CN112509705A - Data analysis method for campus epidemic prevention, control, monitoring and early warning - Google Patents
Data analysis method for campus epidemic prevention, control, monitoring and early warning Download PDFInfo
- Publication number
- CN112509705A CN112509705A CN202011269950.6A CN202011269950A CN112509705A CN 112509705 A CN112509705 A CN 112509705A CN 202011269950 A CN202011269950 A CN 202011269950A CN 112509705 A CN112509705 A CN 112509705A
- Authority
- CN
- China
- Prior art keywords
- data
- monitoring
- model
- early warning
- analysis method
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 62
- 238000012544 monitoring process Methods 0.000 title claims abstract description 58
- 230000002265 prevention Effects 0.000 title claims abstract description 33
- 238000007405 data analysis Methods 0.000 title claims abstract description 19
- 238000004458 analytical method Methods 0.000 claims abstract description 16
- 230000002159 abnormal effect Effects 0.000 claims description 28
- 230000036541 health Effects 0.000 claims description 18
- 230000036760 body temperature Effects 0.000 claims description 13
- 238000012549 training Methods 0.000 claims description 12
- 230000006870 function Effects 0.000 claims description 9
- 238000007476 Maximum Likelihood Methods 0.000 claims description 6
- 206010037660 Pyrexia Diseases 0.000 claims description 6
- 238000009529 body temperature measurement Methods 0.000 claims description 6
- 208000024891 symptom Diseases 0.000 claims description 6
- 208000011580 syndromic disease Diseases 0.000 claims description 6
- 206010011224 Cough Diseases 0.000 claims description 4
- 206010012735 Diarrhoea Diseases 0.000 claims description 3
- 208000035850 clinical syndrome Diseases 0.000 claims description 3
- 238000012937 correction Methods 0.000 claims description 3
- 238000009792 diffusion process Methods 0.000 claims description 3
- 238000002955 isolation Methods 0.000 claims description 3
- 229940036051 sojourn Drugs 0.000 claims description 3
- 238000000528 statistical test Methods 0.000 claims description 3
- 206010035664 Pneumonia Diseases 0.000 description 11
- 230000002776 aggregation Effects 0.000 description 7
- 238000004220 aggregation Methods 0.000 description 7
- 230000008859 change Effects 0.000 description 3
- 238000001514 detection method Methods 0.000 description 3
- 230000009471 action Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 230000007480 spreading Effects 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 201000010099 disease Diseases 0.000 description 1
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012731 temporal analysis Methods 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/80—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for detecting, monitoring or modelling epidemics or pandemics, e.g. flu
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N7/00—Computing arrangements based on specific mathematical models
- G06N7/01—Probabilistic graphical models, e.g. probabilistic networks
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Public Health (AREA)
- Health & Medical Sciences (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Medical Informatics (AREA)
- Theoretical Computer Science (AREA)
- Software Systems (AREA)
- Algebra (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Computational Mathematics (AREA)
- Evolutionary Computation (AREA)
- Mathematical Analysis (AREA)
- Probability & Statistics with Applications (AREA)
- Biomedical Technology (AREA)
- Databases & Information Systems (AREA)
- Artificial Intelligence (AREA)
- Pathology (AREA)
- Epidemiology (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Measuring And Recording Apparatus For Diagnosis (AREA)
Abstract
The invention belongs to the technical field of digital epidemic situation monitoring and early warning, is used for campus epidemic situation prevention, control, monitoring and early warning, and particularly relates to a data analysis method for campus epidemic situation prevention, control, monitoring and early warning. A data analysis method for campus epidemic prevention, control, monitoring and early warning utilizes a multi-source monitoring data analysis technology to carry out multi-dimension, multi-scale and multi-level analysis on obtained individual data and aggregated data in a network form, and estimates the probability of campus epidemic prevention, control and outbreak on the basis of fully considering the complex relation among variables.
Description
Technical Field
The invention belongs to the technical field of digital detection and early warning, and relates to a data analysis method for campus epidemic prevention, control, monitoring and early warning, in particular to a data analysis method for campus epidemic prevention, control, monitoring and early warning.
Background
The method for epidemic situation early warning in the prior art is mainly a pure time aggregation detection method and mainly aims to detect the time point when the number of disease cases in time sequence data is abnormally increased. On the basis of the prior art, the campus epidemic prevention and control space-time aggregation monitoring has the following defects: 1. the possible scale of outbreak of the new coronary pneumonia epidemic can not be predicted; 2. the size of the scanning window is fixed and invariable; 3. some factors cause the natural variation of the occurrence number of new coronary pneumonia epidemic in time and space: such as seasonality, these factors cannot be corrected; 4. the geographical distribution of students is uneven, such as: the number of students in the province is larger than that in the other provinces, and the uneven geographic distribution of the students cannot be corrected.
Disclosure of Invention
The invention mainly aims to provide a data analysis method for campus epidemic prevention, control, monitoring and early warning in order to solve the problem that multidimensional aggregative monitoring cannot be carried out on epidemic in the prior art.
In order to achieve the purpose, the invention specifically adopts the following technical scheme:
1. a data analysis method for campus epidemic prevention, control, monitoring and early warning is characterized by comprising the following steps: the method comprises the following steps:
s1, firstly, data are collected through a campus epidemic prevention and control platform mobile phone end 'health report' or a campus temperature sensing system or nearby epidemic and co-process query software;
s2: tracking and monitoring the abnormal data collected in the step S1;
s3: and performing space-time analysis and early warning statistics on the data detected in the step S2, wherein the method for the space-time analysis and the early warning statistics is a Bayesian space-time analysis method or a hidden Markov multi-source monitoring data method.
Further, in step S1, through the "health report" function of the mobile phone terminal of the campus epidemic prevention and control platform, the collected data includes two types, one type is health data, i.e. it indicates that the physical condition is good, and the other type is abnormal data of the body, including: the patient has symptoms of cough, weakness, fever, diarrhea and the like, and the fever symptom needs to be filled in the body temperature range; body temperature data directly acquired by a temperature sensing system, wherein one type of the body temperature data is data in a normal range; another is data at temperatures above 37.2 ℃; and comparing and analyzing the filtered data by using nearby epidemic situation and co-process query software.
Further, the method for monitoring in step S2 includes the following steps:
s2.1: before returning to school
The students and the teaching staff with abnormal body are tracked and monitored through the health reporting function;
tracking and monitoring students/teachers of epidemic high-risk regional cadavers, students/teachers of middle-risk regional cadavers, students with high-risk sojourn histories after a specific date and teaching workers by regional division and monitoring;
tracking and monitoring students and teaching workers who have suspected contact history in the same district, building and class by using integrated software of 'nearby epidemic situation' and 'same-course query';
s2.2: go back to school
Tracking and monitoring the persons who have suspected contact history of the same-class traffic workers through the integrated software of 'same-course query';
when the temperature is corrected, temperature measurement is carried out, and isolation and tracking monitoring are carried out on the abnormal temperature person;
isolating, tracking and monitoring students who enter the school according to batches and have not relieved abnormal early warning in 14 days;
s2.3: after the correction
The students and the teaching staff with abnormal body are tracked and monitored through the health reporting function;
tracking and monitoring students with abnormal body temperature measured by a daily infrared automatic temperature sensing system;
and tracking and monitoring students who have sufficient reason to leave the school after entering the school after returning.
Further, the bayesian spatio-temporal analysis method described in step S3 includes the steps of:
s3.1: collecting data by the method of S1;
s3.2: selecting indexes; performing prevention and control according to targets of different syndromes, sorting influence factors of various syndromes, considering the dependency relationship among variables, the availability and the quantifiability of related data and the like, and selecting a proper model index;
s3.3: and (3) network structure learning: determining a proper Bayesian network topological structure by utilizing historical data and combining with prior knowledge, selecting a statistical test-based method or a search score-based method according to actual conditions, and dividing model nodes established after the method is selected into the following four parts:
s3.3.1, global node G, namely the characteristics of the school whole population;
s3.3.2, interface nodes I comprise time and places of campus epidemic prevention and control diffusion;
s3.3.3. an individual network P, wherein each sub-network corresponds to an individual and comprises demographic information, clinical syndrome and the like, and the topological network structure of the individual network P is determined by expert judgment;
s3.3.4, a crowd evidence node O comprises abnormal health card data, abnormal temperature measurement of a temperature sensing system, parallel simultaneous query, same unit of nearby epidemic situation and a building;
s3.4: determining network parameters: under the condition of a given Bayesian network topological structure, determining the conditional probability density of each node, determining the parameter learning of the Bayesian network through expert knowledge and training sample learning, then performing parameter estimation by adopting a parameter estimation method, and representing one outbreak when the test probability exceeds a threshold value.
Preferably, the parameter estimation method is maximum likelihood estimation, maximum a posteriori probability or an expectation maximization algorithm.
S3.5: optimizing network structure and parameters; and measuring the false positive rate of the model by adopting a curve, calculating the difference value between the outbreak time of campus epidemic prevention and control and the early warning time of the model, and optimizing the model structure by evaluating the result.
Further, the hidden markov multi-source monitoring data method in step S3 includes the following steps:
s5.1: model training, namely a model parameter estimation problem, namely adjusting the parameters of a model for an initial model and an observation sequence given for training so as to enable the initial model and the observation sequence to be best fit with observation data, and performing the model training through an EM algorithm or a Baum-Welch algorithm;
s5.2: evaluating the model; giving model parameters, calculating the likelihood of an observation sequence, namely calculating a likelihood value or a log-likelihood value, wherein the likelihood value or the log-likelihood value is used for representing the accuracy of parameter fitting data and is carried out by a forward-backward algorithm; the step S5.1 and the step S5.2 need to be repeated, and the parameter with the maximum likelihood value or the maximum log-likelihood value is selected as the final model parameter;
s5.2: estimating a hidden state; given the model parameters and the observation data, the most likely hidden state sequence is estimated based on some optimal criterion, i.e., the most likely path for generating the observation sequence is estimated by the Viterbi algorithm.
Compared with the prior art, the invention has the beneficial effects that: 1) compared with the simple time aggregation analysis, the time-space aggregation analysis information is more detailed, and not only can prompt whether aggregation exists or not, but also can carry out space positioning on the aggregation; 2) the time-space aggregation analysis makes full use of the spatial information in the data, and early warning is more timely.
Drawings
FIG. 1 is a hidden Markov multi-working schematic diagram of multi-source monitoring data.
Fig. 2 is a flow chart of hidden markov model establishment.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments.
Examples
Referring to fig. 1-2, in order to achieve the above purpose, the following technical solutions are specifically adopted in the present invention: a data analysis method for campus epidemic prevention, control, monitoring and early warning is characterized by comprising the following steps: the method comprises the following steps:
s1, firstly, data are collected through a campus epidemic prevention and control platform mobile phone end 'health report' or a campus temperature sensing system or nearby epidemic and co-process query software;
s2: tracking and monitoring the abnormal data collected in the step S1;
s3: and performing space-time analysis and early warning statistics on the data detected in the step S2, wherein the method for the space-time analysis and the early warning statistics is a Bayesian space-time analysis method or a hidden Markov multi-source monitoring data method.
In step S1, through the "health report" function of the mobile phone terminal of the campus epidemic prevention and control platform, the collected data includes two types, one type is health data, i.e. it indicates that the physical condition is good, and the other type is abnormal data of the body, including: the patient has symptoms of cough, weakness, fever, diarrhea and the like, and the fever symptom needs to be filled in the body temperature range; body temperature data directly acquired by a temperature sensing system, wherein one type of the body temperature data is data in a normal range; another is data at temperatures above 37.2 ℃; and comparing and analyzing the filtered data by using nearby epidemic situation and co-process query software.
The method for monitoring in step S2 includes the following steps:
s2.1: before returning to school
The students and the teaching staff with abnormal body are tracked and monitored through the health reporting function;
tracking and monitoring students/teachers of epidemic high-risk regional cadavers, students/teachers of middle-risk regional cadavers, students with high-risk sojourn histories after a specific date and teaching workers by regional division and monitoring;
tracking and monitoring students and teaching workers who have suspected contact history in the same district, building and class by using integrated software of 'nearby epidemic situation' and 'same-course query';
s2.2: go back to school
Tracking and monitoring the persons who have suspected contact history of the same-class traffic workers through the integrated software of 'same-course query';
when the temperature is corrected, temperature measurement is carried out, and isolation and tracking monitoring are carried out on the abnormal temperature person;
isolating, tracking and monitoring students who enter the school according to batches and have not relieved abnormal early warning in 14 days;
s2.3: after the correction
The students and the teaching staff with abnormal body are tracked and monitored through the health reporting function;
tracking and monitoring students with abnormal body temperature measured by a daily infrared automatic temperature sensing system;
and tracking and monitoring students who have sufficient reason to leave the school after entering the school after returning.
The Bayesian spatiotemporal analysis method is implemented by firstly determining basic elements and characteristics of data, wherein the new coronary pneumonia epidemic situation data is symbolic representation of propagation process of the new coronary pneumonia and related factor relationship thereof, and mainly comprises three basic elements and three basic characteristics, wherein the three basic elements comprise tense, position and attribute, and the three basic characteristics comprise multi-dimension, multi-scale and multi-level.
The new crown pneumonia epidemic situation observed at present is the result of a series of complex natural and social factor combined action, and the spreading rule of the new crown pneumonia epidemic situation can be understood as follows: under the action of natural and social multisource complex factors, the space-time change of spreading new coronary pneumonia epidemic situation and the space-time covariant locus of related factors are obtained.
When viewed from a time axis, the method reveals the change of the incidence rate of the new coronary pneumonia along with time, and simultaneously reveals the spatial variation of the incidence rate at each time point;
from the spatial perspective, it was revealed that the incidence of new coronary pneumonia varied from region to region, and that the incidence varied with time in these heterogeneous regions.
Meanwhile, various factors and changes thereof related to the spatio-temporal variation of the incidence rate of the new coronary pneumonia are disclosed.
Multi-dimensional; the time variation of the incidence in each region: the spatial correlation of incidence with the magnitude of heterogeneity and its variation in time.
Multi-scale: the characteristics of the new coronary pneumonia epidemic situation expressed on different empty scales may be different: the characteristic gradual change rule reflected in the process of generalization and refinement.
Multilayer: the individual-class chief and task-epidemic situation control department-school-direct education department-provincial education department-national education department.
The bayesian spatiotemporal analysis method described in step S3 includes the steps of:
s3.1: collecting data by the method of S1;
s3.2: selecting indexes; performing prevention and control according to targets of different syndromes, sorting influence factors of various syndromes, considering the dependency relationship among variables, the availability and the quantifiability of related data and the like, and selecting a proper model index;
s3.3: and (3) network structure learning: determining a proper Bayesian network topological structure by utilizing historical data and combining with prior knowledge, selecting a statistical test-based method or a search score-based method according to actual conditions, and dividing model nodes established after the method is selected into the following four parts:
s3.3.1, global node G, namely the characteristics of the school whole population;
s3.3.2, interface nodes I comprise time and places of campus epidemic prevention and control diffusion;
s3.3.3. an individual network P, wherein each sub-network corresponds to an individual and comprises demographic information, clinical syndrome and the like, and the topological network structure of the individual network P is determined by expert judgment;
s3.3.4, a crowd evidence node O comprises abnormal health card data, abnormal temperature measurement of a temperature sensing system, parallel simultaneous query, same unit of nearby epidemic situation and a building;
s3.4: determining network parameters: under the condition of a given Bayesian network topological structure, determining the conditional probability density of each node, determining the parameter learning of the Bayesian network through expert knowledge and training sample learning, then performing parameter estimation by adopting a parameter estimation method, and representing one outbreak when the test probability exceeds a threshold value.
The parameter estimation method is maximum likelihood estimation, maximum posterior probability or expectation maximization algorithm.
S3.5: optimizing network structure and parameters; and measuring the false positive rate of the model by adopting a curve, calculating the difference value between the outbreak time of campus epidemic prevention and control and the early warning time of the model, and optimizing the model structure by evaluating the result.
The Bayesian network organically combines the individual data and the aggregated data obtained from a plurality of channels in a network form, and estimates the probability of the campus epidemic prevention and control outbreak on the basis of fully considering the complex relation among variables.
Detection principle of hidden markov: as shown in the figure 1, the epidemic situation is assumed to be not directly monitorable, and only some monitoring values corresponding to the epidemic situation, such as abnormal body temperature, cough, parallel in the same course, and the similar unit and building of the nearby epidemic situation, are monitored.
The bottom layer, hidden markov chain model, describes the transitions between states.
And the upper layer is a random model for describing the statistical corresponding relation between the state and the monitoring value.
Hidden Markov modeling is to acquire information from historical data, convert a hidden rule into an actual model parameter, and compare and identify unknown samples by using the historical information and according to a similarity principle.
The hidden markov multi-source monitoring data method in the step S3 comprises the following steps:
s5.1: model training, namely a model parameter estimation problem, namely adjusting the parameters of a model for an initial model and an observation sequence given for training so as to enable the initial model and the observation sequence to be best fit with observation data, and performing the model training through an EM algorithm or a Baum-Welch algorithm;
s5.2: evaluating the model; giving model parameters, calculating the likelihood of an observation sequence, namely calculating a likelihood value or a log-likelihood value, wherein the likelihood value or the log-likelihood value is used for representing the accuracy of parameter fitting data and is carried out by a forward-backward algorithm; the step S5.1 and the step S5.2 need to be repeated, and the parameter with the maximum likelihood value or the maximum log-likelihood value is selected as the final model parameter;
s5.2: estimating a hidden state; given the model parameters and the observation data, the most likely hidden state sequence is estimated based on some optimal criterion, i.e., the most likely path for generating the observation sequence is estimated by the Viterbi algorithm.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents and improvements made within the spirit and principle of the present invention are intended to be included within the scope of the present invention.
Claims (5)
1. A data analysis method for campus epidemic prevention, control, monitoring and early warning is characterized by comprising the following steps: the method comprises the following steps:
s1, firstly, data are collected through a campus epidemic prevention and control platform mobile phone end 'health report' or a campus temperature sensing system or nearby epidemic and co-process query software;
s2: tracking and monitoring the abnormal data collected in the step S1;
s3: and performing space-time analysis and early warning statistics on the data detected in the step S2, wherein the method for the space-time analysis and the early warning statistics is a Bayesian space-time analysis method or a hidden Markov multi-source monitoring data method.
2. The data analysis method for campus epidemic prevention, monitoring and early warning as claimed in claim 1, wherein the data analysis method comprises: in step S1, through the "health report" function of the mobile phone terminal of the campus epidemic prevention and control platform, the collected data includes two types, one type is health data, i.e. it indicates that the physical condition is good, and the other type is abnormal data of the body, including: the patient has symptoms of cough, weakness, fever, diarrhea and the like, and the fever symptom needs to be filled in the body temperature range; body temperature data directly acquired by a temperature sensing system, wherein one type of the body temperature data is data in a normal range; another is data at temperatures above 37.2 ℃; and comparing and analyzing the filtered data by using nearby epidemic situation and co-process query software.
3. The data analysis method for campus epidemic prevention, monitoring and early warning as claimed in claim 1, wherein the data analysis method comprises: the method for monitoring in step S2 includes the following steps:
s2.1: before returning to school
The students and the teaching staff with abnormal body are tracked and monitored through the health reporting function;
tracking and monitoring students/teachers of epidemic high-risk regional cadavers, students/teachers of middle-risk regional cadavers, students with high-risk sojourn histories after a specific date and teaching workers by regional division and monitoring;
tracking and monitoring students and teaching workers who have suspected contact history in the same district, building and class by using integrated software of 'nearby epidemic situation' and 'same-course query';
s2.2: go back to school
Tracking and monitoring the persons who have suspected contact history of the same-class traffic workers through the integrated software of 'same-course query';
when the temperature is corrected, temperature measurement is carried out, and isolation and tracking monitoring are carried out on the abnormal temperature person;
isolating, tracking and monitoring students who enter the school according to batches and have not relieved abnormal early warning in 14 days;
s2.3: after the correction
The students and the teaching staff with abnormal body are tracked and monitored through the health reporting function;
tracking and monitoring students with abnormal body temperature measured by a daily infrared automatic temperature sensing system;
and tracking and monitoring students who have sufficient reason to leave the school after entering the school after returning.
4. The data analysis method for campus epidemic prevention, monitoring and early warning as claimed in claim 1, wherein the data analysis method comprises: the bayesian spatiotemporal analysis method described in step S3 includes the steps of:
s3.1: collecting data by the method of S1;
s3.2: selecting indexes; performing prevention and control according to targets of different syndromes, sorting influence factors of various syndromes, considering the dependency relationship among variables, the availability and the quantifiability of related data and the like, and selecting a proper model index;
s3.3: and (3) network structure learning: determining a proper Bayesian network topological structure by utilizing historical data and combining with prior knowledge, selecting a statistical test-based method or a search score-based method according to actual conditions, and dividing model nodes established after the method is selected into the following four parts:
s3.3.1, global node G, namely the characteristics of the school whole population;
s3.3.2, interface nodes I comprise time and places of campus epidemic prevention and control diffusion;
s3.3.3. an individual network P, wherein each sub-network corresponds to an individual and comprises demographic information, clinical syndrome and the like, and the topological network structure of the individual network P is determined by expert judgment;
s3.3.4, a crowd evidence node O comprises abnormal health card data, abnormal temperature measurement of a temperature sensing system, parallel simultaneous query, same unit of nearby epidemic situation and a building;
s3.4: determining network parameters: determining conditional probability density at each node under the condition of a given Bayesian network topology, determining parameter learning of the Bayesian network by expert knowledge and training sample learning, then performing parameter estimation by adopting a parameter estimation method, and representing an outbreak when the probability exceeds a threshold value, wherein the parameter estimation method comprises the following steps: maximum likelihood estimation, maximum a posteriori probability, or expectation maximization algorithm;
s3.5: optimizing network structure and parameters; and measuring the false positive rate of the model by adopting a curve, calculating the difference value between the outbreak time of campus epidemic prevention and control and the early warning time of the model, and optimizing the model structure by evaluating the result.
5. The data analysis method for campus epidemic prevention, monitoring and early warning as claimed in claim 1, wherein the data analysis method comprises: the hidden markov multi-source monitoring data method in the step S3 comprises the following steps:
s5.1: model training, namely a model parameter estimation problem, namely adjusting the parameters of a model for an initial model and an observation sequence given for training so as to enable the initial model and the observation sequence to be best fit with observation data, wherein the model training can be carried out by an expectation maximization algorithm or an unsupervised learning algorithm;
s5.2: evaluating the model; giving model parameters, calculating the likelihood of an observation sequence, namely calculating a likelihood value or a log-likelihood value, wherein the likelihood value or the log-likelihood value is used for representing the accuracy of parameter fitting data and is carried out by a forward-backward algorithm; the step S5.1 and the step S5.2 need to be repeated, and the parameter with the maximum likelihood value or the maximum log-likelihood value is selected as the final model parameter;
s5.2: estimating a hidden state; given model parameters and observation data, the most likely hidden state sequence is estimated based on some optimal criterion, i.e. the most likely path for generating the observation sequence is estimated, which can be estimated by the viterbi algorithm.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011269950.6A CN112509705A (en) | 2020-11-13 | 2020-11-13 | Data analysis method for campus epidemic prevention, control, monitoring and early warning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011269950.6A CN112509705A (en) | 2020-11-13 | 2020-11-13 | Data analysis method for campus epidemic prevention, control, monitoring and early warning |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112509705A true CN112509705A (en) | 2021-03-16 |
Family
ID=74957526
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011269950.6A Pending CN112509705A (en) | 2020-11-13 | 2020-11-13 | Data analysis method for campus epidemic prevention, control, monitoring and early warning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112509705A (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120244886A1 (en) * | 2009-09-02 | 2012-09-27 | Nokia Corporation | Method and apparatus for tracking and disseminating health information via mobile channels |
| CN111105878A (en) * | 2020-02-14 | 2020-05-05 | 上海金晋智能科技有限公司 | Intelligent system for efficiently preventing epidemic propagation by using mobile phone and network |
| CN111755124A (en) * | 2020-07-01 | 2020-10-09 | 电子科技大学 | Student return management method for epidemic situation period |
-
2020
- 2020-11-13 CN CN202011269950.6A patent/CN112509705A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120244886A1 (en) * | 2009-09-02 | 2012-09-27 | Nokia Corporation | Method and apparatus for tracking and disseminating health information via mobile channels |
| CN111105878A (en) * | 2020-02-14 | 2020-05-05 | 上海金晋智能科技有限公司 | Intelligent system for efficiently preventing epidemic propagation by using mobile phone and network |
| CN111755124A (en) * | 2020-07-01 | 2020-10-09 | 电子科技大学 | Student return management method for epidemic situation period |
Non-Patent Citations (2)
| Title |
|---|
| 刘欣荣 等: "校园疫情全过程防控管理及应对措施研究", 广东职业技术教育与研究, 27 October 2020 (2020-10-27), pages 111 - 115 * |
| 李晓松: "传染病监测数据时空分析及早期预警的统计技术研究进展", pages 13 - 16, Retrieved from the Internet <URL:https://max.book118.com/html/2019/1123/6043112215002124.shtm> * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Wang et al. | Learning to detect patterns of crime | |
| CN112506990B (en) | Hydrological data anomaly detection method based on spatiotemporal information | |
| CN110087207B (en) | Method for reconstructing missing data of wireless sensor network | |
| CN110210681B (en) | Prediction method of PM2.5 value of monitoring station based on distance | |
| CN101354757A (en) | Method for predicting dynamic risk and vulnerability under fine dimension | |
| CN112085621B (en) | Distributed photovoltaic power station fault early warning algorithm based on K-Means-HMM model | |
| CN114141385B (en) | Early warning method, system and readable storage medium for infectious diseases | |
| CN115237717A (en) | Micro-service abnormity detection method and system | |
| CN118051827A (en) | Power grid fault prediction method based on deep learning | |
| CN115470850A (en) | Water quality abnormal event recognition early warning method based on pipe network water quality time-space data | |
| CN115580446A (en) | Non-intrusive load detection method based on decentralized federal learning | |
| CN113269041B (en) | Signal abnormality detection method applied to synchronous device | |
| CN116913098B (en) | Short-time traffic flow prediction method integrating air quality and vehicle flow data | |
| Lei et al. | Prediction of PM2. 5 concentration considering temporal and spatial features: A case study of Fushun, Liaoning Province | |
| CN110662220B (en) | Wireless sensor network anomaly detection method based on time-space correlation and information entropy | |
| CN117272202A (en) | Dam deformation abnormal value identification method and system | |
| Wang et al. | Abnormal event detection in wireless sensor networks based on multiattribute correlation | |
| CN112509705A (en) | Data analysis method for campus epidemic prevention, control, monitoring and early warning | |
| Syed et al. | A novel adaptive neuro-fuzzy inference system-differential evolution (Anfis-DE) assisted software fault-tolerance methodology in wireless sensor network (WSN) | |
| Aman et al. | Influence-driven model for time series prediction from partial observations | |
| CN116028852A (en) | Underground road fire detection method based on BP neural network and D-S evidence theory | |
| Febriansyah et al. | Outlier detection and decision tree for wireless sensor network fault diagnosis | |
| CN114819260A (en) | Dynamic generation method of hydrologic time series prediction model | |
| CN113469228A (en) | Power load abnormal value identification method based on data flow space-time characteristics | |
| CN112561203A (en) | Method and system for realizing water level early warning based on clustering and GRU |
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 |