasets | |
CN117151948B (en) | College student outgoing check-in comprehensive management method and system | |
CN111340288B (en) | Urban air quality time sequence prediction method considering time-space correlation | |
CN112101480A (en) | Multivariate clustering and fused time sequence combined prediction method | |
CN114168906B (en) | Mapping geographic information data acquisition system based on cloud computing | |
CN111897859B (en) | Big data intelligent report platform for enterprise online education | |
CN110889092A (en) | Short-time large-scale activity peripheral track station passenger flow volume prediction method based on track transaction data | |
CN117454255B (en) | Intelligent building energy consumption data optimized storage method | |
CN117290364A (en) | Intelligent market investigation data storage method | |
CN115907822A (en) | Load characteristic index relevance mining method considering region and economic influence | |
CN117132135A (en) | Urban informatization management system and method based on digital twinning | |
CN111931992A (en) | Power load prediction index selection method and device | |
CN105488598A (en) | Medium-and-long time electric power load prediction method based on fuzzy clustering | |
CN108595814B (en) | Method for simulating daily rainfall by using measured perennial rainfall data | |
CN114970956A (en) | OLS and QOIC-based population association analysis and prediction method | |
CN108830405B (en) | Real-time power load prediction system and method based on multi-index dynamic matching | |
CN113408964A (en) | Method for assisting government carbon management based on comprehensive evaluation system | |
CN115511230B (en) | Electric energy substitution potential analysis and prediction method | |
CN116384949A (en) | Intelligent government affair information data management system based on digital management | |
CN105741143A (en) | Load characteristic and cluster analysis based electric power commodity pricing model establishment method | |
CN111797924B (en) | Three-dimensional garden portrait method and system based on clustering algorithm | |
CN114819260A (en) | Dynamic generation method of hydrologic time series prediction model | |
CN117236666B (en) | Emergency material demand analysis method and system | |
CN115813407B (en) | Sleep electroencephalogram staging method based on fuzzy step vector fluctuation |
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 |